856 new of 870 responses total.
I think Dan Cross just fixed the shell problem with bbs.
This response has been erased.
Is anyone else having problems with there .procmail filter? Do I need to change something in my .forward?
When opening pine, get the following:
[Folder vulnerable - directory /var/mail must have 1777 protection]
But I don't know why it needs to be globably writable. :/
BTW, my previous ssh problem, I was using /usr/bin/ssh. /usr/local/bin/ssh
works better. Strange.
I just reported the same problems as in 17 and 18 in coop. And I am not getting ANY mail including two I sent myself just a few minutes ago, one of which should have gone to /dev/null. I will try turning off procmail.
My mail is no longer in /var/spool/mail/k/e/keesan (which I had set up .procmailrc to point at) but in /var/mail/keesan. I have edited .procmailrc and sent myself another test mail.
Changing to /var/mail/keesan did not fix my problem of not getting any of my test mails. I had to turn off procmail by putting # in front of the line in .forward. Procmail is still in /usr/local/bin as it should be. I get mail now but most of it will be spam.
My first spam just arrived, now that I have no filter. Can the inbox limit be temporarily increased until procmail can be made to work again? 50 spams a day won't leave me room for any mail.
I got the following crash dump from Backtalk when I tried to post a
response in the "diy" conference. The response was posted anyway.
Backtalk Crash!
Wed Dec 29 11:41:31 2004
Oops! The authors of Backtalk are deeply sorry to confess that our
program has failed to operate correctly. This is due to an error in
Backtalk itself, or in one of Backtalk's script files, or in the
configuration of the system. It is probably not due to anything you
might have done. The page below contains an error message and a detailed
dump of the state Backtalk was in when the error occurred. This may be
useful to anyone trying to fix the problem.
All this information has been logged, but it may still be helpful for
you to report this error and the circumstances under which it occurred
to the system administrators.
Thanks for your assistance in getting the bugs out of this conferencing
system. We sincerely and humbly beg your forgiveness.
ERROR: index_new_resp called with response number 990052351
executing "post_resp" on line 113 of pistachio/response.bt
Version: Backtalk version 1.3.17
Stack:
(empty)
User Dictionary 0:
bgcolor=(FFFFCC) pw_url=(/cgi-bin/pw/backtalk/) goto=(pistachio/read)
vlinkcolor=() allowhtml=1 bgimage=() logout_url=(/cgi-bin/backtalk/)
post.x=(2\
1) post.y=(12) bodytags=() alinkcolor=() anon_url=(/cgi-bin/backtalk/)
highcolor=(FF0000) pseudo=(David Brodbeck) global_header={...} prevmax=(825)
textcolor=() adm_url=(/cgi-bin/adm/backtalk/) linkcolor=()
global_footer={...}
System Dictionary:
anonymity=0 since=@0 author_erase=1 texttype=(text/plain) amadm=0 mayhide=0
http_headers=1 fw_retitle=1 blindfold=0 resp=0 editdate=@0 hidden=0
id=(gull)
http_expires=@0 uid=9062 newbie=0 linktonew=1 sessions=0
homedir=(/a/g/u/gull)
yapp_format=0 mayread=1 mayedit=0 ifav=1 expire_session=3600 mayretire=0
canspell=1 flavor=(pistachio) lastdate=@1098745982
scriptdir=(/usr/local/lib/b\
acktalk-1.3.17/script) parentresp=0 saverep=0 fishbowl=0 maxresp=825
spell_lang=() expect_resp=826 userlist=0 authorid=() spell_ignore=()
authorname=() fw_erase=0 author_retire=1 csel=() http_location=() maypost=1
conf=(diy) copyright=(Copyright 1996-2004, Jan Wolter and Steve Weiss)
conflist=(/bbs/conflist) authoruid=0 conftitle=(Diy) [=mark isel=() linked=0
`=mark loglevel=0 cpass=() secret=0 canattach=0 fwlist=(keesan, scott)
canemail=0 auto_recompile=1 progname=(backtalk) grouplist=0 exec_limit=0
forgotten=0 maxread=825 mark=mark showforgotten=1 mayseefname=1
publicflavors=(public) author_kill=1 http_no_cache=0 urlarg=()
bbsdir=(/bbs/)
dbtype=() fw_hide=0 allowgraphics=1 alias=(David Brodbeck) stack_limit=0
nopwedit=1 attachments=() gid=1002 item=8 author_retitle=1 author_hide=1
frozen=0 scriptname=(response) mayfreeze=0 editfrozen=1 fw_edit=0
linkdate=@886612066 mayretitle=0
cgiquery=(pseudo=David+Brodbeck&texttype=text\
%2Fplain&post.x=21&post.y=12&conf=diy&csel=&item=8&isel=&rsel=new&goto=pistach\
io%2Fread&prevmax=825&noskip=1&showforgotten=1&text=I+installed+a+Wheelskins+l\
eather+steering+wheel+cover+on+my+car.++It+was%0D%0Asomewhat+time+consuming+%2\
8about+an+hour+and+a+half%29+but+the+results+are%0D%0Aworth+it+for+the+one+par\
t+of+the+car+I+touch+the+most.%0D%0A) newuseropen=0 amfw=0 post_log_file=()
secure=1 rfav=2 http_content_type=(text/html) maxitem=42 author_edit=0
mayresp=1 readdate=@1098799493 author_freeze=1 allowanon=1 title=()
mayseealias=1 maykill=0 noskip=1 confdir=(/bbs/diy) particip=(.fixit.cf)
erased=0 date=@0 favicon=(/usr/local/www/favicon.ico) timezone=()
text=(I inst\
alled a Wheelskins leather steering wheel cover on my car. It
was\nsomewhat t\
ime consuming \(about an hour and a half\( but the results are\nworth it
for t\
he one part of the car I touch the most.\n) retired=0 loginlen=8 rsel=(new)
mayerase=0
Environment: AUTH_TYPE=Basic CONTENT_LENGTH=383
CONTENT_TYPE=application/x-www-form-urlencoded
DOCUMENT_ROOT=/usr/local/www
HTTP_ACCEPT=text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/
plain;q=0.8,image/png,*/*;q=0.5
HTTP_ACCEPT_CHARSET=ISO-8859-1,utf-8;q=0.7,*;q=0.7
HTTP_ACCEPT_ENCODING=gzip,deflate HTTP_ACCEPT_LANGUAGE=en-us,en;q=0.5
HTTP_CACHE_CONTROL=max-age=259200 HTTP_CONNECTION=keep-alive
HTTP_HOST=www.cyberspace.org HTTP_KEEP_ALIVE=300 HTTP_REFERER=http://www.cy
berspace.org/cgi-bin/pw/backtalk/pistachio/read?x.x=0&
x.y=0&conf=diy&csel=&rsel=new&noskip=1&showforgotten=1&isel=8
HTTP_USER_AGENT=Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.5)
Gecko/20041107 Firefox/1.0 HTTP_VIA=1.1 gatekeeper.interclean.com:8080
(squid/2.5.STABLE6) HTTP_X_FORWARDED_FOR=152.160.178.15
PATH=/sbin:/bin:/usr/sbin:/usr/bin REMOTE_ADDR=152.160.178.33 REMOTE_PORT=37109
REMOTE_USER=gull SCRIPT_FILENAME=/usr/local/libexec/cgi-bin/pw/backtalk
SERVER_ADDR=216.93.104.34 SERVER_ADMIN=webmaster@cyberspace.org
SERVER_NAME=www.cyberspace.org SERVER_PORT=80
SERVER_SIGNATURE=<ADDRESS>Apache/1.3.29 Server at <A HREF="mailto:webmast
er@cyberspace.org">www.cyberspace.org</A> Port 80</ADDRESS>
SERVER_SOFTWARE=Apache/1.3.29 (Unix) mod_ssl/2.8.16 OpenSSL/0.9.7c
GATEWAY_INTERFACE=CGI/1.1 SERVER_PROTOCOL=HTTP/1.0 REQUEST_METHOD=POST
QUERY_STRING= REQUEST_URI=/cgi-bin/pw/backtalk/pistachio/response
SCRIPT_NAME=/cgi-bin/pw/backtalk PATH_INFO=/pistachio/response
PATH_TRANSLATED=/usr/local/www/pistachio/response
The backtalk crashs and the weird "response not posted" message from fronttalk are the same problem. They have to do with byte-order problems in the sum files for the conferences. Old and new Grex stored binary files with opposite byte orders. So when we copied over the sum files from the Sun, they arrived all backwards. Theoretically both Picospan and Backtalk are supposed to be able to detect the backwardness of the sum files and compensate for that. Looks like Picospan is succeeding but Backtalk is slightly confused. Easiest fix is to regenerate the sum files. You can just delete them. Picospan will complain about missing sum files until someone does a "fix sum" command in the conference to regenerate it. Backtalk automaticaly regenerates them if they are missing.
This response has been erased.
I removed all the old sum files, and had backtalk regenerate most of them. I think I probably missed regenerating some. If you see a conference where Picospan complains about there being no sum file, there are two fixes. Either a fairwitness can say "set summary" in Picospan, or any user can join the conference with Fronttalk or Backtalk. Picospan will still work without a summary file - it'll just be slower.
While trying to get procmail to work (now it lets through ALL my mails) I attempted to logout and then login as a different user and logout got me a NO CARRIER with no opportunity to login again. I had to redial.
Just "|/usr/local/bin/procmail" without the quotes works fine; I don't know whether the quotes would be needed if you added a parameter.
I just emailed myself from a non-grex account (after making sure keesan was no longer on my white list) with subject line 'viagra cialis meds', all three of which words are in my subject filters, and the mail went to my inbox. I have the double quotes - should I remove them? It is acting as if procmail is not there. I changed both .forward and the MAIL line in .procmailrc and these are both globally readable. I also got another spam not from me in the inbox. My filter, when working, gets about 95% of spams. Would one of you experts mind trying out a very simple procmail setup to see if it works for you?
Yes, new Grex will hang up on logout. I think this is hard to change.
I have my filter working. (I copied yours, commented out everything except the first and last conditions, and changed them to something simple I could work with. My filter puts anything from my spamcop account into a folder named Mail/spamcop [in Maildir format], and forwards anything else to my spamcop account.) MAIL=/var/mail/blaise LOGFILE=$HOME/Mail/from VERBOSE=off :0 * ^From:.*jtrigg.*@spamcop.net $HOME/Mail/spamcop/ :0 !jtrigg+cyber@spamcop.net
I just got one REAL spam (not the ones I was making up and sending from myself) sent to /dev/null. So it appears to be working. I don't know why mails from keesan@ are going into my inbox no matter what the subject line.
Jan, why won't the new grex allow logging out and in again?
Sindi, mails from *@grex are whitelisted before cialis is blacklisted. (The grex whitelist is in the middle of the blacklists.)
This response has been erased.
I just removed from my white list 'org edu grex keesan' as From's and procmail just caught a spam that I sent to myself. But I still cannot view my log file (/mail/from) from within Pine. I can view it with less so it is not a problem but it may be symptomatic of something not being set up right. I could view the first time but ever since it has been blank. I now have a choice of 1 or 2 in 'from', both blank.
It's possible that the new version of pine is more picky about mbox format (which the procmail log isn't but is similar to).
Old Grex was modified not to hang up. It used to confuse a lot of people who were used to the standard behavior. It would probably be possible to figure out how to modify new Grex to do that, but I think it'd be a lot of work. Fronttalk directly executes backtalk when used on Grex.
When I log in I get a message that I have mail, when I go into elm it says I have 0 messages.
Pine tells me that /var/spool/g/l/glenda doesn't exist. I prefer elm but tried pine see if it could find this elusive new mail.
Oops, make that /var/spool/mail/g/l/glenda
In your .login file, remove the line that starts "setenv MAIL". That should fix the problem; if not, replace it with "setenv MAIL /var/spool/mail/glenda".
This response has been erased.
Re the final paragraph of 18: that is what I expected to hear. /usr/bin/ssh is the ssh executable; /usr/local/bin/ssh is a wrapper to get through the kernel blocks. The OpenBSD pf filters work well, but they are different from our home-grown ones. A general note on the pf filters: Every program that provides outbound access to members will have to be linked to the wrapper program. The wrapper _may_ have to be modified as well. So if you find something that worked on the old machine but doesn't work here, please either mention it here or send a message to staff.
sz (zmodem send) either is not present, is named differently, or is not in the default $PATH. kermit send pretends to work, but doesn't actually send anything. At least sftp works, finally! Though I need to find a Windoze client that has download progress reporting. (The version that came with Putty does not.)
/usr/local/bin/sz is linked to /usr/local/grex-scripts/xmodem_wrapper which doesn't exist.
This response has been erased.
I just confirmed the kermit send problem - I got the little bar showing that it reached 100% and then looked and what I sent was not where I sent it. While telnetted here.
Ftp worked perfectly in both directions (put and get) just now. I will be happy to test any fixes to kermit so that non-members can also transfer files.
Non-members can ftp in, they just can't ftp out.
OK, something that either I had set or grex had set was not maintained during the nextgrex migration: I have always been able to count on a new login prompt being issued when I did exit from sh. Now my connection is dropped. What setting do I need under nextgrex to maintain the old behavior?
this b0xen is snappy!! kewlness!
Regarding kermit, I can send and receive files when dialed into grex using kermit to dial. I was going to try telnetting from grex to sdf using kermit telnet but sdf is not answering. When I got the phenomenon of kermit appearing to send but nothing arriving (sending from grex to sdf account) was when I telnetted using the non-kermit telnet program, which even on old grex did not allow me to send files via kermit but on the old grex did not appear to be sending, just timed out. Is this a newer version of kermit? If so I might want to report this as a bug to the kermit people. How do I determine the version? I think I tested kermit send/receive on NextGrex earlier, having telnetted to NextGrex from OldGrex using kermit telnet, and it worked.
I am seeing an awful lot of items in conferences other than agora as new. The first few I looked at seemed to contain responses from me. Is there some general fix for this problem? I have been doing fixseen at each conference (it is nearly instant on this new grex).
The items show up as 'brand new items'.
I can telnet to sdf.lonestar.org or m-net.arbornet.org using the plain telnet program but I CANNOT telnet with kermit - No route to host, every time. I was using kermit telnet on the old grex. So this is two things odd about the new kermit - is it one that was precompiled for OpenBSD? If so I will report it to the kermit support group.
This response has been erased.
Once most of the major transition issues are straightened out, can spamassassin (or a similar tool) be made available to users who wish to invoke it via procmail?
my grex mail dosn't seem to be working properly, I can send mail out but when I e-mail myself from my other address, nothing shows up.
hThe shell problem I reported is fixed. Thanks! However, backspaces still show up as ^H. Plus, the "byte abcd" problem with the pager is still there. I think thes e two problems need global fixes. Looks like good progress so far on this big transition, though! 8-)
Re resp:45: Try WinSCP3. I like it a lot. Re resp:51: It was just explained a few responses up that that was an unusual behavior of old Grex that confused a lot of people. It's unlikely to be duplicated on new Grex. If you explain why you want to log out and then immediately log in again, maybe we can suggest an alternative way to accomplish the same thing.
Why I want to logout and login: to check Jim's mail without a second phone call. Jim also helps out a few people with using grex and sometimes has to go into their accounts to fix things. If I make a change to .cshrc or .login and I cannot remember how to get this change to take effect without logging out and in again (such as experimenting with ^H and backspace to try to help someone else with that problem). Charcat, can you mail yourself from within grex? Do you want me to try mailing you? Do you want to try mailing me from your other address? Why I use Kermit to telnet: I have an account at sdf.lonestar.org (which used to be a lot faster than grex and it has links2 and grex did not, and it has the latest lynx 2.8.5, and I have 100MB storage space and a website with photos) and I can use kermit to telnet there and do file transfers over the same connection, rather than having to exit kermit to do ftp, also one level of membership there allows kermit file transfer but not ftp (the $1 membership). Could the same group with outbound ftp have outbound kermit please? Also I can print my mail on my printer if I am dialed into grex with kermit (or telnetted with kermit to grex) and then telnetted from grex to sdf with kermit, but not with regular telnet. But I think charcat's mail problem should be fixed first, and the ^H backspace.
Offhand, a good way to get the old behavior would be to use "exec login" in place of "logout".
Re resp:62: If all you want to do is rerun your .login, use the command 'source .login'. If you want to log in as someone else, type 'login' and you'll get a new login prompt. Note that this opens a subshell; when you type 'exit', you'll be back in your own account, so don't just turn the computer over to them and leave. ;)
Or you could type 'exec login' if you don't want to open a subshell.
Thanks, I will try all of the above ways to re-login. I just looked at .login and I find /usr/ucb on the path and also erase '^?' and tset in there. Could the script that fixes .login's remove /usr/ucb from the path? I don't know why someone is having trouble with backspace but that appears to be set in .login too. Changing the ? to H might fix the problem for those who have the problem. My backspace works fine. For some reason I have erase '^H' followed by erase '^?'. I can erase what I am typing here with Ctrl-H but not Ctrl-?. Can someone explain?
I tried the following to login to Jim's account and get back to mine: 1. login then logout > NO CARRIER 2. exec login then exit > NO CARRIER 3. login then exit > NO CARRIER exit appears to invoke logout. Luckily we now have unlimited phone service. Did I miss anyone's suggestion of how to get out of Jim's account but stay connected?
This response has been erased.
ssh jdeigert@localhost asked me a question which I had to answer with yes and then it added jdeigert to some list and let me into his account. I could exit with 'exit' and not get disconnected. Thanks. I will make an alias to save typing this out.
A lot of brand new items showed up in other conferences and I could type fixseen to put them back to used items except in 'homme', where I could neither fixseen nor read them but was told 'bad item file header'.
I had forgotten that you are using csh, where login is a builtin. Try "exec /usr/bin/login" -- that should log you off and let someone else log in. Note that once they have logged in, they will have to do the same to let another person log in; if the next person to log in uses logout/exit, then it will disconnect.
Putting literal ANSI escape codes in picospan rsep/isep/ishort (via .cfonce) doesn't work for changing colors in bbs any longer. All I get are the raw codes. My terminal can show colors just fine otherwise.
This response has been erased.
I was running 'bbs' however 'ft' exhibits the same behavior.
This response has been erased.
Does anyone know what causes pine to create these .pine-debug1 files in my home directory? They tend to annoy :)
This response has been erased.
When I just went to run mail I got this: $ mail /tmp: write failed, user disk limit reached mail: /tmp: Disc quota exceeded Wassup with that?!
I am seeing the last line of new / newresponse bbs items missing. I forget what I had set my pager to - more or less - but now instead of seeing "More" I am seeing "byte xxxx". What gives?
Missing the last line of items is often a symptom that the number of rows on your terminal is not the same as the system thinks it is.
(I can't read that last line. :-) If it says something about number of lines per screen in some config file setting somewhere, then since I didn't change anything, it's something that got broken during nextgrex'ing. Should I go on a search and destroy mission for myself (and if so, where?), or will something systemic be fixed?
Gull said: Missing the last line of items is often a symptom that the number of rows on your terminal is not the same as the system thinks it is. What terminal type are you set to? (echo $TERM) This line is for the line-eater (oh, that takes me back...).
This response has been erased.
the !lock_my_terminal and !lock programs are defunct
Intermittently there occurs some sort of condition where (a) Pine does not function, and (b) attempts to enter even a single character into a BBS response result in a "Core dumped" error message. It's not happening now, but did happen last night and another previous time.
I think I fixed kermit. Try it; if it doesn't work send a message to staff, please.
Pine no longer complains about permissions. In my home director I find no pine debug files but there are three '0' length pinerc-numbered files (1 of 3, etc) from June 20 which I presume I can delete. Kermit worked for telnetting to sdf, thanks Joe. I will try exec /usr/bin/login (is this different from /usr/bin/login?) and if it works make an alias. But ssh worked okay to change accounts.
exec /usr/bin/login, logged in as jdeigert, exit > NO CARRIER. I don't think I will bother trying without the exec. ssh works. There are still two unread 'new items' in homme that cannot be read or fixseen'ed.
This response has been erased.
I cannot fixseen, read, or even 'b n' without getting the messages 'bad item file header' for 'homme' - will have to actually read the items.
I forgot I CANNOT read the items so I have no idea what numbers they are. Perhaps I can delete something in my home directory if nobody else has the problem (a conference file).
I deleted two files starting in .homme which made no difference so the problem may be elsewhere than my home directory.
I just went to the homme conference again (after getting that bad item file header message) and this time it was only 1 brand new item and this time I could also read it so the problem appears to be local not global and is fixed.
I noticed that one of the item files in /bbs/homme was owned differently than the others, so I changed it to match.
Thanks, Joe, you must have done that before I tried homme again. Everything is working for me now, how boring. Grex did hang up on me once but that is not new.
re resp:2: Yes, nethack is now working great. Thanks!
Hmmm, I may not be seeing a missing last line now. But anyway, as instructed: $ echo $TERM vt100 $ stty -a speed 9600 baud; 24 rows; 80 columns; Looks pretty standard to me...
I have noticed that login no longer accepts ssh1 (ssh2 now OK - this is just an observation).
!more .cfonce # here is where you can put PicoSpan customization define pager more And under nextgrex, as I mentioned, my "more" prompt says "byte xxx". Also note that when I run mail under nextgrex, and do a reply, when I use ~p to "print" what my reply is going to look like, it doesn't seem to honor a pager setting as it did under old grex.
What do you want your "more" prompt to say? Is the mail problem actually with the program "mail", or another mailreader?
The problem that I mentioned in Response 85 occurred again last night.
Okay, thanks for the question on which pager I was using. It's the subtle things that get you on a system change, it seems. Since more is essentially the same as less now and I had "define pager more" before I made it "define page 'less -R -E'" now and that seems to have fixed the ANSI escape problem I was having with bbs.
This response has been erased.
Charcat please mail me again if you want me to send you a test mail because after I removed grex and cyberspace from my white list, procmail sent your mail to /dev/null. I have to figure out where I am overdoing the spam filter (and why it let through 0ffer on the subject line). I just sent you a mail - did you get it? (6:35 pm Friday).
In terms of the homme cf fix, it's still giving me a bad item number error in bbs and only the first 19 items shown when browsing (though one can read others beyond 19 by directly referring to them). ft has the same issue, it just doesn't say anything about a bad item number.
ebay no longer works with lynx. I cannot sign in. Links won't sign in either - tells me 'no route to host'. With w3m I get a screen telling me to enable cookies or use the latest IE or Netscape. Can w3m actually handle cookies? Did something change at the grex end? Ebay actually looks a bit different.
w3m -cookie www.ebay.com got me signed in. Is lynx set to reject cookies by default? And links?
Links got me 'your browser is rejecting cookies'. Lynx got me nowhere during signin until I did lynx -cookies www.ebay.com and tried to sign in. I got the signin page andthen 'your browser is rejecting cookies'. Has there been some change to the lynx defaults? I would like to set it to accept cookies from ebay and to ask me about cookies from other places, or at least to ask me about cookies like it used to so I can decide at individual sites. Fine with me if it throws out all cookies when I exit lynx.
Links was not set up with a mail program. I type ESC for the menu, S for setup, went down to Mail and telnet, and filled in on the first line Pine % and now g, mailto:keesan@grex.org takes me to the Pine compose screen with that address filled on on the To: line. I think beginners would appreciate having Pine or at least some mail program set up for them. I also added the option of numbering links under HTML options, and something about moving within columns.
What's happening on GrexGallery? I'm not getting to upload a
pic into the GrexGallery but that doesn't work :(
There was a large increase in the rate of porno spam this evening. Not an intrinsic Grex problem, of course, but a problem still.
FTP should be working for inbound connections now. I think I fixed it at least for the moment. More work will have to go into that. Grex crashed tonight because of a resource starvation issue. Obviously its serious or the system wouldn't have crashed but it can be fixed. It's one of the "kern" entries kern.maxfiles I think. I dead tired at the moment but I will be tweaking this tomorrow if someone else doesn't beat me to it.
Any chance of fronttalk/backtalk support for European characters? I tried entering the three extra Norwegian vowels in both fronttalk and backtalk, and neither was able to put them into a response. Even reading in a text file got me "stripping bad characters".
(Staff is welcome to grab the file ~scott/char-test for the Norwegian characters)
Okay, if in Picospan I turn off 'gate' via the command "set noed_always" then start a response, then use ':r char-test' to read your file into the response, the characters are successfully entered, thusly: Test of characters: æ ø å Whether they display correctly on a users' terminal depends on their local terminal settings. I was able to get them to display on mine (Mac OS X 'terminal') by setting the character encoding to ISO Latin 9. When I tried to read the file with 'gate' enabled, it stripped out the characters. Maybe gate has a config option to disable this behavior?
I did an RTFM on the gate man page and found that ':r -s FILE' disables the "unprintable-character"-stripping behavior.
I checked out lynx at sdf.lonestar.org and it works at eBay - asks you whether to accept every cookie, like lynx at grex used to do. Could someone please change lynx.cfg or whatever it is that sets cookie behavior at the new grex? Same for w3m and links - make sure they either accept all cookies (and probably delete at exit) or ask about them.
Re the 'bad item header' problem in the homme conference: I looked through the item files and found that item 22 was corrupt (most of it appeared to be missing, in particular the standard header). So I renamed the file from _22 to BAD_22 and regenerated the summary file for the conference. The error message seems to have gone away. Maybe an intact item 22 is still available on oldgrex and can be restored from there. I'm not sure how to get to oldgrex at the moment, so I can't check.
Got to oldgrex and looked; the copy of homme item 22 is corrupt there too. Prospects for restoration look dim...
When I ssh'ed to check Jim's mail I was told: Error saving configuration ... Disk quota exceeded. Jim has 15 mails in his account, totalling under 100K (not counting headers). I have 750K of mail with no such message. I do not seem to be getting 'new mail' notifications.
I think the message was referring to the total disk usage in his account, not his mail.
Thanks for the international stuff, John. I'll wrassle around with gate.
Last line missing from newresponse bbs items is "back".
I am pretty certain that is the nextgrex version of "more" that is the culprit. I can live with the silly "byte xxx" paging prompt, but it is somehow gobbling up the last line of every page it presents me.
The problem with gate and international characters would seem to be the "isascii" system function, but looking around further it appears that the international characters (a couple dozen of them) should be acceptable.
Re 118 (bad item header in homme cf): it's working as expected now, thanks!
The following e-mail message arrived just after the shift to newGrex. Yet it is dated 24 October. What gives? > Date: 24 Oct 2004 01:45:37 -0400 > From: Mail System Internal Data <MAILER-DAEMON@grex.cyberspace.org> > Subject: DON'T DELETE THIS MESSAGE -- FOLDER INTERNAL DATA > This text is part of the internal format of your mail folder, and is not > a real message. It is created automatically by the mail system > software. If deleted, important folder data will be lost, and it will be > re-created with the data reset to initial values. Does it really have to be left in one's inbox?
I don't know why it showed up, but I think it's safe to delete.
I get those when I've used the web email client for my regular (non-Grex) email. It's basically a way to store various settings (new folders and other stuff) that I might do within the web client. When I get home and download my mail into my regular email client one of those "Folder internal data" emails shows up, and I just delete it.
Could it have been created because I was reading my mail for the first time with newPine?
I appears that the idle zapper is not working right in the following situation: Connected to grex via internet. Logged in via telnet. Did "exec login" to another account. At a certain point the following was displayed: Sat Jan 1 13:58:09 This terminal has been idle 15 minutes. If it remains idle for 5 more minutes it will be logged out by the system. Logged out by the system. But no disconnect was performed. The "Logged out" line was repeated every so often. But my sh was still responding, and whoami showed that I was still the second account.
This response has been erased.
Why would Jim get the disk quota exceeded message only when he used pine, not when he logged into his account?
There are two different quotas for home directory space and mail.
Yes, and he has about 100K of mail so why get a quota message when using pine?
I just deleted 8MB of freedos image files from his home directory but that does not explain why the quota message showed up when going into pine. When will the new disk quotas be enforced? I don't know why Jim did not use sdf to store these files, since he has 30MB space there (for $1).
I don't know what the quota is for mail, but he must be over it.
Re #103:
I am doing the following when the problem occurs:
* Trying to read my mail with pine. Abruptly exits, don't remember the
message.
* Entering a response in BBS. The editor starts up, but immediately dumps
upon my typing the first character of the response.
New information: The problem seems to occur if, and only if, I'm dialed
in direct. I'm entering this response, without problems, via the Net, and Pine
worked normally as well. Just a second ago I tried direct dialing; Pine did
not work and I didn't bother to try using BBS.
I have a problem while using backtalk. The program refuses to recognize that not all the items in a conference are new, even when i click on "mark all responses read".
Hmm; is your terminal setting ('echo $TERM') the same dialling direct as when
connecting over the net?
Assuming you mean me: TERM is vt100 on the net. I'm not at home at the moment to dial in; will check that when I'm there.
When a user goes over quota, they have seven days to get back under the quota. If they don't, quota prevents their creating new files. Several people are over quota right now. They are (mostly) in the grace period. Some people are already contacted staff for help. Others, like Jim through you, find other ways of dealing with the problem.
But why does the 'over quota' message appear when you go into pine and not when you log in? This is confusing, not helpful.
This response has been erased.
Jim had 9MB of files in his home directory and nothing in /tmp and 100K in the inbox. I can try to duplicate the problem by downloading a 2MB file to his home directory if that helps. I went in and out of pine several times and kept getting that over quota message every time. Or maybe he also had the 9MB in /tmp? Does it stay there for a week?
Pine was trying to write to a file in his home directory. That attempt triggered the quota notification.
When dialing direct:
Pine exits with "Cannot open terminal capabilities database".
TERM is set to "dialup". Changing it to "vt100" does *NOT* fix the
problem.
When trying to enter a response in BBS, spacebar enters a new line, and
typing any other character results in a "Core dumped" followed by "OK to enter
this response?".
When connected via the internet:
TERM is set to "vt100".
Pine and BBS work normally.
Scratch that, TERM is now set to "xterm", logged in from home in Putty under Windoze 98SE in VMWare. It was "vt100" in ssh on the Zaurus.
Try editting your .login to remove the '"$TERM"' from the end of the tset command.
What is the user quota on /tmp ? Can it be raised? I'm being thwarted in sending e-mail with long text (a few 100K characters).
The quota on /tmp is 200 KB. The limit is 1100 KB.
This response has been erased.
This response has been erased.
BTW, the command to see _your_ quota is "quota". Personally, I prefer the
"-v" option:
quota -v
var system is full
/var: write failed, file system is full cannot log into party.
> df /dev/sd2d 6092380 6039150 -251388 104% /var heh, how can a file system be 104% full? ;) Anyway, what bru said, /var is full.
It appears to be email spools that are filling up /var. > du -h /var |more 999M /var/mail 942M /var/tmp/mail-spool
Thanks to everyone who got the new machine up and running! Very nice. What became of /usr/local/lib/global.cshrc? Upon logging in, I see a message that there is no such file, and my usual shell prompt and some other things don't work. What should I be using instead?
As a side note, drew, if you're using Windoze under VMWare so you can get at PuTTY whilst using Linux/UNIX, PuTTY is now available for Linux, whilst the BSDs have compatibility packages for Linux software.
The global dot files are in /usr/local/etc/ . However, you may need to add
your own prompt. I added
set prompt="`hostname -s`% "
to my .cshrc file.
Thanks, gelinas. :)
The same thing to me in party:
/var: write failed, file system is full
Not sure if someone posted this already, but I got this when in elm and using vi as my mail editor: Error: /var/tmp/vi.recover: No space left on device. Modifications not recoverable if the session fails. Looks like it's all related.
I am still not getting any 'new mail' messages when I log in, though I have new mail. Could someone please fix this, and also do something so users will be notified they are over quota when they log in, rather than when they use pine. Thanks.
I haven't received any new mail at all today - I suspect the fact that /var is full is preventing mail from being delivered.
This response has been erased.
Hey Mark A. Conger, I've sent you an email regarding my
subscription. I'm waiting for your answer!
I am now getting new mail messages. But I think I figured out why I have to put all my friends on the white list. Any mail not caught by one of my filters or the white list is going to /dev/null. At the end of .procmailrc I have the lines :0: and $MAIL - should I change these to send whatever is not caught by a filter to my mail instead of to /dev/null? The reason spams are getting through is they match something in the white list (for instance 'Michelle' got caught by the whitelist for 'mich' (as in umich.edu). I turned verbose on to determine this. Could I add between my last two lines a blank Subject line? (I forget the exact format, just put in what usually goes in a Subject filter but end in *, or . ).
I have fixed my filter by putting a \ in front of the pipe sign in v[pipesign]agra. For some reason I was getting a whole lot of spam caught by that one filter. Anyone able to explain what caused the problem which I just fixed? I narrowed it down by putting myself on the white list in the middle of my filter , then before the Subject lines (it worked), then in the middle of hte Subject filters (it worked) then near the end (it did not work, I was after the viagra portion). I just apologized to at least five people who sent me new year's greetings and now I can tell them it should not happen again!
The reason lynx is not working for me with cookies is that I have my own lynx.cfg file which says to include the grex file in /usr/local/etc/lynx.cfg which does not exist, in fact I cannot find any lynx.cfg at all besides mine. Where is lynx.cfg? Is there a lynx.cfg? I notice that lynx is set to accept all cookies rather than to ask about them. Does it store them somewhere or delete on exit? If I run lynx without using my own lynx.cfg I can sign in to ebay.
When I do use my own lynx.cfg (lynx -cfg=~lynx.cfg invokes it) in order to get rid of images which otherwise show up as [spacer.gif] etc., and to use links and w3m as downloaders (see my lynx.cfg), lynx options still show it accepting all cookies but for some mysterious reason I cannot sign in to ebay. Anyone want to figure out why this is so? I remarked out the INCLUDE line since I don't know where to point it at.
Help no change, no help 4 days later. Okay, 4 days ago I tried the Staff@cyberspace.org method, just as the June 21, 1999 web page says to. tpryan@cyberspace.org here I can't respond. Something about disk quota. I can't tab. Same settings in terminal emulator My iseps & rseps stoped working (putting headers in color). Do I need a higher disk quota? why? Have you tested by coming in via dial-up? -Tim Ryan
Tim, for me, the solution to the rsep/isep no color problem was to tell the pager to pass raw characters on through (e.g. less -R). Edit your .cfonce file for that (define pager). This worked for both bbs and ft for me. Re 171: try the 'locate' command when you can't find a file like lynx.cfg and see if you don't get something at the system level (in addition to other instances in people's home areas, and in source code-- which you might want to ignore or not depending on what you are looking for). I've tried 'ft' a couple times, but both times it ends up stalling my connection (or maybe it's a coincidence). I'll be reading cfs just fine and then no keystrokes are recognized over the course of several minutes (except ^Z after which I can kill ft). It appears to be when you finish reading one cf and hit return at the prompt to move on. bbs works just fine for me without going south in the middle of a session.
Tim, use the commmmand "quota -v" to see how much disk you are using. Then you can determine for yourself whether your quota needs to be raised. Send a message to staff if you do need more disk space.
Re 157: The 100% in a file system (as reported by df) is based on a smaller figure; there's a margin which only root may allocate, in effect. This may allow root to temporarily create files while cleaning up (say).
Re somewhere way back there (albaugh, I think): You can change that "byte xxxx" prompt. Read the man page for more (or less; it's the same man page).
Re #149:
Done. No effect.
Re #160:
No, I am not using Windoze in VMWare to get at Putty. I am using it in
order to connect through Netzero and still have the system running Linux, and
also limit the exposure of the system to the virtual machine.
Also, MAILER-DAEMON keeps sending me messages telling me not to delete them. What's going on here?
I changed my lynx.cfg to INCLUDE /etc/lynx.cfg (after doing a locate and discovering lots of lynx.cfgs in various places including a few other people's home directories) and now ebay lets me sign in. THANKS! Where does lynx save cookies to, or does it delete on exit?
Re 173: No, we didn't test coming in over dial-up. We knew it was a limitation of the testing and decided that we had to live with it.
I'm getting messages that I have mail, and using !pine from the bbs prompt takes me to pine. At that point, pine tells me I have no mail. Any idea what's causing it?
system is very slow.
guys! I,m just getting my mail right now! mail that I should have recieved a long time ago! whoa! oldGreX must've been screwy!
I just got two test messages from charcat from mich.com (charcat are you reading this) but I cannot reply to him or even mail to myself. When I try: Mail not sent. Sending error 451 Error while writing spool file. What's up now? It worked a couple of hours ago. Is this my account or a general problem? Earlier I got stuck at ebay, slowed to a crawl. I think it was the net connection as I could quit lynx instantly, just took forever to go to the next link. And lynx is still a few years out of date and it still redraws itself 3-5 times each time you go to a new URL.
Net connection appears to be broken. I waited 30 sec trying to ssh out and gave up. Our hardware or the ISP? Probably explains the spool file error. But I did get four incoming mails since the slowdown with lynx a couple of hours ago so it appears to be glacially slow rather than dead.
A couple of minutes ago I couldn't connect via the internet. This is being entered via a dial-up session.
/var is still full. I can't even vi one of my small text files - no space left on device. :-(
You can edit your file, you just can't save recovery information. I power-cycled the modem some time between 22:30 and 23:00 this evening. I left it off for several minutes, while I tried to deal with the full /var partition. I realise now that the state of /var contributed to the network problem: mail was being rejected almost as fast as it was being received because the messages couldn't be written to disk. I have turned off incoming mail for the duration.
The message I get when I log on says Sunday is Jan 2 and also Jan 3. Thanks for the rescue, Joe.
Thanks for the note; I've corrected the dates in motd. :)
I was able to ssh just now out of grex, but it is quite slow. Is this related to the full /var?
Probably: mail was rejected because it couldn't be written to disk. Now it is coming in.
Backtalk is running at OldGrex speeds. The page which used to be the choice for bookmarking, now allows a user to log in but then gives a 404 "The page cannot be found" message: http://www.grex.org/cgi-bin/pw/bt/pistachio/begin The solution is to bookmark this page: http://www.grex.org/cgi-bin/pw/backtalk/pistachio/begin This might be confusing some users.
What happened to all my files?!!!
My files have returned just as mysteriously as they disappeared, and my mail has been restored too. :-)
Under old grex, vi (as it does under Solaris, HP-UX, etc.) when started up showed the number of lines in the file being edited. Under nextgrex, vi instead shows the following: <filename>: unmodified: line 1 Is there some way to get vi to show the number of lines upon startup?
BTW, for me, at the moment, telnetting in to nextgrex from the internet, response is DOG SLOW.
Backtalk got a lot faster in the last hour or so. Yay!
Just now when I tried to connect via the internet, I was shown: telnetd: All network ports in use. Is that a case of staff turning off connections for to do maintenance, or do we have a need for the telnet queue after all? (this is being entered via backtalk)
I got the same "telnetd" message that Kevin did for serveral tries at dialing in. Then finally I was able to dial in. I have *never* gotten this message bfefore. I hope the dialin users won't have to wait in the telnet queue. Also - the ^H backspace problem and the "byte abcd" problem are still with us.
The "all ports in use" situation remains, 2 hours later...
Re resp:127,resp:179: Pine uses those "internal data" messages to store various information, I think. I don't know what happens if you delete one. Re resp:157: In BSD, filesystems have an area that's reserved for only the root user. The percentage full takes into account only the user-accessable part of the filesystem. So if the filesystem is filled by a user, then root (or a process owned by root) puts some stuff on it, it will be over 100%. Re resp:201: The 'Byte abcd' problem isn't really a "problem", it's just a different prompt. more is telling you how many bytes into the file you are, instead of giving an information-free 'More' prompt.
Re #197: I've been using vi since 1992, admittedly mostly on SunOS (both 4.x
and 5.x); I've noticed a line-count at the beginning of an editting session.
The command :number (or :nu) will precede each line with its number.
Re 201: The option "-d" will give you the default "more" prompt. You may
want to set this in your .cshrc (if you use csh), with the command
setenv MORE -d
So I was wrong: vi on on the old machine did give the file name, number of lines and number of characters. On OpenBSD and linux (2.4.26, I don't know which distribution), it doesn't. I note that the man page here, as well as on the Linux machine I now use, is for the "nex/nvi" versions. According to the man page, "Nex/nvi are intended as bug-for-bug compatible replacements for the original Fourth Berkeley Software Distribution (4BSD) ex and vi programs." The man pages for vi and nvi do not hint at how to change the initial information. For what's worth, <CTRL>G still reports the current line number and the total number of lines in the file.
(I now see that I omitted a negative in the first sentence of 204. Ah well.)
What about the ^H that I get on my screen instead of the backspace? With OldGrex, I was told upon login that abackspace was ^H, but it never appered on my screen.
Sindi, two things. First, the easy one: the pipe sign is an OR function in regular expressions, so v|agra matches any line that contains either a v or agra. Second, on the "logout without disconnecting" issue: the "exec /usr/bin/login" is a *workaround*, not a permanent solution. Each time you issue "exec /usr/bin/login" that command will log you out without disconnecting, allowing someone else to log in. If they want to log out without disconnecting, they will have to issue "exec /usr/bin/login".
Re #207: Dave, you can set your backspace character to be either /------------------\ | ^H (ascii 8) or | | DEL (ascii 127). | \------------------/ Some people use one, and some people use another. It only matters that what your terminal program sends is what your account on Grex expects to receive. I have my terminal program set to send ^H, and I have a line in my .login file which says: /-------------------------------------\ | stty intr '^C' kill '^K' erase '^H' | \-------------------------------------/ to tell Grex (among other things) that ^H is my erase key. When you see those ^Hs on the screen it means your terminal is sending ^H but Grex is expecting DEL. So you can either tell your terminal to send DEL, or tell Grex to expect ^H.
The usual motd and login list did not appear this morning.
Is DEL ^? Thanks for explaining that I need to login back to my own account rather than exiting to it. But I set up an alias to use ssh instead. In .procmailrc can I use | to set up one long filter instead of many short ones, for instance v.agra|vi.?agra|viag.?ra (to catch v1agra and vi.agra and viaggra)? In fact is there some way to not have to use three lines starting with :0: to filter on each subject, but put everything on one line, and have one filter for Subject, one for From, one for To, one for message body (0B)? Or even one that combines Subject, From and To: My filter, now that I removed the white list, has been rejecting roughly every other mail from each person so I am turning on the verbose log again but it produces pages of text for each spam and someone just sent seven nearly identical ones. So I would like to shorten .procmailrc. Is anyone working on a grex-wide spam filter yet, that people could modify (for instance spamassassin throws out all my electric bills for five causes).
Talk - Couldn't bind to control socket, couldn't assign requested address. I have not changed my permissions since old grex. .
I just got a tel 'help ^U help EOF (honeycut), in response to my offer of help via tels since talk was not working. My guess is vandalism but still, talk is not working.
The talk program that ships with OpenBSD is probably a different one than Grexers are used to. I see that "ytalk" is not available, but it's one of several superior "talk" replacements.
I s t i l l c a n ' t t a b o r
My spam filter caught a real mail until I removed the filter for message body containing h.?g.?h - There was nothing in the mail that I could find with this pattern which I thought included hgh and h.g.h and htgtht etc. What exactly does .? mean - I thought it means any single character or no character at all. Should it be h\.?g\.?h instead? It also caught something with td in the Cc: line when I thought I was filtering on td in the message body with :0B: - does :0B: also include the header and :0: do header only?
Discussions of regular expression syntax probably belong in another
item, but I'll answer one more here. h.?g.?h would match the word high,
for example, or highest (the first two I think of). Yes, .? means 0-1
instances of any character.
The H flag checks the header and the B flag the body; HB would check
both. (The default is H.)
From the Procmail Tips website
(http://pm-doc.sourceforge.net/pm-tips.html):
5.2 Flags HB at top of recipe (warning)
[Philip] Version 3.22 has a bug that keeps the 'H' flag from being
cleared, such that once you use it, it never gets cleared. Using the 'H'
flag will therefore cause problems with latter recipes that use just the
'B' but not the 'H' flag. Either way, the only time you should use the
'H' flag is on recipes that needs to match against both the header and
the body. If you want a recipe to match only against the body and you're
using 3.22, use the "B ??" modifier on the conditions. See message
<http://mailman.rwth-aachen.de/pipermail/procmail/2002-February/008355.html>
. So to be most pportable possible, convert all previously used condition lines
from:
:0 B
* body-check-here
to use this format:
:0
* B ?? body-check-here
I obviously goofed on my hgh filter! I have been using :0: (not :0) for header only, and :0B: for body only - no H's. I will try again to make some sense of man procmail (and remove most of my filters with .? in them). Thanks.
I set up a test filter with either :0 or :0: on the line before the subject and it worked both ways. man procmail also lists samples with or without the colon. Does anyone else want to start an item on how to use procmail as a spam filter?
Talk is still unable to bind to a control socket.
Dave, the problem was that your .profile had two characters, ^ and H, where
it should have had only one, <CTRL>H. I've taken the liberty of fixing
that problem. (I also fixed the interrupt and kill keys, which had the smae
problem.) I also changed the path of the tset command from /usr/ucb/tset to
/usr/bin/tset.
Tim, you had the same problems with .login. I've fixed them.
I don't know why your tab key isn't working.
Mine works for me.
See: Two tabs, on both this line and the line one above it.
This response has been erased.
It is, as long as you're not using a shell which does something special with '^', in which case you might need to surround the ^H with single quotes.
My agora participation file got munged last night for some reason.
This is more like a little niggle than a problem, but why won't colour work in programs like "most" when I have my terminal set to xterm (I don't know of any other terminal settings where it might work)?
I've also tried "vt100", "vt102", "vt220", and "vt420" to no avail.
Have you tried "ansi"?
re #225: does color work in anything? what happens if you use echo in the shell to echo the escape sequence to change text color? any effect? your pager program might be eating the color sequences if it works in the shell but not in bbs.
When I log in I get the screen ------------------------------------------------------------------ Last login: Wed Jan 5 14:32:51 2005 from pcp09229896pcs.sanarb01.mi.comcast.net OpenBSD 3.5 (GREX) #2: Sun Dec 19 15:10:32 EST 2004 To see statements of grex principles and limits, look at http://cyberspace.org/cgi-bin/grex-principles -Grex Statement of Principles http://cyberspace.org/cgi-bin/grex-limits -Grex Limits You have new mail. /usr/local/lib/global.login: No such file or directory. ------------------------------------------------------------------ No motd, no list of users. Is this a permanent change, or a problem?
This response has been erased.
Thanks. That did it. Now I get the message mesg: Unable to find your tty (ttyq5) in utmp file Is that good or bad or indifferent?
It means that the settings allowing (or disallowing) tels won't be available. We may need to recompile mesg.
Is this related to the problem of talk not working? Tels work for me.
I don't know; I've not tried talk.
No, the problem is that talk is not suid. I don't have time to work on this right now.
I get the same error message trying to talk to people whether or not they are logged in - try a talk keesan (I am logged in) or a talk jdeigert. Something about sockets. Presumably a file in the wrong place or some limit set wrong.
235 slipped in. Does 'not suid' mean permissions for talk need to be changed? I will tell people to email me until talk works again.
Sort of. The program permissions determine who can run it, but what needs changing is what permissions the program can use to run the other processes it needs (like those that allow non-members to use talk to connect to a non-Grex-local machine).
I'm not sure if colors will be at all useful in bbs; but most uses them. w3m does, too, but I don't know if we have that installed. I'll check that, and setting the terminal to ansi, now.
http://www.jewsforjoosuz.org/
Setting the terminal to ansi worked a treat; thanks!
http://www.jewsforjesus.org/
I make heavy use of colors in my Picospan iseps and rseps.
Hmm, I wonder what they are.
http://www.jewsforjesus.org/
You can see them in ~aruba/.cfonce if you'd like. My twitfilter also highlights responses from cetain people by putting them in a different color.
You change that every time someone whens the letter game, Mark? cool. :)
Thanks, mark.
Re #247: I confess I often fall behind. But yes, I have a command to define a command to print out the letter.match# file of the person who is "it" in each of the games.
I'd noticed that Pine creates debug files on another system I use, so I asked
the sysadmins about it. They hadn't done anything special. They also
reported that it behaved similarly on another of their systems. So the
creation of these files appears to be a decision by the Pine developers.
I added an alias for pine to my list of aliases some time back, because I
want to go directly to the index of messages, not the "main" screen. I've
now modified that alias to turn off the debug files:
alias pine 'pine -d 0 -i'
This response has been erased.
Pine and BBS response still do not work when dialed in direct; and sz is still missing.
Does /usr/local/bin/lsz work for you instead? I noticed that there is an sb link in /usr/local/bin/ that points to a non-existent file in grex-scripts. That might confuse some people. Would it make sense to link sb to lsb, sz to lsz, etc?
<twenex muscles in> Generally speaking, I would advise against linking new versions of programs (in the sense that vim is a new version of vi, gnu tar a new version of tar) to the names of the old ones unless the new version provides functionality as near as dammit identical to the old one when called as such. (On a related note: a lot of "easy to use" Linux systems alias "cp" to "cp -i", "rm" to "rm -i", and so on. If this applies to any grexer, I would advise them to realias them back to their default "values". Not doing so runs the risk of coming to a system where they are NOT aliased and accidentally deleting crucial files because you expected the system to ask you if you /really/ wanted to do that, but it didn't. At first I used to disdain aliases for this reason, but then I hit upon the following strategy: Say I type "ls -F" often enough that it becomes a pain to have to type it all the time, especially since i never use "ls" on its own. So I create an alias based on ls which gives a clue as to the extra flags, such as "lsf". That's a simple example, but with longer commands it could be quite useful. Unless you know your system well, it's a pain to have to find the file where the distributors have aliased all the commands, and you run the risk of having to do it again when you upgrade. So just type, for example: alias ls="/path/to/ls" in whatever .profile or .login file your shell uses (the csh syntax might actually be a little different). This forces the shell to look for the command instead of replacing its behaviour with a new one. If you don't want to do this for all the systems and/or account you use, then just do it for root on those systems you have root access to.
In all instances of ".profile" or ".login file" above, substitute "rc file" instead.
I was able to telnet in just now but dialin does not work:telnetd: all Network ports in use. ... Followed a few seconds later by NO CARRIER. Twice. There is considerable telnet lag. There was no wait when I telnetted.
Using picospan, there is a few seconds wait for the last few lines of the next response screen to appear. Vandalism?
asd, aka 'smart' is using 97% of CPU for the past 550 minutes, running 'john'. I will email gelinas. Grex is still usable. Is there some way to limit individual users to 1 or at most 5% of CPU use?
I have in the past couple of days received three fragments of emails (presumed
spam) consisting of just the beginning of the header such as:
From the-concourse-on-high Thu Jan 6 11:01:04 2005
Received: from [201.17.23.27] (helo=c911171b.rjo.virtua.com.br)
by grex.cyberspace.org with smtp (Exim 4.42)
id 1CmZp2-0003qY-M4; Thu, 06 Jan 2005 10:45:13 -0500
Received: (from pyroxenite@201.17.23.27)
by helmholtz5[1
Is this just sloppy spam-writing or something going wrong in the middle of
the mail receiving process at grex? Or vandalism?
Some people may have more problems in their scripts now that we've moved away from a SysV UNIX to a BSD UNIX - the "usr/ucb" directory in a SunOS sytem is where BSD UNIX commands were put in Suns SysV OS. Something like the move from Korn shell scripts to bash. Shell scripts using Sun's SysV commands may not work the same in BSD (or be missing entirely) so be aware. Also, there are more differences in the directory structure and the whole environment and deamon setup that may affect scripts written in a SysV system. Better test your scripts before trusting them. In my case (so far) I had to remove all my aliases and removed the '/usr/ucb' reference in two places in my .profile. I also notice git is gone and without mc or git Grex is not very easy to use as file management is a pain without a file manager. I'll keep playing though.
general question: is there a way to spam-filter for plain-ol-mail? pine is a pain (imnsho) even with its 'advanced' features (bloat?). plain-0L-mail is soooooooooooooooooooo easy to use/learn adn it soes not suck up f feast-full of quota blocks, either.
re #261: the tool you want to use is probably procmail. you can pass mail off to a filter before it is ever delivered into your inbox and the read the mail that passes the filter with whatever MUA (mail user agent, aka "mail reader") you want. Sindi can probably tell you how to set up an elaborate system of procmail rules to try to screen out spam or you can wait until staff have time to install a system like SpamAssassin or other anti-spam package and have procmail use that.
I don't see color in any command (ls, links, lynx, more(less), w3m, etc.) anymore. I forced pine to use color. Still using Putty which means TERM is set to xterm. On the Sun aliasing "ls -color" worked but not on OpenBSD.
Set TERM to ansi. You can do this in PuTTY too.
A friend reports that Pine spellcheck is not working. I tried it and got a message about something alternate and 255. Please feel free to copy /a/k/e/keesan/.forward and also /a/k/e/keesan/.procmailrc but change keesan in the latter file to your own login and delete all lines starting with # and also change my whitelist (the remaining parts with $MAIL on the last of three lines) to your whitelist by putting in the addresses of friends who write you rather than the From:'s that I have chosen to let through from my friends. And email keesan for help. The complicated bit about Nigeria is to send Nigeria spams to a nigeria folder and also to polygon who posts them at his website.
Re 260: I think /usr/ucb is a Sun-ism. It's certainly not SYSV, and the SunOS we were running before was a BSD, pre-Solaris SunOS.
Re #253:
lsz works. One major issue solved, one (or two) more to go.
Party (hayz) is now splitting 98% of CPU usage with smart/asd. Could some staff member kindly delete both accounts?
load averages: 2.70, 2.73, 2.77 18:31:34 138 processes: 3 running, 135 idle CPU states: 94.2% user, 0.0% nice, 5.6% system, 0.2% interrupt, 0.0% idle Memory: Real: 71M/210M act/tot Free: 1301M Swap: 0K/3072M used/tot PID USERNAME PRI NICE SIZE RES STATE WAIT TIME CPU COMMAND 20421 smart 64 0 5668K 5980K run - 953:53 49.37% john 11064 hayz3141 64 0 300K 896K run - 44:26 45.80% party 29978 _mysql 2 0 34M 17M sleep poll 1:29 0.00% mysqld 32358 _syslogd 2 0 164K 488K sleep poll 1:16 0.00% syslogd 3601 named 2 0 2516K 2852K sleep select 1:05 0.00% named 20957 exim 2 0 580K 696K sleep select 0:27 0.00% exim-4.42-2 15359 _pflogd 4 0 512K 328K sleep bpf 0:08 0.00% pflogd 15771 root 2 0 284K 1000K idle select 0:05 0.00% sshd 10834 root 2 0 1092K 1932K sleep select 0:04 0.00% httpd
re #268: quoting % CPU usage is not enough to establish that a user is abusing the system. At the very least we'd need to know the load average on the system as well. When not very much is going on it's not unusual for a single process or a few processes to appear to hog the CPU. It doesn't necessarily mean they're starving other jobs.
Is it considered good manners to be logged on for 954 (minutes, hours?) using this much CPU time? Smart appears to have been logged on since very early Thursday (finger) so I don't really know what the TIME column means.
That's 954 minutes of CPU time. It's the cumulative amount of time that process has used the CPU, which is not necessarily the same as the amount of time since it was started. (In fact, unless the process has continuously used 100% of the CPU, it will always be lower.)
Is that not a lot of CPU time to be used by one user? Until recently that process was using about 97% of cpu time. Could you take a look at what is going on? User smart lists as name 'asd' which is one of many things on my twit filter so this is likely to be some sort of 'joke'.
Yes /usr/ucb is a "Sun-ism". That stands for "University of California at Berkeley", i.e. BSD's original "home". SunOS _is_ SysV, has been since I started using it in 1988. You'll find SysV commands in /usr/bin and BSD commands in /usr/ucb. You'l find the SysV conventional directory setup including /dev/tty??? (which is not in BSD). Here's a script (everytty - which lists all the ports in use by a particular user) which I have run on the old Sun box - it's pure SysV (will even work on Linux which follows SysV conventions). It won't run under any BSD derivative. It's called by "everytty username". #!/bin/sh for a in `find /dev/tty[a-s]? -user $1 -print` do a=`basename $a` echo " $a:" ps -xt$a done Sun did the /usr/ucb thing to add BSD-isms to its OS and allow users to choose which "flavor" they wanted to see by changing the order of their PATH, either putting /usr/bin or /usr/ucb first in their PATH. Hope that clears things up.
http://kb.indiana.edu/data/agjq.html?cust=023362.83043.131 SunOS is a Sun Microsystems implementation of the Unix operating system. Solaris is SunOS packaged with a number of additional tools, and a graphical user interface (GUI) environment. Since Sun Microsystems did not offer the Solaris product until SunOS 4, SunOS and Solaris have different version conventions (e.g., Solaris 1 includes SunOS 4, while Solaris 2 includes SunOS 5). To further confuse the naming scheme, Sun now refers to Solaris by just its point release (e.g., Solaris 7, 8, or 9 instead of 2.7, 2.8, or 2.9). When Bill Joy, one of the main programmers of the Berkeley Software Distribution (BSD), helped found Sun in 1982, he brought with him the elements for the first release of SunOS. Up through version 4.1.x (Solaris 1.x), SunOS remained a heavily BSD-influenced Unix implementation. However, in the late '80s, Sun entered into a partnership with AT&T, which was then developing the other major Unix flavor, System V. The result was System V release 4 (SVR4), which incorporated BSD as well as SunOS extensions (e.g., NFS). Subsequently, with its version 5.x (Solaris 2.x) releases, SunOS shifted from its BSD origins to SVR4.
http://www.bsdforjesus.org/
Our site:
Linkname: Grex Web Server Statistics
URL: http://www.cyberspace.org/stats/
is currently reporting for time ranges no later than:
October 03-October 09 2004
UH OH
OH DEAR>
How do I turn on mail size limits in Pine? All I could find in Setup was sorting by size. I just got a 176K 'file.zip' attachment purportedly from China. In the old grex the per-mail limit was 100K and anything larger would bounce then people would write to complain and be instructed to send me plain text, or just one small jpeg not a 3MB one. Thanks to whoever got the load averages back from 3 to 1. Speedy!
> How do I turn on mail size limits in Pine? All I could find in Setup > was sorting by size. You don't. By the time Pine (or any other mail reader) gets to see how large a message is it's too late; the message has already been delivered to your mailbox and then read by your e-mail program. If you want to reject all mail above a certain size you can easily do that with procmail before it gets delivered to your inbox. A recipe like "* > 100000" will match messages over 100,000 bytes. Once you match them you can decide what you want done with them.
You mean something like: * > 100000 /dev/null (I am not sure if the :0: on my first line got into this response). Why is grex no longer placing a limit on mail size? Someone else complained recently about having to empty spams from his mailbox several times a day or it would fill up with junk like this. Or maybe it is a virus. I only know how to filter on headers (:0:) or message body (:0B:). If grex is going to allow these large mails to get through now, could someone possibly write up a script to simplify rejecting them with procmail, which is not for beginners?
For some reason, the script in #274 didn't work for me until I added a
line:
#!/bin/sh
for a in `find /dev/tty[a-s]? -user $1 -print`
do
b=`basename $a`
echo "
$b:"
a=${b:3}
ps -xt$a
done
(Apprently, in ps -t, the 'tty' is implied, so including it in the
variable results in the shell attempting to parse 'ps ttytty[a-s]?'
rather than 'ps tty[a-s]?')
Heh.
This response has been erased.
I cannot telnet to grex. I type my login and then nothing happens, or sometimes I can get as far as typing my password. I had to use backtalk (vanilla, lynx) to post this.
The telnet problem just fixed itself (or someone fixed it).
Grex is now immediately disconnecting when I dial in direct.
I just dialed in direct 45 minutes later. How do I change Pine's behavior back to letting me decide whether to forward a message as an attachment? I usually say 'no' so that I can remove most of the message by leaving it as message body (such as all the email addresses in it). I searched on 'forward' in Setup with no luck. Could this be put back to default behavior, along with 100K mail limit?
This response has been erased.
This response has been erased.
Sooner or later, we'll get the configuration of exim right. When we do, a limit on size will probably be included. Asking for it repeatedly will not make it happen any sooner.
Okay, I will try to clean out my mailbox more often until it happens. Do we still have a 1MB mailbox limit? 176K spams or viruses fill that fast.
I cannot access 7 or 8 ftp sites in a row including ftp.slackware.com. I am a paid member which means I should have ftp access and these sites are accessible from another shell account elsewhere. Do I need to change something in lynx.cfg or somewhere?
DITTO Looking up ftp.slackware.com ftp.slackware.com Making FTP connection to ftp.slackware.com Alert!: Unable to connect to FTP host. Looking up ftp.slackware.com ftp.slackware.com Making FTP connection to ftp.slackware.com Alert!: Unable to connect to FTP host. Can't Access `ftp://ftp.slackware.com/' Alert!: Unable to access document. lynx: Can't access startfile
resp:291 Thanks for installing _some_ of the GNU stuff. If we don't have mc (Midnight Commander) how about installing git (GNU Interactive Tools) as a file manager. Of course mc from "ports" would be better, hint :-)
I can now do a Kermit telnet to another shell account but kermit file transfer between them does not seem to work. I had to exit the telnet connection and do ftp instead. I think it used to let me kermit telnet and then file transfer between these accounts. (I had gone off to another account to get a file at an ftp site because lynx won't work now to get files from ftp sites).
This response has been erased.
Re. #119: remmers wrote:
"Got to oldgrex and looked;"
How does one get to oldgrex?
This response has been erased.
I don't know why ftp wasn't working before, but is now: ] Respond, pass, forget, quit, or ? for more options? !ftp ftp.slackware.com ] Connected to ftp.slackware.com. ] 220 207.173.172.131 FTP server ready ] Name (ftp.slackware.com:gelinas): ^C! ] ] Respond, pass, forget, quit, or ? for more options?
ftp to grex and (binary) get seems to have *really* screwed up ... staff has mail ....
ftp to ftp.slackware.com works but I am trying to use lynx lynx ftp.slackware.com That still does not work. I will try to figure out how to use straight ftp (login anonymous password Enter?).
login anonymous, enter your complete email address as password, worked. But ls does not work - Entering extended something-or-other mode. It is easier to fetch files with lynx/google from anonymous sites like this.
Its even easier using 'wget'. .dunno if its instaled here.
I think my problem with not being able to using kermit file transfer between sdf and here is that I was telnetted to grex at the time. Just now I was able to dial in to grex (with kermit) and then ssh to sdf and then do a kermit file transfer and amazingly the file did not get renamed L12344.htm but kept its original name and ended up not at grex but at my own computer. Lovely! In order to fetch a file with wget I would first need to use lynx or links or w3m and do a google search to find the file name and path, then exit the browser to use w3m (or at least shell out of it). Simpler to download the file with the browser so I hope this can be fixed. If I am telnetted to grex I guess I will just have to use ftp to move files from sdf to grex. Instead of having to exit my telnet connection to do this I can use the 'screen' program which lets me do several things at once at grex such as telnet in one screen and ftp in the other. (Or if I am using linux instead of DOS to telnet in the first place I can ftp from another linux terminal). It will still be nice to get lynx working again at ftp sites but there are a lot of ways around this. The large email problem should certainly be attacked first for the benefit of the grexer who complained of getting several a day and having to empty the inbox several times a day.
For what it's worth...I found that grex now requires sftp-2.
What is sftp?
SecureFileTransferProtocol. It encrypts the transfer session.
ftp.slackware.com is not an HTTP site, so to access it through lynx you need to use a complete URI -- ftp://ftp.slackware.com.
I can type lynx ftp.slackware.com at my shell account at sdf.lonestar.org and it works. Here it does not work, nor does it work to click on a link to the site from another site. I don't recall having this problem at old grex. Thanks to whoever fixed the pine settings so that it now asks again whether to forward a message as an attachment so I can answer N and edit it first.
lynx ftp://ftp.slackware.com gave me the same result as without the ftp://. Unable to access ftp site. Perhaps it only works for staff?
links ftp.slackware.com at sdf - connects immediately. Here at grex 'No route to host'. w3m is taking forever here and at sdf so I stopped waiting. Odd.
It doesn't work for me, either. Now, a question: why use a web browser when something designed for the job, to wit, ftp, is available? That's a lot like using a spoon to cut something while holding a knife is in your other hand.
Because M$ has managed to convince everyone that the Windows philosophy* works better than the Unix philosophy? *To wit: Do a zillion things, and do them as badly and inflexibly as possible.
There are two secure general purpose file transfer protocols: ftps or ftp-ssh -- Plain old FTP over a secure connection sftp -- a new protocol with an unfortunate name that makes it sound like it's FTP, but it's not. Other than the name, it has nothing to do with FTP. I'd thought all of the free world had agreed to have nothing to do with sftp. However, my knowledge of these two protocols does not go far beyond what I have written in this response.
re #314 Some of us like to use cursors rather than type a command for every detail.
not really a problem but the Grex Faq says it's running SunOs.
Yeah, updating the web pages is on my list of things started.
OK, I tried fronttalk. Because I'd just read stuff with picospan, there was nothing new to read. So I tried "read since 1/9" (and subsequently "read since 1/9/2005"), and it gave me errors 'Don't understand "/9"' and 'Don't understand "/9/2005"' (respectively). After the errors, it jumped me into the middle (!) of old stuff (item 1 response 118), for some reason.
When I tried "read since 1/9/2005", it gave the error Dave reports but took me to the beginning of item 1. When I tried "read since jan 9 2005", it took me first to item 9 and then item 1. I suppose we should read the ft help.
Re: #314. Those models are still around, they have their uses. Any movement on the problem I've reported twice yet?
resp:298 Thanks - mc works great!
Re resp:316: I know a fair number of sites that use SFTP exclusively, to get around the problems of FTP -- mostly, its insecurity, and its inability to play nice with firewalls. FTP is still used for anonymous file access sites because SFTP doesn't make any sense in that situation. That usage seems to be slowly disappearing in favor of HTTP, though. You're correct that SFTP doesn't resembly FTP much, protocol-wise. It's really an extension of SSH. It is, however, a File Transfer Protocol. ;)
And client user interfaces are usually designed to make it "look like" ftp.
Yes, if you're used to using FTP graphically. If you're used to using FTP on the command line, you're going to notice differences.
I was actually thinking of command line interfaces. But we digress.
Regarding using ftp instead of lynx to fetch files from ftp sites, that is like putting down your spoon and going off to look for a knife. I am already using lynx to find the file I need and it would be simpler, rather than writing down the full pathname and filename and exiting lynx and then typing it all into ftp (or even using 'screen' and copying it that way) to simply fetch the file with lynx, or links. Links is even set up to let you download a file while continuing to browse so no need to use 'screen'. In this case I was trying to find a small DOS program for the Bulgarian neighbor to use to type in Cyrillic and I found a few but they were at ftp sites (file archives).
zerofill is running multiple copies of vvopenbsd since 11:11 using up all the cpu time.
I haven't noticed any difference using sftp rather than ftp - but then, I'm just using things like ls, rm, put, and get.
Thanks to whoever just fixed the vandal! Load average down from 60 to 0.0.
<remmers bows>
This response has been erased.
I connected via ssh a few minutes ago and got a garbled mess following entering my password. I'm dialed in now, although I also got that garble on my first dialin attempt. It seems to be working OK now.
I'm experiencng a *huge* spam problem. After beingoff the system for 3 days, I had over 300 e-mails waiting! I never had this many with Old Grex. Can someone move over whateever bilderfilter OldGrex was using? The present situation really cannot continue. Mail is virtually useless to me know.
Re 322: Which problem is that, twenex? Re 334: That's a problem of not getting a tty. I don't know why it happens, but I've had it happen a couple or three times. Re 335: No, the mail-filters that were in use on the old machine cannot be moved over. We have to recreate them.
Backtalk refuses to recognize that I've read some responses in agora. It always comes up with "X new respondes, 0 read" where X = the total number of items.
http://www.xnewrespondesforjesus.org/
Anyone who wants is welcome to copy over my .forward and .procmailrc, change keesan to their own login, delete all lines starting with # and replace my (now very short) white list (sections ending in $MAIL other than the one at the very end) with the addresses of friends they know will be sending them mail, and be prepared to check the log every day and write anyone you missed whose mail got thrown out. Eventually everyone will be on your white list. I just deleted my white list and am rebuilding it but fixed one problem with the filter that was throwing out all mail from AOL because it filtered on 0pt (as in 0pt 0ut but also 10pt, found in a lot of HTML). I am now getting only about 3-5 spams/day and losing a couple of real mails - just put the sender on your white list and ask them to mail again - the log file shows you who wrote. I zap about 40 spams a day.
So - how soon can we re-create the mail filters that were used on OldGrex?
I'll try to take a look at them this evening.
Thanks, Joe. Today my spam filter threw out mail from two of four friends and one of the other two was on my white list. I have added too much to it in a hurry. I would appreciate suggestions on how to fix it: /a/k/e/keesan/.procmailrc. (Ignore any lines with # - the old whitelist).
I just discovered that non-spam can contain 3Dhttp when a helpful friend who uses AOL sent me a mail with that in it. (She has been saving her mails and resends them so I can run them through my filter after putting her on the white list). What does this do? I removed a few other 3D filters too.
I tried to fetch an ftp file with wget: failed.. No route to host.
Which sounds like a problem which wouldn't be limited to wget. If you try to ping the same host do you get a similar error or does ping succeed?
ping command not found I was looking for alternate ways to fetch ftp files since lynx ftp and links ftp are not working yet here. I was able to use straight ftp at this site (simtel) but not: lynx, links, wget. Just now I tried to look at my energy bill www.mydteenergy.com with links, which works for this purpose (well, part of it works, I can at least sign in and look at one of my two bills) at sdf, but at grex 'no route to host'. Gelinas tried to fix this problem during beta testing but I don't think succeeded. Yes I am a grex paying member with ftp out privileges. I probably typed it correctly because substituting ftp for wget and removing the file name worked to get me into the correct directory.
Progress on making a better spam filter. An AOL-using friend helped me to determine that non-spam AOL mails can contain, without the sender's knowledge, 3dhttp <td (as in tables) and 'click on', none of which were visible to me. It might be simplest to just put all people I know using AOL on the whitelist since they can't figure out how not to send HTML.
(birthday wisher broken?)
My mistake. The ping binary is around in /sbin/ping but execute permission is reserved for root and staff.
If there are questions about setting up Exim I can try to assist. I've done quite a bit with Exim, though never on a system that gets the kind of mail volume Grex does.
I fixed talk. The reason I didn't port the oldGrex spam filters over is that I was unable to discover what they were. I don't have either source code or documentation for them. I have a very long list of things that need fixing and very little time to work on things. I appreciate your patience.
Many thanks for talk. Yesterday I had two long exchanges of info via tel (which is even more confusing than usual when grex is running at load average 60 or the language of communication is Russian in the Roman alphabet).
Is it your birthday, Anne?
Lynx still offers me the option of sending a downloaded file to my computer via kermit/telnet but I don't think that ever worked. Could lynx.cfg be changed to reflect reality? Or grex be changed to allow kermit file transfer via telnet? Maybe it once used to be slower than ftp but it is just as speedy now and often more convenient than saving a file to a home directory, and then starting a new connection to grex in order to ftp it. I was downloading a few photos of someone I had a one-hour tel exchange with yesterday who had ideas on how to improve procmail filtering. (His idea was to throw out all mail from aol, yahoo, msn, earthlink - he said this gets 90% of the spam).
Re#353- Yup. :) I'm 30 now!
You're more than ten years older than me!
Happy Birthday !
The "no route to host" message is often a result of the packet filter blocking outbound packets. OpenBSD's packet filters don't look at all the groups a person belongs to, so we have to link programs to "inet_wrap", which then sets the group appropriately and calls the real program. wget is not on the 'approved' list. links is, but it does not seem to be set up properly. The problem with DTE appears to be malformed html: they don't properly separate the file name from the host name. I don't know why links on sdf can parse the link. If I recall correctly, links reports the name of the host it cannot reach. If you look at that domain name, the problem should be obvious.
Links at sdf works on the dteenergy site on the second try (but the site has other problems so I may give up on it). I will use straight ftp until lynx and wget work again with ftp sites. DTE says they asked the company doing the website to fix some problems but it just keeps getting worse instead of better. Thanks Joe for spending so much time on fixing these small problems.
I fixed the birthday-greeter. Thanks for reporting its failure, Anne. And happy birthday. :)
re #356- Yeah, nyah nyah! ;) re#357- thanks. :) Re #360- you're welcome, and thanks! :)
Backtalk authentication problem: Using the vanilla interface, some of the links at the bottom of a page when browsing new responses to an item are formatted such that even if the user is connecting via https, the link will be formatted for http, which results in a browser asking for cleartext authentication before submitting a "forget" request, for example.
Backtalk authentication problem: Also, response 'Post' buttons submit via http rather than https.
Re way back (out of my scrollback buffer) re my problem with ft "read since": Joe, I did read the ft help. (It didn't differ from my memory of Picospan help on this issue, but it's been a few years since I looked at that.) "help read" referred me to (I think) "help range", which gave the format I was trying as an example.
My mailbox this morning was full because a friend sent me two photos, one of which was about 400K and the other a bit smaller. Please could the mail size limit be given a high priority? McNally, until then, could you explain precisely how to set up a .procmailrc that will dump such mails? I did not follow the last explanation. For the benefit of all email users here, please also include the first three lines of .procmailrc and the last two. .forward is, I think |/usr/bin/procmail right? This sends all incoming mail to procmail to be filtered. Maybe someone could run a script to set up this sort of mail size filter for everyone until the exim one is working? Why would 'cpl' dump a returned message from me?
is there any sort of filtering mechanism creatable (by user or system) that can mostly de-spam the stuff in 'plain ol mail'? same as dpc and keesan, i am *really* getting slammmmmmmed/spammmmmed!
If I could just filter on "cum", it would take care of about half my spam.
> My mailbox this morning was full because a friend sent me two photos, > one of which was about 400K and the other a bit smaller. Please > could the mail size limit be given a high priority? > > McNally, until then, could you explain precisely how to set up a > .procmailrc that will dump such mails? Try adding a recipe like this to test with: :0fw * > 50000 /a/k/e/keesan/bigmail Then send yourself two messages, one small one, one with at least 50 kbytes of content. The larger message should wind up in ~keesan/bigmail and not in your inbox. Once you're satisfied that it filters as you expect you can change the destination file to /dev/null.
367: Do you get a lot of spam in Latin?
I get a lot of spam with utf or iso on subject lines and just now one of my two latest spams was about a cumshot but that is only about 5% of the spam. Is there some way to set a filter to forward mails over 100K to another address instead of a folder here? They are usually photos from friends.
Thanks Mike, that worked and I am now sending mails over 100K to /dev/null. I notice that in my filter I am sending Larry Nigeria spam with the string ! polygon@.. WOuld I just put something like that in instead of the /dev/null on the third line to forward big mails to myself at some other address? If anyone else wants to use Mike's filter, be sure to also include the first three lines of my .procmailrc and the last three lines (along with this filter which is now in there) so that the rest of your mail will go to your inbox, and change keesan to your login. The only problem with this method is that the mail does not bounce back to the sender so they don't write and ask how come the mail bounced, but you can check the log file daily and write people you know whose large mails bounced. You also need the .forward file so that mail will go through procmail.
Minor problem with Mike's filter: Extraneous filter-flag ignored. Does this mean one of the characters in :0fw should be removed? The filter works anyway, letting through a small mail but diverting a big one.
From the Procmail Quick Start (http://www.ii.com/internet/robots/procmail/qs): Matching a Word If you want a recipe to match only the word test, rather than the string test, surround test with the Procmail word delimiters \< and \> as in the following: :0: * ^Subject:.*\<test\> IN-testing [end quote] So to do what Rane is looking for would be :0: * ^Subject:.*\<cum\> /dev/null as a Procmail rule. I highly recommend that anyone who wants to use Procmail for filtering (whether spam or otherwise) read the Quick Start; it is very well arranged to walk people new to Procmail through setting it up.
Thanks for that info. I have not needed to filter on cum because mail containing it includes so many other easily filtered strings. I removed fw from Mike's filter and it still works and no longer tells me there is an extraneous filter-flag: First line :0, second line * > 100000 third line /dev/null (I cannot type a colon at line beginning here).
The best I can tell, you don't need either the f or the w flag in the size-catching recipe. Those are only needed when the action of the recipe uses an external program, from what I can find.
Note that the recipe I gave will match "cum" *in the subject line only*, and *only as a word by itself*. (For example, I didn't want to match "cumulative"...)
Handy recipe, thanks. I just replaced /dev/null with ! followed by my other email address and the log file now tells me it sent my mail over 100K to Folder: /usr/sbin/senmail -oi keesan@ (my other address). Does this mean it was forwarded to that address? It was not there when I checked shortly afterwards. mailsize was 267K. But if Grex puts a limit on outgoing mail size will this no longer work? At that point I guess I could just remove the filter and let friends write to ask me why their mail bounced. Or is the limit only on mail going into the inbox, not mail being forwarded? Is there some way to filter on words starting in c.al so as to dump mail with variations on cialis but not special? Perhaps we should have a new mail filtering item, at least until grex gets a spam filter working again.
I was able to forward the large mail to my other address with First line: :0 Second line: * > 100000 THird line: ! (my other address) It would be nice if this continued to work after grex got back its mail size limits so that big mails would go to me instead of the sender.
Sorry about the extraneous flags, I just quickly adapted from another recipe in my .procmailrc and since I'm at work I didn't have time to test fully. Glad it helped, though..
Yes, you can filter on words starting in c.al, but that will catch (for example) "challenge"... If you really want to, it would be "\<c.al". (The \< and \> don't need to be matched, any more than ^ and $ need to. \< means the start of a word, just as ^ means the start of a line, and \> means the end of a word, just as $ means the end of a line.)
So how does \< differ from '\ ' (a blank space)? TO whoever said r and R behave differently in pine (one replies to everyone and one to just the sender) they behave the same for me. Maybe you have configured your Pine to act differently. I was expecting my new filter to forward a mail with attached .tif file someone said he just sent me here, but he forgot to attach the .tif file.
Could we find another item to discuss spam filter rules and procmail syntax? I'm scanning this item looking for system bugs to fix, and wonderful though procmail is, it isn't really in that category, and it's about half this item. Thanks.
Sorry, I suggested that and then continued to answer the questions. I'll stop now.
I fixed the https/http problem in vanilla. Backtalk's "read since" logic has been flakey for a while. I need to revisit it.
> TO whoever said r and R behave differently in pine (one replies to everyone and one to just the sender) they behave the same for me. < I didn't say that about pine, I said that about mail. Yes, the little ole, stupid, command line oriented vanilla mail reader/sender.
Is there some way to get Pine to stop offering to reply to all recipients?
No, there isn't, Sindi. Since it doesn't have separate commands for "reply to all" and "reply to sender," it has to ask every time. Dave, my comment on reading the ft help was directed as much at myself as it was at you. I also read it, without finding the answer to date ranges. :(
JANC!! HOW TO YOU GET PARTY TO HOG CPU LIKE HAYZ DID!?! THANKS
load averages: 7.91, 6.84, 4.86 00:50:38 182 processes: 6 running, 172 idle, 3 stopped, 1 zombie CPU states: 34.3% user, 0.0% nice, 65.1% system, 0.6% interrupt, 0.0% idle Memory: Real: 76M/219M act/tot Free: 1293M Swap: 0K/3072M used/tot PID USERNAME PRI NICE SIZE RES STATE WAIT TIME CPU COMMAND 14640 azure 64 0 1460K 1680K run - 64:46 33.69% nethack 29978 _mysql 2 0 34M 17M sleep poll 4:35 0.00% mysqld 32358 _syslogd 2 0 164K 484K sleep poll 3:38 0.00% syslogd 3601 named 2 0 2620K 2968K sleep select 2:55 0.00% named 20957 exim 2 0 580K 696K sleep select 1:17 0.00% exim-4.42-2 whoa
whoa
This response has been erased.
cpu usage about 96%: nethack and party - 3 processes split 3 ways.
If you don't calm down, I'll split you three ways.
Thanks for the hint, Dan; reduces the guilt of killing processes. :)
> #11 of 30: by JERKS (qqqqqqqqqqqqqqqqqqqqqqqqqqqqqqq) How is it that the grex userID, shown between the (), can be more than 8 characters? Is that an edit that Backtalk allows?
The new Grex operating system allows account names to be up to 32 characters long, apparently.
Since I don't recall that being announced before or after nextgrex was implemented, I'm wondering if that is something we can arbitrarily enforce. There is no reason that we can't keep the newuser mechanism from limiting IDs to 8 characters, even if the O/S doesn't have that restriction. I'll take this up in coop.
Doesn't seem like a problem to me... my usual PhPBB nickname is around 10 characters long.
Error writing to /tmp from lynx so I could not send mail until I changed the path to my own directory (of the file I was writing). Disk quota exceeded. What does this mean? I don't think I have anything else in /tmp but if so how would I check this? (I sort of recall trying to download something yesterday via tmp that was too large).
I searched /tmp for zip files and found the one that I got about 1MB of yesterday and deleted it. How often is /tmp emptied? (My usual 'ISP' - WCC - was dead for a while yesterday and I needed DOS xpdf in a hurry but there is also the free access-4-free.com for such occasions).
I still have the via menu notice that says I have mail, and when I check it with !pine, I'm told i have zero messages. I checked, as far as I'm able, with my .login and .cfonse(?) files ,and i can't see ANYTHING that would cause it. Any suggestions?
lowclass, try "!less /var/mail/lowclass" I think you'll find that the file is not empty, but it's not something that pine will show you. If the file is not empty, login reports that you have mail.
GREX just developed a bad participation file.
s/GREX/Agora/
http://www.agoraforjesus.org/
Grex keeps freezing up on me. And I think the participation file got trashed.
-bash-2.05b$ party
Welcome to PARTY! Type '?' for help:
/var: write failed, file system is full
---- nharmon leaving (Jan 15 16:54)
nharmonthefag: take the time to find out
nharmonthefag: i am going to contact your job and see if grexing while at work
is considered to be what the company had in mind when they hired
you
and then show them the logs where you insult your boss
nharmonthefag: and ask them if that is also what they are paying you to do
nharmonthefag: something tells me it's not
nharmonthefag: i will let them decide
nharmonthefag: maybe they will give you a raise or something
---- alistair joining (Jan 15 16:57)
---- nharmon joining (Jan 15 16:59)
romantic:
---- mfp leaving (Jan 15 17:25)
/var: write failed, file system is full
-bash-2.05b$ tel zoezig77
Telegram to zoezig77 on ttyq8...
Msg: the place where it keeps its logs is full! so, it's broken!
SENT
/var is full again.
Mail not sent, error 451 writing spool file - is that on /var? I could Postpone the message at least.
Mutt is working today but pine seems not to be - I just get a blinking cursor.
Lynx google: Can't access start file (it did try google.com). ssh or telnet: Temporary failure in name resolution. And the only two mails I got were returned when grex could not send them due to a problem in finding the host. (No spams in at least 12 hours).
I just wandered down to the pumpkin and power cycled the DSL box. Looks like we're back on the 'net.
Sorry, but as of now we're still off the Net. I can't reach www.cyberspace.org from either my wife's machine or my own.
Multi-minute lag. It is no longer optional to find a way to throttle bandwitdh on the DSL link. It is essential.
Massive, crippling network lag. This is the worst I've seen it in a long time. Are we ever going to get the DSL connection fixed?
what gull said.
Won't the colocation take care of DSL problems?
Yes, it will. Is that planned for the near future?
grex is freezing a LOT for long enough to invoke the idle killer..*excep** that logging in elsewhere shows no idle time ...
wanted first reposne to be written before this connectoin dies of some internal delay mechanism. uptimnes are low, low, low and top shows no adverse sitau6tion that i recognize.
soemthing, is bogging grex BigTime (tm).
I don't know what is going on. I've had to kill several connections in the past hour. I can get to another machine in the pumpkin, without problem. Maybe things have cleared up.
The connection problems seem to extend to incoming mail. It just took 5 minutes to get the previous line to echo.
Further information: The lag is on a per-connection basis; while an ssh session is locked up, I can ping Grex with sub-200 msec ping times and establish another. So, something is hanging individual connections. This points to a stack bug or configuration issue.
Terrible lag on ssh, no lag on dialup.
Could we get the mail system to do without that silly MAILER-DEMON Do Not Delete This Message file that it insists on having? The mail system on Old Grex managed without it.
terrible bogging on direct telnet as well .. dunno about dialup (so far).
I'm having *serious* lag when I come into the conferences through
backtalk. At about 9:55 a.m. this morning I got to the "Welcome
to Grex's Backtalk Conferencing System!" page without too much
delay. I pressed the button "Enter Conferences". The system
was trying to reach ~/cgi-bin/pw/backtalk/pistachio/begin", but
it took *nearly two minutes* for the "Connect to" box to pop up.
I entered my loginid and password, and there was another
very long delay before I was able to reach the entry page for
the conferences.
Is this "normal" behavior?
Long delays still occurring.
yup ...... still - any staff have this info yet?
Temporarily disabling the automatic idle log-off program, might be a useful palliative measure. Even though not treating the cause, it would at least treat a major symptom.
Another fact: While my connection looks "stuck" and is failing to echo, I can type and watch packets go out and return in sync with the keys. The problem looks like a dropped packet without a proper retransmit.
Welcome to the Winter 2004/2005 Agora! Your Host: Katie Geddes New Users: Welcome to Grex! This conference has been restarted. Type "join fall" for the old version. 80 brandnew items First item 1, last 81 Is it just me, or did everyone have their "what have your read" info trashed?
Yea, I had to do a fix and "r since jan 17"
Mail to Grex is starting to bounce because of the network problems.
Iam dialed in and the speed is great and pine works but still no new mail. Any chance of moving early to provide.net?
I can still only use Grex via dialin. Backtalk never really gets anywhere, ssh connections die slow and painful deaths. Anyone have a clue what the issue is?
Same question - would staff please let everyone know what is going on, and the prospects for fixes?
This response has been erased.
blindingly fast right now.
(I shouldn't say blindingly.)
Mail continues to bounce due to inability to get SMTP connections. Backtalk is timing out on both anonymous and "login" access attempts. Grex is effectively unusable by any means of connection, for all purposes. Re #439: That explanation is certainly wrong. I can send pings and open new connections even while one connection is hung. The link is up, but something is happening to halt traffic on individual connections. This is probably why SMTP is failing. If nothing is done to fix the problem of connections hanging, a move to a faster line at provide.net will not help anything.
This response has been erased.
Then why doesn't retry fix the problem with the hung connection? I can type things and see echo packets all the while.
This response has been erased.
Dan, what do you do on gryps? Anything other than ssh on to grex? I've not been doing anything strenuous on groupie, but it does seem responsive when grex is not. Sometimes, I've been able to open a new connection to grex while the old one is hung. Sometimes, I've seen one connection hang and another not. I'm not convinced it's a problem affecting all machines on our network; I've insufficient evidence. To those complaining: yes, staff is aware of the problem and is trying to figure it out. We simply don't have anything else to report. :(
I am telnetted and grex is usable at the moment (for bbs, anyway).
I just got my first mail in about two days here so perhaps things are fixed. (I hope my answer got sent okay). There was about a 20 sec freezeup while I was answering it but otherwise just the usual telnet lag. (I have to slow down my typing speed to about 2/3 usual).
Another mail just arrived
Re 426: the "Do Not Delete" message is an artifact of Pine. If you don't use Pine, delete it. If you do use Pine, you won't see it.
Three more mails arrived - is grex cured? I could also send mails out.
Pine works and I think I replied to a mail, but a mail forwarded from here elsewhere did not arrive. I can now ssh elsewhere too. (I am dialed in). If someone fixed grex, thanks. If not, let's hope it stays fixed for two more weeks.
GreX is fast and then locks. That is all.
grex is fast and fine as of this entry ... whatever was done - thankxx!
Yup, enough of the problem has been fixed to allow mail to flow again.
oh, did i type too fast in #453 .... bogged & idleded & disconnected about 3 minutes later ...
It seemed suspiciously like a path MTU discovery problem of some sort. Pings and such worked, but anything that used larger packets failed.
right. number of bytes.
Yeah. Initially I thought I was having a rerun of an MTU problem I used to have on my own connection, but then I noticed every site I tried except Grex was working fine.
The lag problem and the mail-delivery problem appear to have been fixed completely.
Whoa.
wanna bet? took nearly 30 seconds to establish the telnet connection over my eathernet from work.
The system is deadly slow again, both via putty and Backtalk. I assume someone is aware there is a problem, but there's very little information, it5 seems, about what is being done about it, or if anything is being done. Any idea abuot what is causing the problem?
This response has been erased.
Still very slow.
BAcktalk is barely responding at all. Half the time the pages don't load and when they do it takes five minutes or more. I couldn't even get to the agora screen, I had to telnet in finally just to post this
How long will it be before we move to colo? Is it really not worth even trying to get the upstream problem resolved in the mean time? Right now I'm even getting intermittent lag on the *dialin*. This morning dialin was giving me a 1-2 second delay for each character.
Re. #463 & #466, postings by cross and gull, referring to "a problem with an upstream ISP": Does "a problem with an upstream ISP" mean: unusual use or misuse of the dsl service by some other customer of the service in such a way as to limit Grex' normal use of the service?
What does an upstream ISP have to do with slow dialin, which I just had for the past 10 minutes (wait 20 sec to see the cursor move)? If the DSL line is the problem can we cancel wtihout having to pay for the rest of the contract?
This response has been erased.
#469...so you are saying its just a *coincidence* that these major lag issues occurred with in days after grex moving to the new server? The timing has to make you wonder. maybe grex would be better off back on the old server for the time being. maybe nextgrex can't handle its job
This response has been erased.
Conditions have deteriorated the point that mail delivery seems to have halted some time Thursday morning. Lag is bad now and I jave had to drop one connection because it was stuck.
Dialed in. Load average 028. It just took me three tries to type that number and I have to wait 10 sec or so to see what I typed appear onscreen. 0.28. Sometimes things go full speed, sometimes a great deal of lag. No mail came in recently (for a few hours). How would ISP problems affect dialins?
(richard, we've been dealing with network problems since last summer, long before the move to the new machine.)
re #474: > we've been dealing with network problems since last summer the recent problems seem qualitatively different than the ongoing outages that were occurring previously. re #470: > maybe grex would be better off back on the old server for the time being. That would take a major amount of work to move back to the old server and then another major amount of work when the new server was judged to be ready for prime time. Worse, one might complete all that work only to find that the source of the problem lay elsewhere, as seems to be the current prevailing opinion among staff.
#474...gelinas, what mcnally said...the problems have clearly gotten worse, much worse, since the switch to the new server. I understand the work that would be involved in going back to the old server, but how else can the new server's role in the current problems be conclusively proven?
Regarding the particular variety of intermittent screen freezing (if there are different kinds) that we have been observing recently, when I'm logged onto Grex and my screen happens to be frozen, are all the other screens of other logged on Grex users also frozen <<at that moment>>, or at any given time are some Grex screens frozen and some not frozen?
Please have some patience, folks. Please. There are multiple reasons for the problems Grex is having and staff is as anxious to get them under control as everyone else. It's my undertanding the machine is moving to co-lo this weekend, if at all possible. Some of the lag will be helped when robocop is up and working and Jan is working on that. And the setting are still being fine-tuned but doing so under the current lag makes that all the more difficult. Staff is not ignoring Grex right now, really, they're not. Let's try to be supportive - especially now, when they have so much to do. I suppose we could go to our DSL folks and insist it's their fault and consider not paying the bill, but first we'd have to be sure it's their problem, entirely. I'd rather we simply work on a solution and get Grex working right.
Re #477: Some are frozen and some are ot frozen. (Speaking from experience here.)
In other news, there appears to have been no mail all of yesterday afternoon, evening and last night.
And this morning.
The problem before was frequent sudden freezeups for mayb 30 seconds while telnetted. Right now (I am dialed in) things are okay for a minute or so then I will type something and wait 5 sec for half a line to appear, then after 5 sec another chunk appears A 5-10 character string will appear, then in 5 sec a few more characters, and then suddenly it all works okay again. This is entirely different from before. Is there some way to hook up any other grex-type setup to the same DSL modem and phone line to see if the problem carries over? (An httpd server? Telnetd server? Do I have the terms right?). At this point nobody would be terribly inconvenienced (more than currently) if grex were off the air for an hour or two to do such a test. Can some other computer (say a laptop) be plugged into the serial port via a serial cable to see if it can telnet to grex? Or work as a dumb terminal? (I even had this working on my little linux setup once, both ways).
It's funny how m-net's bbs is getting more activity.
This response has been erased.
Re. #479 in which russ wrote:
Re #477: Some are frozen and some are ot frozen. (Speaking
from experience here.)
If not all screens are frozen at the same time, then would it be possible
for an online user whose screen is <<not>> frozen to run a Unix script
that would unfreeze the screen of another online user whose screen <<is>>
frozen?
We will move the machine tomorrow, which should, we hope, solve this problem.
resp:486 Thanks Joe! Are you going to do that in the snow?
I ignore snow.
yea to the keepers, fixers and movers of grex, your hard work is appreciated
=^o.-^=
Grex is working relatively okay at the moment (the usual telnet lag) but no incoming mail.
Would staff please check out www.badcaps. net to see if the freezeup problem is related to bad capacitors? It can Many companies have been putting in defective caps since 2000. Smptom is 'random and frequent freezes, or freezes under heavy board activity'. We revived one dead board by replacing 6 cheap caps. Flaky power supply could cause similar problems - measure voltage. Measure cpu temperature (we have a remote thermometer). Jim fixed two such boards and is free this weekend. What manufacturer and model board? What brand are the large capacitors neqar the power supply (probably four in a row).
Grex has been moved, to judge by the new IP address, and seems much happier. Thanks, staff!
The problems with Grex are not related to bad capacitors. The story of them has been blown completely out of proportion compared with what has actually been experienced. Motherboards with weird looking capacitors are most often the fault of poorly made units, and not the dreaded improperly built type that do exist, but not in the amounts reported. Thanks to the staff people who moved Grex. Certainly an order of magnatude faster, if not more than that.
Joe moved Grex during a blizzard to keep the rest of us connected and happy. Thank you, Joe.
No kidding. Thanks a bazillion Joe!
Wow, on time even! The board we just fixed had several bulging capacitors out of 10. Jim replaced 6. The bulging ones tested okay, one that looked okay was completely shorted. You cannot tell just by looking at caps. Obviously grex is much happier at this new address. I still get the usual telnet lag but it can nearly keep up with my typing now. About a 2 sec wait for the cursor to move down one line. Well, maybe it is half as fast as my typing.
I don't think the telnet lag is any worse than it was last year so the problem that was plaguing us at the pumpkin seems to be fixed. I am using a 56K modem (which tends to connect at 33K or at most 39K) and dialins are generally alot faster. Joe, many thanks for your work on this.
Backtalk is working fine for the first time in a few days. Thanks Joe! for the move.
Backtalk is working great right now. Hail to the staff, particularly gelinas who moved Grex to the co-lo! Thanks very much!
Grex is almost as fast as M-Net now. Thanks Joe!
hey what do you mean "almost"? Lets have a contest, pick a file of a specific size, and download it to grex and mnet at the same time and see which chokes it down the fastest. I'd bet five bucks on Grex. :)
Yea, lets see which one goes over its monthly bandwidth quota quicker. *snort* Good idea, Richard.
This response has been erased.
Let's hope folks don't turn Grex into an ftp warez site since it has a bandwidth quota.
This response has been erased.
Telnet will not access grex right now - I tried two different programs. One told me it is not receiving the expected responses. ssh worked instantly.
Pine is not working again - the cursor just blinks. Mail works. Pine worked this morning when telnet worked. I recall it not working a week or so ago. I did not test mutt.
Whoa! -bash-2.05b$ w 9:17PM up 16 mins, 4 users, load averages: 0.32, 0.47, 0.44 USER TTY FROM LOGIN@ IDLE WHAT jor p4 adsl-69-212-43-1 9:13PM 1 -csh charcat p5 64.31.11.168 9:15PM 0 /usr/local/bin/party_ krj p6 dialup-4.229.123 9:16PM 0 /usr/local/bin/party_ bookie p8 4.229.180.231 9:17PM 0 /usr/local/bin/bbs -bash-2.05b$ party party_ error: Cannot find your tty (/dev/ttyp7) in utmp I'm here, but I'm NOT here.
Not anymore :(
Attempts to access http://www.cyberspace.org/ from the Grex Lynx command
line yields:
Making HTTP connection to 216.93.104.35
followed by:
Alert!: Unable to connect to remote host.
The web proxy is still on the old network connection.
re #507: correction -- Pine *does* work, but when I launched it earlier today it took an inordinately long time to load (from 3-5 minutes..)
Since I advised him to enter it in the "System Problems" item in Agora without giving sufficient info on how to do that, I'm reposting a response entered by user cobalt in another Agora item: > (89) #11/11: CoBALT eins261 (cobalt) Mon, Jan 24, 2005 (02:51) > Please pardon my off-topic question and I do not know how to post a new BBS > statement. I hope someone will assist me. My problem is simple: prior to > the GREX move to PROVIDE.NET I was able to "sign on" to Google. > > Now I cannot. LYNX won't even access the url, and LINKS says "no route > to host" when I attempt to access the SIGN IN link. > > I am not aware of any imposition of restrictions regarding sign on to > Google. > > Please someone tell me what is wrong. > Thank you.
grex's move is spectacular - thankxx to gelinas & others
The problem with google is probably related to the web proxy, which is still on the DSL link.
Logging in (via ssh), I saw:
mesg: Unable to find your tty (ttyp1) in utmp file
There is a mesg command in my .profile, but I don't understand this error.
I think I've seen it once or twice since the change to new Grex, but not
consistently. (But I might be wrong about either part of that.)
mesg has to find our entries in utmp in order to properly update wrttmp. Jan discovered that OpenBSD apparenlty writes data to utmp later than it probably should, so he added a delay and retry to mesg. Either the delay is not sufficiently long, or something else is going wrong.
Others have said it already, but thanks for the move, Joe. Backtalk's running faster than I've ever seen it.
This response has been erased.
I have not received any mail from the outside world since some time yesterday. I can send myself mail from grex. Pine works. A message sent to staff from my non-grex address bounced back at 4:37 am this morning after trying for four hours. Has anyone else received outside mail today?
I have confirmed Sindi's problem with mail by doing my own test. About an hour ago I sent a message to Grex from kpunet.net. Checking the outgoing mail queue on the kpunet server I can see that my mailer's connection to grex has been timed out and the message is deferred: DNS from the remote host: > # nslookup > > set type=mx > > cyberspace.org > Server: localhost > Address: 127.0.0.1 > > cyberspace.org preference = 10, mail exchanger = grex.cyberspace.org > > > set type=a > > grex.cyberspace.org > > Name: grex.cyberspace.org > Address: 216.86.77.194 Attempt to connect to Grex's mail daemon: > # telnet grex.cyberspace.org smtp > Trying 216.86.77.194... > telnet: Unable to connect to remote host: Connection timed out
tcptraceroute to Grex on port 25 (the SMTP port) ends at ypsi-sfld.provide.net (216.86.64.2) tcptraceroute to Grex on port 80 shows that ypsi-sfld.provide.net is the last hop before Grex. Looks like Grex's firewall is blocking port 25.
The following happened when telnetting and trying to connect as usual via
cyberspace.org The problem did not happen when immediately afterward I
connected via grex.org :
Connected to cyberspace.org
Grex central timekeeping. At the beep, the time is
2:25PM on Monday, 24 January 2005
New to grex? Type help at the login prompt
(ttyt6) grex login: albaugh
albaugh's Password:
This is NOT grex. Please use the domain name, not the IP address.
Dial-in Users: It will take a little bit longer to get the terminal
server properly configured. We apologise for the inconvenience.
- Grex Staff.
Connection closed by foreign host.
I just verified that the problem is repeatable - it wasn't a one-time glitch.
What the hell happened? I came back to bbs, and it says 91 new items! What keeps destroying my agora participation file?!!!
re #525: I haven't had any problems with my Agora participation file. That proves nothing, of course, but it does suggest that people who are having problems should probably at the very least add the name of the program they're using to read Agora to help staff debug. I'm using "bbs" (whatever that's a front-end to these days) but others may be using backtalk, fronttalk, or possibly other programs to munge their conference participation files.
I'm using "bbs' and my .agora??.cf sems to have
gone away amd, more amazingly, come back. Twice.
Re resp:523: Looks like the name server you're using is still caching Grex's old IP address. The problem may resolve itself given a day or so.
I always use bbs, from a shell session, unless backtalk (web) must be used due to shell inavailability. Under nextgrex, what happens to your (a) conference participation file if you get idle logged out? In old grex, that was never a problem (you would have to re-read the things before the logout happened, but you would lose your entire participation record).
I'm using bbs right now, but every time i try using the web interface, I get the same problem as albaugh.
http://www.thesameproblemasalbaughforjesus.org/
Bbs, earlier today I also lost my agora.cf file and did a fixseen.
I think I fixed the terminal server to connect to the new machine last night, so dial-in should be working again. I also think I fixed the web proxy to accept requests from the new machine. No, you aren't seeing double. I repeated the report here hoping to keep any counter-examples to this, the "Systems Problems" item, and out of the "Systems Announcements" item.
After a several-day-long mail drought I've received several messages (Spam, of course..) Was it the reboot or was something else changed network-wise?
I don't know; I had to start mail by hand after the reboot because it's not starting automatically.
a) cyberspace.org still doesn't come here - only grex.org b) insult to injury - my grex mailbox got filled up, but every time I try to run mail to delete stuff, grex tells me /tmp quota exceeded, so I'm hosed. I sent an e-mail to staff for ideas. Anyone else have any?
paladin: no i'll be put in jail for beating someone to death with a keyboard ---- romantic joining from channel rightnow (Jan 25 16:47) hayz3141: what! paladin is a violent man! scholar: stop carrying on with a married man, pete! scholar: what! karenz: that would be a good bludgeoning instrument scholar: why are you so violent, paldin?! hayz3141: help! help scholar: that won't help you get out of jail, you know! scholar: karen! hayz3141: paladin is a violent pyscho with a keyboard! help! scholar: stop dishonouring your husband by cuckholding him! scholar: help! scholar: i'm going to post that in the system problems item, hayz!
re #536: depending on how much space you need, and whether you can find another world-writable temporary directory, possibly you could bypass the /tmp quota problem by setting the TMPDIR environment variable to something else. I haven't checked how the filesystem is laid out on Grex to see whether this would work..
This response has been erased.
I suppose I could have gathered that from the startup message, but thanks..
Re 536: Can you tell which name server(s) you are using? All that I can check have the right address for cyberspace.org.
It looks like the changes to the terminal server didn't 'take.' I'll power-cycle it the next time I'm in the Pumpkin, which will probably be tomorrow morning.
I can now forward mail without having to make it into an attachment - thanks to whoever fixed this. I exceeded my /tmp quota one time by trying to save a large file to it. If you tried to use /tmp recently how can you figure out which files in there you still own so as to delete them? I. e., list all files that you own in /tmp. Surely there is some command for this. How often is /tmp automatically emptied?
w3m www.google.com -- opening socket............... lynx google - tries to access the IP number and nothing happens links - g - www.google.com - works instantly
The command you want, Sindi, is
find /tmp -user keesan -ls
It looks like I'll also have to kill and restart the web proxy.
Why is links working if lynx is not?
"links - g - www.google.com - works instantly" Yes, but when you attempt to access the sign in link it fails with error "no route to host" "lynx google - tries to access the IP number and nothing happens" --the system seems to be attempting to access Grex's OLD ip address. Perhaps this is a DNS proxy issue with a stale cache? Thanks for attention to this matter.
Re #545, I copied a small file to /tmp, ls -l showed I had read-write-execute permissions but others had just read-execute permissions, and then find /tmp -user keesan -ls listed a bunch of files for all but one of which I was denied permission so I deleted the one that I had permission for (which I had copied). This is the answer to how to empty your overflowing mailbox.
No, it's not trying to reach grex's old IP address. The old address was 216.93.104.34; the address lynx is trying to reach is 216.93.104.35, which is the web proxy. The web proxy is still on the DSL link. I don't know if the problem you are seeing is caused by the DSL link being flakey or by the web proxy not being configured for the new subnet. I added the new subnet to the web proxy's configuration last night, but the "kill -HUP" may not have been enough to cause the proxy to reread its configuration file.
I don't see how the find command is "the answer to how to empty your
overflowing mailbox." It's just the way to find the files in /tmp that you
own. The error message
find: /tmp/lynx-mLGAA21624: Permission denied
means that we don't have sufficient access to the directory
/tmp/lynx-mLGAA21624 to see the files within it.
re 536 What! How could you e-mail staff if you could not run mail ?
-bash-2.05b$ w 12:00AM up 11:19, 20 users, load averages: 0.65, 0.62, 0.59 USER TTY FROM LOGIN@ IDLE WHAT cross p0 fubar.cshack.net Tue05PM 2:22 -tcsh russ p1 sf31-d20.dynamic Tue11PM 3 -csh qw2 p2 e81-197-68-102.e Tue11PM 1 menu cobalt p3 218.13.94.233 Tue10PM 10 lynx contaxes p4 174-227.lib.umic Tue10PM 2 mail rasputin p5 dialup-4.247.140 12:00AM 0 lynx asddsaasddsaasddsa p7 - 12:00AM 0 w ryan p9 67.39.176.100 Tue11PM 0 all your base are belong to us ketsara pa 210.246.163.253 Tue11PM 16 -tcsh naftee pc 55sub16.edunet.u Tue11PM 0 -bash charcat pd host145.64-31-10 Tue11PM 0 /usr/local/bin/party_ pez pf ip68-228-34-112. Tue11PM 0 /usr/local/bin/party_ gelinas q0 216.86.65.248 Tue09PM 0 bbs richard q3 pool-70-19-93-15 Tue10PM 0 /usr/local/bin/party_ janc q4 h-69-3-69-232.sf Tue04PM 57 -csh jmhut q6 157.95.31.61 Tue11PM 15 (pine) gelinas qa 216.86.65.248 Tue10PM 2 lynx http://google.com srw qc pcp05308365pcs.w Tue10PM 1:33 -csh erikb qf a80-127-224-218. Tue10PM 16 mutt rcurl r1 pcp09229896pcs.s Tue11PM 0 pico /a/r/c/rcurl/.cfdir/cf.buffer Whoa.
whoa!
Whoa!
1. Yes, of course, this was normal prior to the move 2. I have checked again, the problem still exists, as explicated on the BBS System Problems Board, Item #544 and response. SUMMARY: links : can access google but forbids sign on, "no route to host" lynx : attempts to access old proxy then timeout. Thanks you for attention to this.
You can *send* mail to others without any bearing on your inbox!
Oh, I got the impression that you could not run mail.
When you run mail without a recipient list (or other command line parameters) it invokes it in "read the inbox" mode. Setting TMPDIR to my home directory also causes quota exceeded on mail startup. Who is going to fix this? (!) It is intolerable that I can't fix my situation due to system setup "flaws".
(In reference to Item #555 ): The Google link rejected by LINKS is an https link. I hope that https links are not restricted on this new server. Thank you for your attention to this.
re 558 So it can't "read the inbox" if your quota has exceeded ?
sorry: clarification of Item #559- LINKS loads www.google.com without error (I believe another user has mentioned this) and it is the sign-on that fails with error "no route to host" but the sign-on link being accessed by LINKS is an HTTPS link. So seeing that LINKS can access http://www.google.com there seems to be no problem there whatsoever, but actually the problem PERHAPS is the LINKS rejection of an HTTPS page. Please respond at your convenience. Thank you for attention to this matter.
Re 554 - proof that 8 character logins are enough for normal users. And that longer ones don't display properly in certain programs. Longer ones are also a nuisance to add to the twit filter. I thought that the mail problem has to do with /tmp being over quota and removing files owned by the user in /tmp would fix it. I just had a 20 sec freeze. Shades of old grex.
There was some kind of routing problem at provide.net about an hour ago. It may come back; I don't know. Ah. Yes, clearing out your files from /tmp would give you more room for mail's temporary files.
I'm seeing: & a p o s ; instead of a single quote character, using Backtalk this morning, only on Grex. (I don't get the same thing on M-Net.)
Oops, apparently posted in the wrong item- but yeah, having the same problem as jep. (Maybe it's a Perry thing. ;) )
No, I am having the same problem
I just got a legitimate email in 8859-9 (Turkish - someone sent me a tel and I guessed he was Turkish and emailed back what I thought meant 'I don't speak Turkish' but maybe I left out the 'not') so please don't set the spam filter to rule out subject lines in other charsets. I got a 'From' in Chinese recently and a message body in Russian. bbs is working fine. Procmail just threw out 6 spams in 2 hours and if this is typical I must be getting about 70/day. Any progress towards choosing a spam filter or at least porting Marcus's minimal one?
5 more spams since 10:30 (in one hour).
cyberspace.org now yields grex
&%$&%$&* grex froze while I was entering the following: I have *no* files on /tmp!!! The problem is that my filled mailbox, which apparently mail wants to replicate somewhere while it's running, is larger than my quota on /tmp - and also larger than my quota in my home directory. Cannot someone on staff simply pick a larger, more suitable quota on /tmp to address this situation, and deploy it? A larger personal quota for me should also allow me to clean out my mailbox, but it is not addressing the underlying problem of incorrect settings for mailbox versus /tmp quota. Note: I'm not faulting anyone. But now that we know there's a problem, it should be corrected. P.S. Curiously, and different from old grex, since my .forward file has mail left in my grex mailbox as well as being forwarded off-grex, I continue to receive the forwarded mail at my non-grex account. I assume that the messages are *not* stored in my filled mailbox, and I wonder if senders are getting "mailbox full" bounces...
My arrow keys on 101 or 84-key keyboard do not work to move around within the text in fronttalk with vt100 or linux, sshed with linux to grex. [D beep etc. I will try dialing directly with kermit when that is possible. vt320.
It's great that Grex is up at Provide.Net and that Grex via the Web is working fine. Unfortunately, the dialins are still hosed. Here is what I just got when I tried to log in: Welcome to Grex! It may take a few seconds to connect. Grex is the Midnight Snack of Champs New to grex? Type help at the login prompt groupie.cyberspace.org login: dpc dpc's Password: This is NOT grex. Please use the domain name, not the IP address. Dial-in Users End of message. I have gotten this several times over the past week. I have no idea what "groupie.cyberspace.org" is. Also, the "Dial-in Users" line was truncated, and I was promptly logged off. I was dialing 761-3000, so I am sure that I did in fact reach Grex. What's happening here?
Re 572: kingjon reports the same thing, as of (I think) last night.
What's happening is that "groupie" is now using the IP address that used to belong to "grex." I *tried* to tell the terminal server to use the new IP address for grex, but it doesn't seem to have taken. I'll try again. The machine known as "groupie" was known as "grex" until a month ago, BTW. I put it back on its old address because we need a name server at that address until I get the domain name registry updated. I'll post the full text of its "nologin" file as soon as I can log on to groupie.
havign a problem loging in just now vie telnet.
Now experiencing dns name resolution timeouts for lynx and links. Thank you for your attention to this.
Similar delays (presumably DNS timeouts) earlier today when launching pine. No delay just now.
The full text of groupie's "nologin" file is:
This is NOT grex. Please use the domain name, not the IP address.
Dial-in Users: It will take a little bit longer to get the terminal
server properly configured. We apologise for the inconvenience.
- Grex Staff.
DNS timeouts for LINKS and LYNX I am posting this only because I see DNS timeouts noted for pine, but there is no further information. Thank you for your attention to this.
If someone has time, could they please test LINKS access to www.google.com and attempt to "sign on" (even if you do not have an account). The link is https and the failure reports "no route to host." LYNX still attempts to access old server proxy? Making HTTP connection to 216.93.104.35 ? Any information on this ? Thanks (everyone).
BTW, for better or worse, my filled up mailbox, which I can't access *still* due to the /tmp quota, keeps increasing in size. So incoming messages continued to be added to it, even though admonition messages seem to imply that they won't be.
Could someone please kill the pgreen and erica accounts? I type bbs and get 'I don't understand "pgreen" and (no conf). Fronttalk works but I still get this message and one about erica, 2x each.
I have now confirmed that LINKS cannot access any https links; can this feature be activated/toggled ? Hope someone can help when there is time.
When I do r n and there are new responses, using fronttalk, it automatically scrolls to the last page so I lose the first part unless I scroll back up. I don't recall this happening before - vandalism? NOt that I have been using ft until now.
That's what happens when you use less as your pager and you run a twit filter.
The problem is that the "twit filter" line in your .cfonce has been broken into three lines. Rejoin it into one and your problem whould go away. Killing other people's accoutns will do nothing to make the error you are seing go away.
Thanks, that fixed it. I would not need to edit .cfonce with pico -w if the twits did not have multiple personalities. How did you figure this out? Is there some way to read the last response in an item if it is not new, other than r, q, then the number of the last response?
Thanks to Joe-of-staff for addressing my mailbox problem.
Thanks, Joe!
The dial-ins are still giving me "This is NOT Grex...".
I used "less" on your .cfonce, Sindi. In picospan, I use "r X new" and then "-1". In FrontTalk, I use "r X nor" and then "-1".
Thanks, joe!
Is anyone else unable to telnet or ssh to grex today? (I cannot reach sdf either).
I reached grex eventually but cannot FTP upload to grex and the telnet lag is awful. top does not reveal any problems.
I'm here via ssh.
The system seems REALLY slow, but I don't know what's causing it. Could it be someone consuimg a lot of resources?
Speed problem is fixed, thanks.
I cannot do an ftp download. All I see is S. A few days ago I could not do an ftp upload. I am a member with ftp privileges.
Yes, a vandal wass running a fork-bomb. 'Twas killed at or about 08:56. (Which the observant will note is the time Sindi reported her dial-in attempt failing. Correlation is not causation, but . . . ;)
Sindi also emailed staff shortly after the dial-in failed and the telnet succeeded but was 89 load average. Should I bother reporting vandals or does staff notice them anyway? I tried again to dial in just now and got Welcome to grex. It may take a few seconds to connect...... And after a minute or so I gave up. ssh connected instantly.
You need more patience. The modems are still on the far side of the DSL.
This response is made by sshing from my machine to oldGrex and then from oldGrex to Grex. So it passes over the old, flakey DSL twice. It is, as should be expected, quite slow. A good way to learn patience. :)
As if we have not had enough practice with patience in the past couple of months! Telnet is not bad now.
Are there plans to move the dialin lines to the colo site, or will they be dropped when the lease expires on the Pumpkin?
I'll be faxing the information to transfer the lines to the telephone company later today. I am sorry that this has taken so long. :(
From one who is "spoiled" by SiteScope in the workplace: Is there enough spare grex hardware lying around to construct a "monitor" box whose sole purpose would be to monitor grex and when it detects something irregular (such as no response within a reasonable period) to notify whomever it would be willing to be so notified?
My arrow keys still do not work in fronttalk with my 84-key keyboard (or probably with the 101-key keyboard either). Yes it said Fronttalk (it scrolled by too quickly to read anything else) when I typed ft and I do not have a scroll buffer in linux except in vga and I am using mda.
I see robocop is running now. Or was running. Things just speeded up.
When scribbling a recent response that wasn't very interesting:
Respond, pass, forget, quit? scribble 23
Segmentation fault
grex%
I've got "limit coredumpsize 0" set in my .tcshrc, so no core file
(not that anyone would have time to track it down..) Is there a
problem with the perms on the censored log, perhaps?
This response has been erased.
pine didn't start for me after 2 minutes
This looks like a problem: PID TT STAT TIME COMMAND 25997 r3- I 0:00.61 ./psybnc 25353 r3- I 0:00.63 ./psybnc 21795 r3- I 0:00.58 ./psybnc 5143 r3- I 0:00.55 ./psybnc 27215 r3 Is 0:00.02 -bash (bash) 24250 r3 I+ 0:00.41 pico index2.html
This response has been erased.
Grex ssh worked last night but this morning: remote host failed or refused to allocate a pseudo tty. (A bunch more lines and then a blinking cursor).
Now I could ssh and login but then 'unable to find your tty (ttyr4) in utmp file. Things seem to be working okay anyway.
Lynx is working, ridiculously fast! I keep overshooting. But ftp (for members) is not working with lynx.
I don't find it too surprising that ftp doesn't work within lynx; lynx isn't linked to the necessary wrapper.
When I ssh here 'unable to find your tty in utmp' again but it works. Ftp has not been working in the browsers for a while now and I hope it can be fixed soon so that I will not have to exit lynx to use ftp once I find the directory I am interested in getting a file from - last night it was a modem upgrade file from usr that I could not access on my own slower connection (it took so long that lynx on my computer started to look for www.usr.com.edu). I can ftp between my computer and grex now.
I mention this because I don't know whether something was done since the last
mention (but if it was, it didn't work):
I just got
mesg: Unable to find your tty (ttyq6) in utmp file
when logging in.
Sindi slipped in, with probably the same problem (with mesg).
Just as another data point, I'm still receiving the type of error when
ssh'ing into grex that is described in resp:614 and others, above. Here
is the oddly formatted text that is written to the screen when this happens:
<pasted>
Server refused to allocate pty
Warning: no access to tty (Bad file descriptor).
Thus no job control in
this shell.
stty: mesg: I/O stream 2 improperly redirected
not a terminal
stty: not
a terminal
tset: standard error: Inappropriate ioctl for device
stty: not a terminal
mesg: I/O stream 2 improperly redirected
watch: Could not identify your terminal
</paste>
Note: this error doesn't always happen but when it does it seems
impossible to ssh at all.
I would appreciate any explanation as to why this occurs, still. On the
old Grex server, this would happen a lot and everytime someone asked,
we'd get the pat explanation that "just wait until next Grex; all these
problems will go away."
Thank you for your attention to this matter, volunteers.
whuddup -bash-2.05b$ w wing21 7:44PM up 6 days, 23:58, 16 users, load averages: 0.78, 0.83, 0.80 USER TTY FROM LOGIN@ IDLE WHAT wing21 ft 200.175.43.115.d Sat10PM 0 - wing21 ft 200.175.43.115.d Sat09PM 0 - wing21 ft 200.175.43.115.d Sat10PM 0 -
I thought he mentioned it somewhere above, but perhaps folks missed it: OpenBSD doesn't write information to the utmp file as quickly as it could. So things that look at utmp, like mesg, during login sometimes fail. So Jan added a one-second retry to mesg. Sometimes, even that one second isn't enough. Personally, I've taken to running the mesg command from my .login by hand when I see that message, figuring that by the time I can enter commands by hand utmp has been updated. On oldGrex, the "failure to allocate pty" error was caused by the telnet queue. NewGrex doesn't have the telnet queue, so we didn't expect to have the "failure to allocate pty" error. However, sometimes no ptys are available, with the result reported in #621 above and elsewhere. Right now, we don't know why ptys aren't available. We have tried increasing the number. We've not stopped investigating the problem.
This would be a really low priority item compared to all of the rest, but is there any chance of upping the time limit you are given to authenticate before ssh times out? The old grex used to be quite generous in this regard, now I find that if I start ssh and switch to another window I often run into the limit before switching back.
Any progress on moving over the modems? We know at least one grexer with no other way to reach his mail, and he assumes it is now full of spam and bouncing all his real mail.
Well, let's see . . . Three modems (one a spare) are now at Provide.Net.
One is attached to grex. The other needs a different cable, because there
wasn't enough space for two DB25-DB9 adapters on grex's serial ports.
I borrowed a telephone cord from the Provide.Net people, to connect that modem
to the telco. (Amazingly, it hasn't been called yet. :)
I think we need to turn on the tty for the connected modem. I also think it
is on tty00. Unless I guessed wrong, in which case it is on tty01. I don't
know yet if turning it on will require a reboot. Whether it does or no, I'm
going to turn on both at the same time.
Before my next visit to Provide.Net, I need to get:
two 25-foot telephone cords
A DB9-DB9 serial cable, preferrably male-female.
The last could be replaced by a DB9F-DB25M serial cable.
BTW, a serial-to-USB adapter at Radio Shack was $41.xx, just a little more
than a new, external USB modem.
Don't buy at Radio Shack! Try Best Buy, CompUSA, and the next Gibraltar Trade Center computer show. Or local computer stores.
This response has been erased.
We have a serial to USB adaptor that could be donated to grex if the serial ports cannot be made to work, and probably a few long phone cords. We have at least two spare 9F-25M modem cables and Kiwanis probably has 9-9s (I am using our only one with a newer modem that needs it). Give us a call (see Deigert in the phone book) if you want these tomorrow.
I am dialed into grex!!!!! No telnet lag. I had last tried about 7 pm and it just rang and rang. We have one 9-9/25 modem cable that grex can at least borrow until we find another at Kiwanis. Jim says he can make whatever length phone cords are needed - he has a tool and the ends to put on with it. We have some 9-pin female to 25-pin male (computer to modem) that grex is welcome to keep as we know we can get more of these at Kiwanis. Joe, what time are you going to provide.net tomorrow? Yippe!!!!!
I can even ssh from grex to somewhere else after dialing in.
I don't know that I will go tomorrow, much as I'd like to. When I went over this evening, I got lucky: some other folks had just arrived and were calling to be let in. Had they not been there, I'd have not been able to get in. While access is available 8am-10pm, it's not automatic. I'll poke around my own basemenet first. I know I have some long telephone cords, and I have the means to make more, too.
Why am I logged in from -? I think it used to be 'groovy'. Does this reflect the change from a computer to a serial port?
632 slipped in. Joe, let us know about modem cables. I have a long and a short ready for you (F/M).
Yes, "-" reflects the absence of the terminal server. I'll let you know, one or the other, later tonight, Sindi.
I'd like to make sure that someone who is NOT a member can't ssh out, too. (I expect that they can't.)
getting an error message that says "write failed, file system is full" uh oh
however I got that error in party, if the file system really is full, why can I still post on agora?
ditto #637
ditto #charcat's mom
Re #638: When it says "file system", think "disk partition". Grex's disks are multi-partitioned. Party and bbs use different partitions for writing data. None of the partitions are full now, so someone must've taken care of the problem.
I just tried to dial in again and got a long wait and a NO CARRIER.
With only one modem on a two-line hunt group, unanswered calls aren't too surprising. We can also expect busy signals the first week of dial-in access: A lot of demand has been dammed up; the flood won't be pleasant. ;/ But neither will (should) it be long. :)
I have not been able to dial grex since 10 pm last night. I just checked, nobody else is dialed in. It dials, makes some noises, then dials again, then disconnects. Has anyone else dialed in today? I tried at least 5 times.
Yeah, looks liek there's a problem. I'll check on it tomorrow, after I visit Kiwanis.
I was able to dial in this morning, but melanief was not, and in the afternoon neither of us could dial in. Trying now - the phone seems to be answered, but there is no connection beep.
No luck this morning for me (Monday). The Kiwanis visit is because a grexer who runs the computer/electronic dept. there has some ideas about how better to set up the modems with twisted-pair phone cable instead of 25' phone cords to eliminate interference and resulting problems. He is totally dependent upon grex dialins for internet access so has a strong interest in getting this working soon. Joe, thanks again for all your time.
Twisted pair *is* better than flat cord, for avoiding interference. CAT3 is often used for phone wiring.
I keep getting this error: login as: naftee naftee@grex.org's password: Server refused to allocate pty
I don't know why ptys aren't being allocated; we only had 26 users on. I went with flat cord for now; 'twas available.
re #649 DITTO
Re. 650: it's because people are breeding too quickly. :( (also remember that people can use up ptys merely by telneting in. they don't actually have to enter their username and password.)
Thanks, I am dialed in now. Why did you need to turn off grex to plug in the modems?
I still can't dial in. The modem tries to connect - I get the beep and the answering beep, then some transmission noise, and then a higher beep which never stops. The connection is never completed. Joe, are these two of the same modems that we were using before? Did you change any settings?
I had that problem for a while at work. Some modems just couldn't dial in to our modem -- they'd start the handshake and never complete it, with the same sort of high-pitched beep. I switched to a more expensive modem, one meant for dialin use, and that solved it. (Specifically, I switched from a $50 US Robotics Sportster to a $300 US Robotics Courier. Sportsters are strictly consumer-grade modems, and not very good ones at that, as it turns out.)
Re: #653. I assume because the modems in question are not plugged into serial ports, but into the expansion ports at the back of the computer, which require you to take the case off to get at, and turn the power off to avoid shorting the motherboard. If this is the case, then frankly, I'm amazed that any recent PCI modems have been found that aren't those accursed WinModem things, which, as the name suggests, are crippled to only work with That Other OS. This is done by moving most of the (already limited) intelligence of modems into proprietary modem driver software written for Windows, and distributed without source code. It serves Microsoft's purposes because it makes them hard (but not impossible! 8D) to use with Linux, et al., and serves the purposes of the manufacturers because it makes them cheaper to produce whilst slightly raising profit margins. Thankyou Microsoft, thankyou very bloody much. I don't think. (As an aside, reading, in Knoppix Hacks, about the limitations of the Windows Recovery CD made me think of a brilliant strategy for punishing Microsoft for their illegal behaviour towards other OS and application vendors: They should simply be forced to use their own OS.)
Re resp:656: I'm pretty sure Grex is using external modems, unless something has changed.
Interesting. Perhaps then turning Grex off was just to be on the safe side. Only Joe can enlighten us, really.
http://www.safesideforjesus.org/
re #659...we seem to have a problem with certain users or user posting fake url's. A user posting fake url's is purposely posting bad information on the system. Which in turn cause people who come here to see these links and hit on them. If grex is full of posts with bad url links, it makes grex itself look bad. IMO staff does not need to be condoning such activity.
Primarily, I turned grex off because I wanted to plug it into the powerstrip I have the modems plugged into. When I went in on Saturday, I plugged the powerstrip in to the UPS. It just made sense to me to have everything plugged into the powerstrip. These are the GVC external modems we were using at the Pumpkin.
if this is the case, we must first as ourselves, why are people hitting on links instead of on buns? a neoclassical formulation would b e: because they're women! but this was busted up by the implicationf of implied homosexuality in the work of anciebnt greek scholars combined with the people who supported "greek life". so, the next formulation was: because they'er women and abnormal men! this held true until the famous french midget jaques derrida came along and DECONSTRUCTED the idea of gender, positing that women were the euqal and opposite reaction (i.e. social newtonianism) and that this allowed us to bust up the gender binary, and come up with he current forulation: some PEOPLE like hitting on lin ks, some like hitting on buns, some don't like hitting on either, and some like hitting on both. needless to say, richard shoulod be ashamed of his discriminatory/sexist/genderbinaryist views of hitting on links.
What?!
deliberately posting a bad link on grex on purpose should be seen as just as bad as deliberately posting bad code on grex. Or deliberately posting any sort of information known beforehand to be bad on grex. It is corrupting the quality of the bbs. Grex should pride itself on the quality of its posts, and that means staff taking action when it sees users deliberately and purposely posting bad information repeatedly.
Re #661 - something must have changed, because I could connect to the Pumpkin but I can't connect to Grex in its new home. Nothing's changed on my end except the phone number I'm dialing. It does sound like there's a little bit of line noise, but I wouldn't think that'd be enough to cause a problem. BTW Joe - do these lines hunt? I.e., if I dial 0512, and it's busy, will I be forwarded to 0513? Also, the !phones command needs to be updated.
The 'phone company tells me they hunt. The phones script is updated.
http://www.richardscrusade.org/
Re #664: Richard, don't feed the troll!
I tried just now and could not connect via dialing. I might have been using a different computer earlier when I connected. I have mostly USR Sportsters flashed from x2 to v90. This one is a 28.8K.
When I get back I could try to dig up my old Newton-based modem configuration.
re resp:664: Grex should pride itself on the quality of it's posts? Do you read most of the responses? The URL postings are below average in creativity and information content but they are hardly the bottom of the barrel here. Just filter the author, as I do. Then you don't have to worry about them any more.
you filter?
He's a backtalk user.
filters are for pussies.
picospan still segvs when I scribble a response.
Re. 674: AHAH< AND FOR CUNTS
HAHAHA>LOL
This response has been erased.
I can't run Pine because (I think) Grex is unable to find my terminal in the utmp file. What does that mean?
Re resp:664: Honestly, I think staff has better things to do than censoring people's posts.
Caveat re resp:678 - Fronttalk is beta, so expect some rough edges if you use it. However, Jan's been putting a lot of work into it recently and it's getting to be in good shape. There's an item in the Garage conference to discuss Fronttalk bugs and features (item:garage,5).
I can't dial or telnet to grex now.
So far, only the first modem has been used.
Telnet and ssh both work for me. I want to get some cat3 telephone cable. I have cat5, but I don't think that will work for telephone. Of course, I could put two jacks on each end of the cat5, instead of one, and use one cable for both modems. :)
:)
CAT5 will work fine. It's just more expensive. CAT3 and CAT5 both have four twisted pairs (eight conductors), incidentally -- CAT5 just allows for more bandwidth. In either case it's just a matter of crimping the right connectors on the ends (or attaching the right jacks, as the case may be). For best noise rejection, don't "split pairs" -- use two wires from the cable that are twisted around each other for the phone line. Only the two center conductors of a phone connector actually do anything, on a single-line phone, and in fact you'll find a lot of commercial phone cords only have two wires. So, for a single-line phone, the connector would look like this: --AA-- (Where - is empty, and A are wires from one pair of the cable.) For a two-line phone, it would look like this: -BAAB- (Where A is from one pair, and B is from another.) Incidentally, the two wires that are used on a phone connector correspond with the two unused positions on a 10baseT ethernet connector. I suspect this is to help protect ethernet equipment from damage when some clueless person accidentally plugs a phone line into it.
Hmm. I just found this website that probably explains it in a less confusing way than I just did: http://www.ling.upenn.edu/~kurisuto/phone_wiring.html
Jim and Kiwanis probably have this cable.
Kiwanis did not.
ummmmmmmmmmmmmmmmm, ethernet wiring, iirc, is *not* wired correctly for telephone use. oh, wait, that is incorrect since polarity no longer matters to the modern phone newtwork. this applies to 4-conductor cable. gull's #686 is correct for phones, single-pair and two-pair.
It's true that ethernet wiring isn't quite the same as phone wiring, but the cable itself is fine -- the difference is in how the ends are attached. You're going to have to change the ends anyway, unless your phone stuff has RJ-45 jacks on it. (Which isn't unheard of, mind you, but it's rare.) There are still some things out there that care about polarity, though you're unlikely to run into them. I found out the hard way this week that polarity matters a lot to DID lines. Ended up making a custom cable for a new fax modem because the jack at work was wired backwards, to compensate for some old equipment that was also wired backwards. DID lines are their own special nastiness, though; take everything about how phone signalling normally works and stand it on its head, and you have a good idea how a DID line is set up.
having no luck at dialing in on the new number I listened in while dialing and heard the recording "you must dial a 1 and the area code" does this meen the call is now long distance for us western washtenaw users? If so I'm upset. It basicaly makes my other computer and my mother's useless. Why couldn't we keep and ann arbor phone?
Leslie/arabella's .login file doesn't run when she logs in. This broke with the migration to the new machine. The commands in the file all seem to run fine if she types "source .login" I can't figure out what's different between her setup and mine; my .login has always run as expected.
How many hours a month do you normally dial in to grex from those two computers? Access-4-free offers ten free hours/month internet access from which you could telnet ($5 signup fee). And I ran across one long distance phone company which was under 3 cents/minute within MI. I had no idea Ypsi was a long distance call from Chelsea.
ypsi is long distance for Chelsea, Dexter, Manchester I'm not sure about Saline.
Is ypsi at least local for some towns that Ann Arbor is not, east of ypsi? Steve, do you have a dsl line at your usual computer and if so, does this mean you cannot also dial in to the isp from other locations? I suspect another friend in Chelsea is going to drop her grex membership if she has to pay for an ISP to reach her mail here. If the old number is forwarding, is that original number still a local call from Chelsea? That would at least give people 90 days of free calls while they found another way to access mail. I still cannot dial in to grex. Worked right after the modems were reinstalled but not since then.
I looked at some login files:
drwxr-xr-x 30 root wheel 512 Dec 28 01:17 /a
drwxr-xr-x 38 root wheel 512 Jul 16 2001 /a/k
drwxr-xr-x 38 root wheel 512 Jul 16 2001 /a/a
drwxr-xr-x 46 root wheel 1024 Feb 8 16:53 /a/k/r
drwxr-xr-x 63 root wheel 1024 Feb 1 14:35 /a/a/r
drwxr-xr-x 10 krj people 9728 Feb 9 00:44 /a/k/r/krj
drwxr-xr-x 3 arabella people 3072 Feb 9 00:36 /a/a/r/arabella
-rw-r--r-- 1 krj people 712 Dec 28 23:19 /a/k/r/krj/.login
-rw-r--r-- 1 arabella people 333 Feb 9 00:16 /a/a/r/arabella/.login
Just wondering if some protection was different somewhere in the
paths. Making the .login executable might be worth the trouble.
I have a very dim memory, of the hash mark "#" as the
first char on the first line of a script, in some cases,
for some shells, also making a difference *on the second line*.
I can't find any corroboration but it's worth a try.
I'm having a problem in that when I log in on occasion, it treats all items in Winter like a brand new item. Is this due to rebooting or what, and how can I fix it? Or more importantly, how much of a bribe to get a staff member to fix it for me?
If there are enough members who need local access to grex, we might be able to justify the expense of having some sort of call forwarding from Ann Arbor.
How are the new phone lines working for dial-in users who've tried 'em? My modem can't seem to connect (endless training tones) and it sounds like the lines have fairly bad static.
I was just able to dial in now. Instant connection once the modem stopped making noises.
Disk quotas are too low on /tmp; I can't even gzip a dump and put it there temporarily to download.
This response has been erased.
resp #696 yes the old number was a local call. I don't know why I'm so upset about this, I can use the internet to connect from my home computer but I know 2 people who can't. How much extra would it cost grex to keep an ann arbor phone or call forwarding?
I don't know, but I can probably find out how much it would cost to set up
permanent forwarding. I'll try to call SBC tomorrow.
One difference between ~krj/.login and ~arbella/.login is:
if ($?prompt) then
. . .
endif
I don't know that that is the problem, though.
I was able to dial in last night, but not this morning. melanief called to tell me she still can't connect. Joe, I think some of the GVC modems were suspect, and we had the best of them in service before the move. Are the two that are connected now from the pool of modems that were in use before, or were they idle?
I could not dial in this morning either. Steve, are you using dialin service to your ISP or broadband? If broadband, don't ISPs also let you dialin as well so you can connect when away from home?
Can't speak for all, but Comcast and WOW do not.
> WARNING: Your mailbox is 87% full. > If it reachs 100%, you will no longer be able to receive mail. It is my experience on nextgrex that this is not true: Due to being sent several large attachments at once and SPAM, I know I've well exceeded the inbox limit (still 1MB?), and yet have been able to receive more mail. That has proven to be very useful, as there was a time I couldn't run mail to clear out stuff, due to the low /tmp quota, yet this didn't start bouncing mail to senders. OTOH, if the inbox limit isn't enforced, that could allow a big pile up of junk sent to defunct accounts etc.
I got 'network not supported' first time I tried to ssh here, then the second time it worked. Is this something to do with where I am ssh'ing from? Still cannot dial in.
The modems are from the active pool. Last I looked, the mail quota was 5MB.
Dial-in/modems/problems: What i'm hearing definitely sounds like continuous, bad line noise. I've no problems (& don't hear line noise) dialing other places. Dunno...i'd say it must be in the trunk line (vs. "last mile") from what folks are saying, but i'd also thought that noisy trunk lines were a thing of the past.
Could the noise be from using 25' phone cords? I still can't dial in either. It has worked twice in the past few days.
Possible, especially if the cord is bundled up.
> Last I looked, the mail quota was 5MB. At the time I was given the "100% full" message yesterday, I'm pretty sure I had nowhere near 5MB worth of mail in the inbox.
You probably know this, but
ls -l /var/mail/<loginid>
will show the size of <loginid>'s inbox.
Jim offers to put together a regular phone cable that will plug into the phone jack at one end and have two jacks for SHORT phone lines at the other end (or perhaps the cable could plug directly into two modems? - how far apart are they?). But first he has to recover from the flu as he cannot do this too well while in bed.
weird here ....
grex% w
3:52PM up 3 days, 9:40, 36 users, load averages: 1.08, 0.86, 0.84
USER TTY FROM LOGIN@ IDLE WHAT
volt p0 tran.adsl.mustan 2:59PM 5 -tcsh
politie p1 gn-lwb-5c43.adsl 3:49PM 0 (pine)
cross p3 sl-aafes1-40-0.s 2:42PM 4 -tcsh
tsty p4 pcp03303701pcs.y 3:45PM 0 w
jvmv p5 200-203-131-127. 3:16PM 0 gate /c/j/v/jvmv/cf.buffer
smoke p7 ACD89C61.ipt.aol 3:51PM 1 -bash
jn p8 port1364.ds1-by. 3:48PM 1 -bash
gudri p9 - 3:30PM 0 /usr/local/bin/bbs
vallica pa 81.12.221.23 3:23PM 0 -bash
test19 pb pcp745288pcs.res 3:10PM 0 vi testing.txt
unix22 pc dup-200-65-13-13 3:36PM 8 vi copiar
cross pd sl-aafes1-40-0.s 3:46PM 0 screen -r
cross pe fubar.cshack.net Wed09AM 0 vi minunit.h
marcvh pf c-24-18-218-42.c 3:27PM 15 -bash
frank60 q0 130.182.168.175 2:45PM 0 -bash
aym q1 130.182.168.209 2:46PM 0 /bin/sh /usr/local/bin/help
radi q2 83.238.4.98 3:46PM 0 lynx
steve q4 pcp04149896pcs.s 12:01AM 0 -sh
q76 q5 user6.47.udn.pl 3:51PM 0 -bash
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
hackerpt ft bl6-11-28.dsl.te Fri05PM 0 -
grex%
plus ...............
grex% f -m hackerpt
Login: hackerpt Name: hackerpt
Directory: /a/h/a/hackerpt Shell: /bin/bash
On since Fri Feb 11 17:04 (EST) on ftp13449 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:08 (EST) on ftp10539 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:08 (EST) on ftp1084 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp3456 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp21571 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp21872 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp15683 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp17847 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp4871 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp26602 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp440 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp24252 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp25520 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp19024 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp14084 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:09 (EST) on ftp19333 (messages off)
from bl6-11-28.dsl.telepac.pt
On since Fri Feb 11 17:14 (EST) on ftp16240 (messages off)
from bl6-11-28.dsl.telepac.pt
No Mail.
No Plan.
seems to be stuck somehow ...
portugal has arrivedd
This response has been erased.
What a crock.
This response has been erased.
An update on the modems: I didn't get a chance to call SBC on Thursday or Friday. I did find 140 feet of cat5 cable in my basement. I _couldn't_ find my crimper, so I plan to buy a new one tomorrow, make a couple of cat5 telephone cords of appropriate length, and then get to Provide.Net to install them. The last step may not happen until Monday, though.
whip me up a batch of pancakes, cross.
Jim has a crimper you can borrow, also some of the round grey cable normally used to run phone lines, whatever it happens to be. I think his idea was to make one line with something at the end to plug two phone cords into. Or if the modems are fairly close, perhaps he can make something that plugs directly into the modems (soldered instead of plugged, better contact).
Splitters, which can be added to the end of an ordinary phone cord, are "dirt cheap". It isn't worth the time to make custom cords.
It's telling me that my mailbox is 98% full, when I have just over one megabyte of mail in it. Furthermore, for the last several days it has refused to let me delete anything (if I was able to read it at all) because I was over quota in /temp. Yesterday and today, also getting the incredible line noise that won't let me dial in.
(To add to #726, as one who watched and tried to help:
(1) On dialin, the modem connects, but it gets what appears to be continuous,
severe line noise, never a visible dialin prompt of any kind. As with
others who have reported this, the only difference on our end is the
number being dialed.
(2) Grace was able to use elm to read her mail, but when she exited, if she'd
made any changes (deleted any messages), it told her that her limit was
exceeded on /tmp. To add insult to injury, or maybe more injury, it left
the original copy of her mailbox in /tmp, blocking future attempts to even
run elm. That limit needs to be at least doubled, I'm afraid.
(3) Her problems with elm were greatly exacerbated by the amount of spam Grex
is letting in. The mailbox keeps getting bigger & bigger, but the limit
on /tmp won't let her delete any of it.
Is it possible to hand-edit the mailbox? Or maybe to download it first and then read it and delete it at grex? Please try my .forward and .procmailrc as a filter - just the first 20 filters (and the first and last three lines or so) of .procmailrc are needed and change keesan to davel or gracel.). My long filter catches 95% of spam but the first 20 parts should do about 90%. A quarter or so of my spam is sent to one of about ten other grex addresses in a batch mailing, most of the rest has java or images or a few key words like medication.
yahoo email is free.
I get lots of spam with header 'To:': benny, amit, ebsy, eh, kleptic, krex, makji and hal9000, none of whom exist at grex. Could someone set up a systemwide procmail filter on these? (I also get spam addressed to at least 6 currently active grexers). Putting Subject:...prescripiton and Subject:...taladafil and X-Message-Info, img.src, and java would catch at least half of the rest of the spam. Does anyone WANT to get emails with these at grex? In my case at least nobody but spammers writes me about medications or meds, but maybe some people discuss those with their friends, along with cumshots and mortgage refinancing.
Any about cumshot refinancing?
Just a heads-up: I'm not going to get to Provide.Net today. :(
I got there today. My Cat5 cable contained two eight-conductor cables: one blue and one yellow. I put an RJ-11 connector at each end of each of them (leaving four wires unused in each cable) and put that in place of the separate telephone cords we had been using. So that part of the problem should be fixed. I also plugged a telepone directly into the test jacks: one (0512) was dead, and the other (0513) was so noisy that the telephone wouldn't dial through. I called SBC from another line in the facility to report the problem. They promised by repair by tomorrow evening. That should take care of the second part of the problem. Since only one line was working, 0513, and only tty00 has been used, I changed the order of the modems so that 0512 would be on tty00 and 0513 would be on tty01. Now we wait for the lines to be fixed to see if I fixed that part of the problem, as well.
Thanks a lot Joe. I hope SBC will also refund us the charges for the period when the lines were not usable. I look forward to dialing in again. Maybe then I will be able to print out my email again? When I try (with % or with Y after enabling that), the file just scrolls across the screen but does not print to 'attache-to-ansi' - or maybe I simply don't know how to print in linux, though I did insmod three modules - parport, parport_pc and lp - do I also need lpr? Normally I just dial in with DOS and kermit.
My experience with SBC is that you should expect it to take several tries to get a reliable line. Hopefully they'll do a better job for you, though.
Ok, once again when I logged on, it showed all the items in the winter agora as being brand new. What's up with that?
I've been out of town, and unable to telnet in, for several days. Just now, when I logged in, I was told (quite understandably) that my mailbox was full, and that I would not be able to receive any more mail until I removed some of it. So - wait for it - I tried to read my mail and was told that /tmp was full, and my quota was exceeded! There is no way I can read any mail. Would someone with a clue *puh-leez* increase the disk quotas for mail so that I (and others) can read it? Thanx!
Thanks Joe, for dealing with the modems.
Thanks Mark!
Grex won't let me save changes to the HTML version of a conf login.html file
dpc is suffering the same thing I went through, until my /tmp quota was increased. Is this how it's going to be - one by one, affected users have to ask for help to solve something they could have solved for themselves?
Do you have spam filters?
A short spam filter that catches about 75% of my spam -
delete lines starting in #, which I put in for information only.
.forward (in home directory):
|/usr/local/bin/procmail
.procmailrc (in home directory):
MAIL=/var/mail/yourlogin ##change this!!!
LOGFILE=$HOME/mail/from
VERBOSE=on
##The above sets your mail directory and produces a logfile
##with a list of what mail went where and why -- ~/mail/from
##which you can read with pine, L, as 'from'. You can see if
##the filter caught a real mail and why, then either put the
##mailer on your white list or remove the filter that caught it.
## This first filter forwards mails over 100K to my other address.
:0:
* >100000
! keesan@myotheraddress.org
#The next two are 'white list', mail to be let through:
#Change from carole to the names of people who write you
#if your filter catches their mail.
:0:
* ^From:.*carole
$MAIL
#Change to subjects that you want to be sure not to accidentally filter
:0:
* ^Subject:.*ebay
$MAIL
#This filter lets through mail from people using pine.
:0:
* ^References:.*pine
$MAIL
#The following is a sample of how to filter on mails with word
# 'filter' in message body and send them to mail folder filter.
# Useful if you want to save mail from a list.
:0B:
* ^*filter
${HOME}/mail/filter
#The rest of this throws out (to /dev/null) spam.
#All mail with java in it is spam.
:0:
* ^Content-type:.*java
/dev/null
#I don't know what this is, but it is always spam.
:0:
* ^X-Message-Info:
/dev/null
#All mail with embedded images is spam.
:0B:
* ^.*img.src
/dev/null
#All mail sent to these characters is spam, unless you are carson.
:0:
* ^To:.*benny
/dev/null
:0:
* ^To:.*carson
/dev/null
:0:
* ^From:.*carson
/dev/null
:0:
* ^To:.*cme
/dev/null
:0:
* ^To:.*gmike
/dev/null
:0:
* ^To:.*hal9000
/dev/null
:0:
* ^To:.*krex
/dev/null
:0:
* ^To:.*makji
/dev/null
:0:
* ^To:.*pez
/dev/null
:0:
* ^Cc:.*true
/dev/null
:0:
* ^Subject:.*medication
/dev/null
#perscription, prescripiton
:0:
* ^Subject:.*p..scrip..on
/dev/null
:0:
* ^Subject:.*\|
/dev/null
## The following lines (0B) filter on message bodies only:
##The first will catch cialis, c!ial1s, etc.
:0B:
* ^.*c.al.s
/dev/null
:0B:
* ^.*deals
/dev/null
:0B:
* ^.*medication
/dev/null
:0B:
* ^.*meds
/dev/null
#the . is any character, the ? means character before it is optional
# so this catches mortgage, mor.tg.agge etc.
:0B:
* ^*mor.?tg.?agg?e
/dev/null
:0B:
* ^.*pills
/dev/null
:0B:
* ^.*\%.rate
/dev/null
:0B:
* ^.*refinance
/dev/null
:0B:
* ^.*rx
/dev/null
:0B:
* ^.*s0ft
/dev/null
#leave the next line so real mail will arrive safely
:0:
$MAIL
That's fine, but one's mailbox filling up isn't always about SPAM: Friends could send you several large file attachments, and that would do it.
That is why I started the filter with a section that forwards large mails to some other address. You could also just set the filter to send the large mails to /dev/null if you don't have a forwarding address. I added this filter (with much help from mcnally) after a friend sent me a 700K photo (which she following with 1.3 and 1.4MB). See the line with 1000 in it.
I think it'd be better to fix the underlying problem, which is that the /tmp quota is apparently smaller than the mailbox quota.
SBC is at Provide.Net working on the lines now. Apparently, the problem is extensive; they hope to get it fixed tonight, but if they can't, they'll be back on it first thing in the morning. Right now, the lines are ringing open.
Thanks for the update, Joe. I wonder what the problem was? Bad wires, I suppose.
"Bad dates"
They have to replace the cable and will be out tomorrow to finish the job.
Thanks to whoever fixed my /tmp quota!
I've now reset the quotas on /tmp for all users who have logged in since the middle of December (Dec 19 14:45 2004, to be precise). I've also set newuser to assign the new quota to any new users.
I note that there were a couple of connections on tty01 today, before SBC took the lines down for repair. So I did get the modems connected to the 'right' lines this time. :)
Thanks again Joe. Clever of you not to assign new quotas to non-using users who would just get more spam. Today my abridged spam filter posted above caught 13 of my 15 spams that arrived since midnight: 3 X-message, 3 img.src, 1 java, 2 medication, 2 meds, 1 pills, 1 prescripiton, also (using the longer filter) a generic.drug and a To: true. Before I rearranged the order of filters half of the spams were caught because of the To: line. It is interesting how most of my current spam is due to our lack of national health insurance or care.
I think SBC has finished the work on our telephone lines.
I've got a "system non-problem" to report. Typically I get at least 50 spam e-mails per day. This morning there were only 5 spams. And just now I checked, and I have received *no* mail at all! What happened? Did someone connect a spam filter?
I tried to send an e-mail (from within Pine) and failed on three successive attempts. The third time I noticed a briefly flashed message in the status line at the bottom, something about "Error 451, error writing spool file, mail not sent." (not an exact quote.)
Make that: [Mail not sent. Sending error: 451 Error while writing spool file]
Probable cause: /dev/sd2d 6092380 6069840 -282078 105% /var
yep, just got /var: write failed, file system is full
in party
There are some overgrown log files in /var/log that could probably be rotated and compressed to help alleviate the crunch.
I dialed in! Thanks Joe, for dealing with SBC.
Thanks aruba!
Thanks, Jim!
Thanks, by the way!
Thanks, naftee!
Thanks, EVERYBODY!
Yeah!
someone please fix it, so I can save changes to the graphics and photo conf login.html files
I have been dialing in since last night.
I'm not familiar with that file, so I can't help, eprom. Drop a line to either cfadm@cyberspace.org or staff@cyberspace.org.
I can't telnet to grex, but ssh works. Telnet is failing from different machines in different places.
Yep, cannot reach cyberspace.org via telnet. This is being posted via backtalk (web).
This morning someone originally from Samarkand sent me a tel asking if I was admin (maybe because I was dialled in) and trying to say something about a problem with inetd and port settings. He was I think trying to let me know how to fix the problem so I asked him to explain to staff@. After which we discussed languages and cooking for 3 hours! Jews from Samarkand speak Bokhari as well as about 5 other local languages some of which I never heard of. His family was killed in the 90s and he escaped via Greece, where the local Uzbeks were threatening him so he came here and is doing something with software that he finds boring but which lets him chat all morning. Does Win98 have ssh capabilities? A friend just called to ask how she can access grex now that Chelsea is a long-distance call and telnet is dead. She has been sending in her dues faithfully and hoping grex would stabilize, silly girl.
You can download putty for win98.
same thing as marcvh noted a few hours ago. grex does not respond to telnet, and hasnt been responding to telnet all evening, but does respond to ssh (which Im on currently) and backtalk. why would this be happening?
Backtalk is crashing nastily for some reason, at least for anonymous access.
Unlike ssh and http, telnet is started out of inetd. For some reason, inetd was hosed. "kill -HUP" didn't help, so I killed it dead and restarted it.
Thanks Joe! Don't know what grex would do without you. My friend has no idea how to download Putty for Win98, or even what a file is. Someone set her up to dial provide.net and use telnet.
thanks keesan!
Heh, looks like provide.net will be getting a lot of business from Grexers now. Now if only I could get their "TurboConnect" to stop dropping calls on me...
This response has been erased.
I just telnetted in no prob. Whoever fixtit: THANKS!
Re 783: See #778. Joe fixed things - as usual.
thanks tod!
THANKS JIM
/var is full again.
Next time any disk work is done, would it be possible to either increase the size of the /var partition above 3GB or to put /var/mail (or /var/spool/mail, or wherever OpenBSD puts it) on its own partition?
Slash var slash mail is on its own partition!
if you google putty download you'll get a page where you can download the small .exe. just run it.
re #789: you're right -- somehow I overlooked it before.. it'd still be nice to either have a bigger /var or have whatever keeps filling it up isolated elsewhere..
I got /var down to 67% full by moving stuff over to /grex/zcore. Hopefully i didn't screw too much up...
Re putty download, my friend has no idea how to download anything. We tried several times to teach her to download files from grex and gave up. She does not know what a file is. Or where the files are in the first place. I tried to draw pictures. Or how to run a program. We set her up to type grex in order to reach her email here with pine, after turning on the computer. Someone else managed to set her up with IE and telnet. I tried to explain how to change from white on black to black on white printing. She lives several hours bike ride from here and it is snowy out.
thanks, i!
party comes up with this message /var: write failed, file system is full
Yes, it's full again.
.
EOT
&
/var: write failed, file system is full
/var: write failed, file system is full
pre
& h
tough to send email wehn this happens ....
!date
Fri Feb 25 05:35:14 EST 2005
Ditto #792, /var's at 65% now.
still can't get backtalk to save configurations changes...can someone look into this????
yeah ?????
I think it's going to have to wait for Jan, Jeffrey.
Wow. And here I thought we'd be done with the whole 'disk full' thing for a while after getting the new system.
Latest problem is that seven of us were waiting for someone with a key to show up at the pumpkin. Jim and I walked to my apartment to clean out the yard so it will hold a desk and three big chairs. We can hold onto grex stuff for a month at most while people decide whether to pick anything up. Metal shelving, mounting rack, but mostly SUN stuff and maybe 3 mono monitors. A few phones. Don't know what the movers will keep.
GrexStand
The move went great, once the key issues was resolved (thanks Aruba!). I snagged my old 5-line office phone and several big Sun VME cards.
Scott, what are you going to do with the big cards? We had hoped to recycle
the rest of them but our neighbor stopped by to see if anything was worth
putting on eBay and informed us that Friedman's Nonferrous Metals Recycling
closed a month ago. There are rumors they may have opened a different place
in Ypsi, which is not much help to us.
We spent a few hours taking apart: 2 keyboards (cracked, missing
keys), 2 IBM PCs (one with a hard card, the other with 2 full-height floppy
drives), a 9600 bps external modem (Practical Peripherals), several drive
enclosures for full-height scsi drives, a 9-in TI printer. THe box of
diskette labels did not have an inventory number. We still have the previous
grex computer, lots of cards, some T1 stuff, something that you can plug about
10 modems into, two SPARC terminals, two Ann Arbor Ambassador dumb terminals,
three monitors that might be TTL with burnin, a Sparc keyboard, another
keyboard, a few smaller boxes we cannot yet identify, some wires, one internal
modem (14.4 or slower?), a 486 motherboard, and a desk and three big chairs
and maybe some shelving. Anyone want any of this? A Sparc mouse with three
elongated buttons, cute. Various metal plates that may have gone with the
rack that went to provide.net. We are wondering WHY grex kept some of the
above when it was over 20 years old.
We are putting the smaller metal pieces into the recycle bin but does
anyone know of a place that accepts ferrous metal and does not charge to take
it? (They used to pay, but we lost both local junkyards). Kiwanis takes its
computer stuff (and will take our monitors and terminals) to a place that pays
5 cents/pound - maybe they will come pick up our boards and larger cases.
We used to get about 15-20 cents/lb for 'greenboard'.
It should be a while until the home-directory partitions fill up. However, /var seems to be something else.
it's ur mom
Re #807:
I might have a use for the 9600 baud modem, as a backup for my 33.6K.
I took those VME cards for decorative purposes. :)
Drew, we recycled that modem but you are welcome to one of our 14.4K modems - internal or external. We also recycled most of our 14.4 externals and only kept 4-5. And about 10 internals.
It was pointed out to me by someone on my blog that when one clicks on links that are on Grex's home page (http://grex.cyberspace.org), you get a "back talk crash" Specifically, this happens when one clicks on the links to sample the conferences. Example: http://grex.cyberspace.org/cgi-bin/backtalk/peek:agora
Dude.b
A report from party is that FTP is not working for removing files from Grex. I tried it myself and got a timeout.
Is anyone planning to fix lynx, links and w3m so they can access ftp sites if you are a paying member of grex?
And hopefully also upgrade lynx to 2.8.5, which has been the latest stable version for a while now. Thanks.
Lynx 2.8.5 stable release came out Feb 2004 and they are now working on 2.9. 2.8.5 has a lot of bug fixes and performance improvements and some new features. Is there a precompiled version for OpenBSD?
OpenBSD 3.6 includes lynx 2.8.5rel2 - has grex upgraded to 3.6 yet? 3.5 includes lynx 2.8.4rel1. How often does a new OpenBSd come out?
calm down, keesan
This response has been erased.
keesan demands complainace - immediately! her orders are issued! comply! oh, wait ..... envermind ,
follow her instructions ! ..
Picospan and Fronttalk seem to have a different idea of how many newresponse items I have in this conference. Pico says 151, FT says 21. FT is correct, I think. I've been reading the conferences mostly with Backtalk.
This response has been erased.
Not quite. Soon, maybe.
i hope we still have link to picospan
Yes, you will still be able to run picospan, with the command "picospan".
Oh, cool! But I still haven't seen the self-destruction of picospan functionality with my own eyes, as reported in this item.
Yeah. That's bull-, and possibly horse-, shit.
I fully agree with scholar's mammalian dissertation.
While it was mammalian, the more specific theme I was going for was ungulate. But good try, man, good try.
Seeing as you do have hooved feet, I guess I forgot about your affinity for your cousin ungulata. Apologies.
Is it just me? Grex is timing out... :(
Timing out how? I note that it has been up since Saturday.
I just restarted inetd; apparently it was hung. It is possible that that was what was affecting FTP, too.
Re. #834: If by "timing out" you mean that Grex is (or was) not accepting telnet connections, no, it is not just you; you are not alone, which (nonfunctionality of telnet) necessitates my present access of the Agora via Backtalk.
"laston" is not working. Attempting to go to Backtalk via the cyberspace.org home page results in a Backtalk crash. Still. Weeks after the first notice. (How hard can this be to fix?)
are you the famous scholar r. k. sawyer jr.?! wow! hey, guys! this one isn't a fake!
This response has been erased.
strange. 'laston' worked for me . . once. or twice.
then no more.
Gaston.
When was it laston? ;-)
This response has been erased.
tell russ to stop arguing, then.
dude.
big guy.
Don't know if it counts as a "problem", exactly, but the help text for the
"Respond or Pass" prompt in pico contains a typo:
<integer> - <inteber> -- displays just that range of responses.
Also, 'STop' should probably only have one capital, unless it's being used
to indicate a shortcut in a way that none of the other items are.
On the bright side, there's no "teh" anywhere.
jiucy
Re. #839: Probably not, but if so, it's only because I'm a midget standing on the shoulders of giants, some of whom post in these fori.
Jim has me on his .yeswrite list for tels and talks but when I typed talk jdeigert 'your party is refusing messages'. Did we do something wrong? I just checked it. messages no (exceptions) - I am one of two. mode any bells yes. This should be in .login for jdeigert. He can request a talk with me but not vice versa. We are both telnetted in on the same connection over my new little network and we are sitting side by side and practicing 'talk'.
HE"S CHEATING ON YOU.
His .yeswrite file lists keesan.
but his penis says rebecca.
The only thing I can see that *might* cause trouble is the blank lines at the end of his .yeswrite. But that may not matter; I don't know.
If you count the number of blank lines you'll ascertain the length of his mistress' name.
When I get disconnected from Grex while I'm reading mail, and then reconnect ad go back to reading mail, Pine marks my mailbox as read-only, because it says it's open in another process. It has always done this, but lately it seems to take a really long time for it to stop doing it. Is that because the idle-zapper is taking a long time to kill my old login session? It's been idle for 57 minutes now, but still present. Can I kill it myself?
Yes, you should be able to kill it yourself. Try:
ps -auxwwww | grep aruba
to get the processid of the old pine session and then
kill <processid>
You may need to
kill -9 <processid>
Thanks! That worked.
Thanks aruba!
This response has been erased.
Just letting someone who reads this know that the 'Computers & Staff' page has not been updated to reflect the new system (on Grex.org) Thank you
I'll never understand why Jan writes gobs and gobs of text on the abortion issue and the end of the oil era, but not on the new GreX machine :(
I've been getting increasing amounts of junk/spam email on my grex email. Has anyone else been getting a lot of junk email? Is it possible that some junk emailer is automatically targeting random grex addresses?
Jim has not received any at all so probably not.
I have no junk mail.
This response has been erased.
What are you filtering on? I have not been receiving truncated fragments (first three lines of a header) recently. I just added a filter on X-RBL-Warning, which removes anything on the spamcop blacklist, which is about 1/3 of my spam or maybe more.
I think I've noticed a decrease, but I haven't really been keeping track.
You have several choices: