This item is for system announcements (new computer equipment on Grex, 
system upgrades, Grex meetings, etc.).  Personal announcements should go 
back in item 2; Grex system *problems* belong in the next item (#4). 

29 responses total.

I've installed a new version of 'nethack' on Grex.  There have been some
changes, including new character types.  Unlike the old version, the new
one is Y2K compliant.

All old saved games have been erased.  Sorry, but they would have stopped
working in the new year anyway due to Y2K bugs in the save file format.

I've installed a new version of the 'elm' mailer on Grex.  The old version
was not Y2K compliant.  Let me know if there are any problems.

Valerie has installed the special Y2K compatible "splishsplash" command.
Try

     !splishsplash

Valeries says she has no clue who Tom is.

wht's going on ?

splishsplash is very neat - thanks Valerie (:

Yeah... though I must admit, I was a little worried. It appears a little too
close to the bottom of the screen for my comfort. They usually advise against
doing that if it's that shallow. Now, if there were a tragic version of it,
I'd be a little happier. Not that I'd get off on the carnage, but at least
it would be a faithful representation.

Jan, you might have wanted to wait until 3.3.1 came out for nethack, like I
mentioned in the garage conference. There was a 3.2.3 patch that would have
made the old version Y2K compliant. 3.3.0 has some major bugs, including an
armor bug (with the right (easy to make) conditions, you can wear *anything*)
which can make for some huge abuses (such as near-immortality). But other than
that, 3.3.0 does add some nice features.

We'll install 3.3.1 when it comes out.  Meanwhile, have fun exploiting
the bugs in 3.3.0.

Fair enough.

        Well, that was fun. Interesting to see in the scroll back.

How was that fun?

Re 11, I assume that #10 is talking about !splishsplash .  See also 
!macarena .

The Grex Board of Directors meets this Wednesday, January 12, 7:30-9:30,
upstairs at Zingerman's Next Door, 422 Detroit Street, Ann Arbor. The
public is invited. See Item 150 in Coop (item:coop,150) for the agenda.

   Grex had a disk disaster this weekend, but in the continuing tradition
of managing to squeak by them without actually losing data.

   It started Saturday morning at 10:26, when the disk that holds the
mail directories suddenly became unwilling to talk with the disk controller.
Not suspecting anything out of the oridinary the system came back up
when rebooted and all seemed well.  Although rare, there are times when
Grex crashes for no apparent reason.  Unfortunately I discovered the reason
Saturday evening when I foud the error messages from the disk about its
not being there.  Those messages always fill me with dread, because they
usually bring the winds of disaster with them.  Not long after seeing the
previous errors, the system went down again.  Scott shut the system down
and when I got to the pumpkin Sunday morning, the mail disk appeared to
be fine.  Taking the oppurtunity to make a backup of it while things
all still appeared to work, said backup went fine.  The disk continued
to behave itself, until scg and I started to wonder if the disk simply
hadn't overheated, as one of the side panels to the disk enclosure box
wasn't on, altering air flow.  In the process of testing a new disk for
a new larger /c partition, the mail disk errors came back, badly enough
that the system was unstable once again--with mail data unreadable.  Taking
poor litle sd4h offline for the last time, we unhooked it from Grex and
said farewell to yet another 2G disk.  Using the 'new' disk, we started
a restore of the mail data only to find some bizarre problem with restore,
but found that the interactive version of a restoration worked, and did
that.  Several hours later, after a nice flyby from krj with some cold
Coke, the restore finished, and all looked OK--I saw no problems with
restore, and all the mail files I checked had the correct permissions
and ownerships.  Bringing the system back up resulting in the usual
attack from other sites waiting to dump mail.  I call this a mailstorm
but interestingly this was wasn't as bad as others in the past.  The
highested number of sendmails I saw running at once was only 124, which
was a far cry from the 301 I once saw.  After letting Grex catch up on
mail for a couple of hours it came back online about 11:40pm.

        Good show!

STeve, you're an alarmist.  "Disaster" means you lost some data.  "Emergency
disk replacement" is more like it.
(And thanks VERY much for managing it!!!!!!!!!!!!!!!)

Thank you!

  Nice Going STeve. Thanks a lot./

        Non-scheduled drive maintenance.

Job well done!

thanks, wizards of grex  :)   .

  Ha! Non-scheduled drive maintenance!  Excellent!

Where exactly is this linked to?

Item 86 in the Helpers conference.

This response has been erased.

   Grex now has another disk on it for home directories.  It's
called /c just like the old /c, except its about 2.5 times bigger
and will beable to hold quite a few accounts, and we *hope* that
the storage problems we've been having will go away.

Thanks to all.  At around 2 am when we stopped by to offer Kiwanis leftovers
to grex staffers, Jan showed us around the modem collection etc.  It is very
hot in the pumpkin.  Computers make good space heaters.  I hear that the next
plan is to set up a spare grex computer to do formatting etc. on rather than
having to shut down grex for this.

Filesystem            kbytes    used   avail capacity  Mounted on
/dev/sd0h             284215  240494   15300    94%    /bbs
/dev/sd4a            1944365  773287  976642    44%    /c
/dev/sd1c            1944365 1748977     952   100%    /a

It's great that we got the new disc!  Are we planning on migrating some
things from /a to /c?  And is b/bbs at 94% something to be concerned with?

It is, but we've got a plan....  The old /c partition is now free.  We will
probably be moving a lot of stuff around to shift things to bigger partitions.
Some stuff will probably move off of /a, but the next few reaps will also
help.  All this means more down time will be needed.

We plan to build a second machine for development work and for doing things
like formatting and testing disks.  Unfortunately, we'll need to format and
test a disk for the new machine, and this means more down time for Grex. 
Sorry.

We now have 5 disks and a tape drive on Grex's SCSI chain.  We can't add many
more drives.  We'll want to install a second SCSI controller and put the
drives on two different chains.  We have extra controllers, but have no
experience setting them up.  This is another thing we want the development
machine for.

Response not possible - You must register and login before posting.

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss