Welcome to Temporary News
Terry Knab
get news fixed at Nyx.
A few words of warning:
Nyx.* groups are set to propogate everywhere (I run nntpcache and can NOT
change it easily).
None of the prior posts exist currently.
Nyx's normal rules apply to all news usage.
I am offering this as a service to Nyx's users. Please do not abuse it.
Abuse will be dealt with as per usual Nyx Policy.
--
-------------------------------------------------------------------------------
Terry E. Knab
News Administrator
Nyx Public Access Unix
Eric Pass
Hi Terry, I was wondering if you could elaborate just a little
on the news server situation and status. Is there any more information
available other than that in the motd?
Thanks,
Eric
--
Eric W. Pass (ep...@nyx.net) | Lots of game stuff for sale on my
http://www.nyx.net/~epass/home.html | Web page, see URL at left!
---------------------------------------------------------------------------
"Bother", said Pooh, as his bloody gloves dripped on the floor.
Terry Knab
: Hi Terry, I was wondering if you could elaborate just a little
: on the news server situation and status. Is there any more information
: available other than that in the motd?
I'll try my best. :)
Basically what has happened is that the machines have had ongoing problems
with their hard drives. The plan is to rebuild them with new hard
drives, and start over...
the tree by the river
Eric Pass <ep...@nyx.nyx.net> wrote:
>
>Hi Terry, I was wondering if you could elaborate just a little
>on the news server situation and status. Is there any more information
>available other than that in the motd?
I'm up here at Nyx HQ on the new news machine at the moment; I've
only done the hardware and basic OS and software installation,
so it still awaits the nimble fingers of our admins to get it
serving news properly again.
In case you're curious, I went ahead and set up one of the dual
Pentium-II linux machines I'd been putting together for Nyx's
new login machines as the news-server-to-be. One of the handy
things about swapping the machine out with one of the new ones
is that all the machines are configured with exactly the same
hardware except for the number of disk drives--and so, we have
a spare system or two already set up that can be swapped for
any of these systems should we experience a hardware failure.
--
Soc.singles FAQ: < http://www.trygve.com/ssfaq.html > || "A minor setback..."
Personal webpage: < http://www.trygve.com > || -- Evil the Cat
Trygve Lode, president, Nyx Net, public access internet < http://www.nyx.net >
"Some say prayers, some say prayers...I say, 'Mine.'" -- SoM
Darlene Cypser
>Eric Pass <ep...@nyx.nyx.net> wrote:
>: Hi Terry, I was wondering if you could elaborate just a little
>: on the news server situation and status. Is there any more information
>: available other than that in the motd?
Trygve believes a "feature" of the debian Linux may have allowed the
drives to wrap after one gig and write over the operating system, etc.
His solution was to put the operating system in a partition less than 1
gig.
Elie disputes that this is the cause. However, we could come up with no
other good explanation as to why two machines with different motherboard,
bios, and harddrives (one IDE and one SCSI) would end up in precisely the
same state without any intervention. The "state" was mangled data.
(Fortunately Trygve believes that he was able to review the actual
software that runs news.)
Any other theories are welcome.
--
,___, ,___,
(o o) Darlene Cypser dcy...@nyx.net (o o)
/))^_) Secretary/Treasurer http://www.nyx.net (_^((\
// " " Nyx Net = Public Internet Access " " \\
Gary Heston
Darlene Cypser <dcy...@blackhole.nyx.net> wrote:
>Trygve believes a "feature" of the debian Linux may have allowed the
>drives to wrap after one gig and write over the operating system, etc.
>His solution was to put the operating system in a partition less than 1
>gig.
This is a fairly common bug in BIOSses which can't handle drives with more
than 1024 cylinders. I've never heard of it being a problem with any version
of Linux, although most of my experience is with Slackware. (Which I need to
play with some more--ten F/W 2.1GB SCSI drives should make a nice RAID for
the home network.)
>Elie disputes that this is the cause. However, we could come up with no
>other good explanation as to why two machines with different motherboard,
>bios, and harddrives (one IDE and one SCSI) would end up in precisely the
>same state without any intervention. The "state" was mangled data.
>(Fortunately Trygve believes that he was able to review the actual
>software that runs news.)
Was LBA (or is it LHA) enabled in the BIOSes in each system? If not, that
is likely the culprit.
I generally make root pretty small--200MB at most--and mount everything
like /usr, /var, all the spool and tmp directories, and home directories.
>Any other theories are welcome.
I vote for misconfigured BIOS.
Gary
--
Gary Heston ghe...@nyx.nyx.net Disclaimer, datclaimer...
Windows: User friendly, admin hostile, reliability challenged.