Very Computer

by **Brian Kram** » Mon, 27 Mar 1995 12:57:13

I've been experiencing strange behavior on my linux box since I repartitioned
the drives to allocate more space for news. I'm looking for suggestions
since I am stumped.

Basically..the machine locks up for no aparent reason and gives no
warning. It acts as my news server and www server as well as handling uucp.
I'm running inn and ncsa's server.

What would be places to look to try to isolate the problem?

I'm running 1.1.35 with the quota patch. Up until making these changes
it was working fine (2-3 week uptimes).

Here's the setup.

Gateway 2000 (Micronic MB) 486/66
40 MB Ram
1 340 MB IDE
1 420 MB IDE
1 1800 Meg Quantum SCSI
Adaptec 1542CF (Bios enabled, 5 meg/s, fast+sync) I did recently change
these from bios disabled, and no fast+sync).
3 Com Etherlink
Boca Board 16 port serial board

I have also recently upgraded to a T1..and news processing has gone
from an article every 1.5 secs to 2 every sec.

--

New Jersey Computer Connection - Public Access Unix Site - pluto.njcc.com
Voice: 609-896-2799 - Fax: 609-896-2994 - Dialups: 609-896-3191 & 896-2722
Dialup or Telnet to pluto.njcc.com and log in as guest for more information.

by **Daniel Panca** » Tue, 28 Mar 1995 15:06:30

>I've been experiencing strange behavior on my linux box since I repartitioned
>the drives to allocate more space for news. I'm looking for suggestions
>since I am stumped.
>Basically..the machine locks up for no aparent reason and gives no
>warning. It acts as my news server and www server as well as handling uucp.
>I'm running inn and ncsa's server.
>What would be places to look to try to isolate the problem?
>I'm running 1.1.35 with the quota patch. Up until making these changes
>it was working fine (2-3 week uptimes).
>Here's the setup.
>Gateway 2000 (Micronic MB) 486/66
>40 MB Ram
>1 340 MB IDE
>1 420 MB IDE
>1 1800 Meg Quantum SCSI
>Adaptec 1542CF (Bios enabled, 5 meg/s, fast+sync) I did recently change
>these from bios disabled, and no fast+sync).
>3 Com Etherlink
>Boca Board 16 port serial board
>I have also recently upgraded to a T1..and news processing has gone
>from an article every 1.5 secs to 2 every sec.

Join the crowd....Search for the keywords "hang", "SUDDEN DEATH"
"lock" in al the linux groups. One thing I have in common with you is
the BOCA2016, and I am running a news server, however, it's my linux
dialin machine that locks... This is the 1st time that I have heard
of a 3 Com having this problem... most of us are using NE2000 cards...

I was told that it was a IRQ conflict.... So I changed my IRQ from 15
to 9... uptime = 1 day 5 hours.... I will need at least 4 days uptime
before I am convinced...

If you find out any more please let us all know....

Dan
--------------------------------------------------------------------------
Daniel Pancamo (713)286-0399 voice InfoNet Communications (ISP)

19.95 Month SLIP http://www.infocom.net Houston, Texas 77259-0343

by **Brian Krame** » Tue, 28 Mar 1995 10:35:43

> See if you can correlate your freezes with heavy IDE disk usage. Can you
> ping your machine when it `freezes'?

Yes to both..it happens mostly during expires. I'm thinking about getting
rid of the ide drives...do you think that will help?

Quote:> I recompiled expire without optimization and with debugging info and for some mysterious
> reason I've been able to run expire under gdb successfully (no freeze) about 5 times
> and have never had a freeze running expire under gdb. If I go another couple of days
> without a freeze, I'm going to try running the newly compiled expire without gdb to
> get rid of any effect that gdb might be having.

I assume you are running inn 1.4. I'd appreciate it if you could send me
a binary when you do that. I don't have the source handy...

Quote:> At first, I thought that I was having problems with the SCSI subsystem, but I started
> running expire with the -z option and it would still crash. All that it was doing was
> reading from and writing to one of my IDE drives.

I think it is IDE related too....the load goes way up cause of the ides.
I run BSDI on my other machine, and have had 42+ days uptime since
the last reboot. If I can't get this fixed soon, I'll have to switch.

New Jersey Computer Connection - Public Access Unix Site - pluto.njcc.com
Voice: 609-896-2799 - Fax: 609-896-2994 - Dialups: 609-896-3191 & 896-2722
Dialup or Telnet to pluto.njcc.com and log in as guest for more information.

Very Computer

Sudden Lockups With No Clue As To Cause

Sudden Lockups With No Clue As To Cause

Sudden Lockups With No Clue As To Cause

Sudden Lockups With No Clue As To Cause