Re: 2.6.10-rc3-mm1-V0.33-04 (private thread?)
From: Gene Heskett
Date: Mon Dec 20 2004 - 14:22:27 EST
On Sunday 19 December 2004 08:56, Gene Heskett wrote:
>On Saturday 18 December 2004 07:48, Gene Heskett wrote:
>>On Saturday 18 December 2004 07:08, Gene Heskett wrote:
>>>I must report a couple of anomalies while running
>>CORRECTION! 2.6.10-rc3-mm1-V0.33-0, not 1!
>And I am now officially confused. These messages *are* coming back
>from the server, but I am also getting delayed by a day, bounce
>messages from the server? Anybody have any idea whats fubared?
>Also, in the FWIW dept, I'm back on 2.6.10-rc3,
>realtime-preempt-2.6.10-rc3-mm1-V0.33-04 also went away in the
> middle of an amanda run yesterday moring after an uptime of about
> 45 minutes, leaving no trail whatsoever in the messages log, and
> the only possible clue was that the last runtapes session amanda
> ran, with should have taken in the vacinity of 400 seconds, took
> only .389 seconds, but no error was logged in the amanda-dbg
> directory. So I have no clue. The only consistent clue is that an
> lsof locks itself till ctrl-c'd, and an amcheck run cannot wake the
> client on the server, but connects with the client on the firewall
> box just fine. And it works after a reboot.
>>>Twice now, odd goings on, such as lsof just locks itself till you
>>>give it a ctrl-c, Only a reboot fixes this.
>>>And I assume its related, but NDI where, amanda's amandad gets
>>> stuck and this machine was not backed up this morning. The
>>>/tmp/amanda-dbug/* files do not seem to offer any clues as to why,
>>>other than an occasional timeout. The other client machine on
>>> this network, my firewall, was backed up normally, but nothing
>>> here made it to tape. This occured once before on a previous
>>> version, 32-19 I think but won't swear to, and I believe if I
>>> reboot right now, and restart amdump, that it will work, so
>>> something would appear to be uptime sensitive.
>>>kmail has repeatedly lost its connection to outgoing.verizon.net,
>>> but a restart of kmail seems to restore that (if its not verizons
>>> fault, they *were* playing with it earlier today). But this has
>>> been an ongoing problem.
>>>I'll go get the latest and install it for effects.
>See above, 33-04 was also a no-show.
And, I've rebuilt it again, and it seems to be working. The only
gotcha I've noted so far is that I watched about 20 minutes worth of
tv with tvtime, which worked just fine AFAIWC, but then the log
spits out in the next time update, that my clock is nearly a minute
fast! Running with a tickadj setting of 9925 here, its been staying
within +-.4 seconds per hour, so tvtime is doing something odd when
an ntpdate run says the clock needs adjusted by
Dec 20 12:40:40 coyote ntpdate: step time server 188.8.131.52
offset -60.516102 sec
in an hour of elapsed time.
But so far, lsof and amcheck are running ok. That was my pair of
clues that a lockup was emminent before.
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
99.30% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attorneys please note, additions to this message
by Gene Heskett are:
Copyright 2004 by Maurice Eugene Heskett, all rights reserved.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/