Author: Tom Robinson <tom.robinson AT motec.com DOT au>
Date: Tue, 01 Sep 2009 13:51:45 +1000
Hi, Thanks for the feedback Frank. I am running dump. After re-nicing the sendbackups and dumpers to first -1 and then -3 the load average still hovers at zero: load average: 0.00, 0.00, 0.00 Re-nici
Author: "Dustin J. Mitchell" <dustin AT zmanda DOT com>
Date: Tue, 1 Sep 2009 00:25:41 -0400
On Mon, Aug 31, 2009 at 11:51 PM, Tom Robinson<tom.robinson AT motec.com DOT Hmm, the causation may be going the other way -- if the disk is generating too many IRQs for the CPU to handle, then netwo
Author: Jean-Louis Martineau <martineau AT zmanda DOT com>
Date: Tue, 01 Sep 2009 07:45:33 -0400
We need to find a minimal patch that fix the problem. Cat you try the attached patch? Jean-Louis Michael Burk wrote: I applied the patches to the 0827 version again (had to do the sendbackup-dump.c p
This is starting to look like a kernel bug -- is there an associated OpenBSD bug or something that we could reference in comments in the code to explain this strange formulation? Also, we'll probab
Author: "Dustin J. Mitchell" <dustin AT zmanda DOT com>
Date: Tue, 1 Sep 2009 10:33:06 -0400
This is starting to look like a kernel bug -- is there an associated OpenBSD bug or something that we could reference in comments in the code to explain this strange formulation? Also, we'll probably
Since the logs indicate an error with the "index tee", I thought I'd try turning off the index generation. The backup still failed, but with a different error: 1251826429.341455: sendbackup: pid 2672
I applied the 3-line patch to the 0831 snapshot and ran a full backup on both machines, with 4 file systems each. All 8 completed successfully with no "strange" messages. Next, I commented out the 3
Author: Jean-Louis Martineau <martineau AT zmanda DOT com>
Date: Tue, 01 Sep 2009 15:18:26 -0400
I have nothing else to try. The order of system call is a follow: In amandad process: pipe(pipefd) dup2(pipefd[1],b) fork in the child: exec sendbackup process dup2(b,c) write c #fail with EAGAIN Th
Author: "Dustin J. Mitchell" <dustin AT zmanda DOT com>
Date: Tue, 1 Sep 2009 15:25:23 -0400
If it's not too hard, it would be nice to have a ktrace or equivalent of this, first to look at here, and second to take to the OpenBSD list. I know that's tricky since this is a daemon process.. Dus
[...] I'm no expert on this topic, but if I were investigating something like this, I'd be curious to know if all three of the lines in the patch were necessary for the fix. Since, as you pointed out
If it's not too hard, it would be nice to have a ktrace or equivalent of this, first to look at here, and second to take to the OpenBSD list. I know that's tricky since this is a daemon process.. D
Author: "Dustin J. Mitchell" <dustin AT zmanda DOT com>
Date: Wed, 2 Sep 2009 13:39:03 -0400
Can you have amandad sleep for, say, 120 seconds just before it launches sendbackup, and somenow notify you of the pid to which you should attach ktrace? Maybe by writing it to /dev/console or to sys
Author: Brian Cuttler <brian AT wadsworth DOT org>
Date: Wed, 2 Sep 2009 13:51:08 -0400
Amanda 2.6.1 on Solaris 10/Sparc Amanda 2.6.1, Solaris 10x86 Server has 21 clients with a total of 109 DLEs. One of the client systems has 51 DLEs, 1 ufs and 50 zfs partitions. The partitions/DLE are
Author: Tom Robinson <tom.robinson AT motec.com DOT au>
Date: Thu, 03 Sep 2009 14:28:39 +1000
Thanks Dustin, I've found that our very old (RH7.1 seawolf), running a very old kernel (2.4.20) has a bug in the ide driver. I can't say categorically that this is the root cause of the dump issue I
Author: Nathan Stratton Treadway <nathanst AT ontko DOT com>
Date: Fri, 4 Sep 2009 10:14:38 -0400
Are you referring back to the results you posted to here (to amanda-users) on Sept 1? (That is, this message: http://www.mail-archive.com/amanda-users AT amanda DOT org/msg42749.html ) I take it you
Author: "Dustin J. Mitchell" <dustin AT zmanda DOT com>
Date: Fri, 4 Sep 2009 13:11:39 -0400
Attached is a test program I just put together which does about what Jean-Louis specified above (with the addition of some closed fd's). This works fine on my mac, which is the closest approximation
Author: Nathan Stratton Treadway <nathanst AT ontko DOT com>
Date: Fri, 4 Sep 2009 17:01:46 -0400
I tried to understand the program flow for sendbackup.c and sendbackup_dump.c, and I agree with Dustin that it is "pretty crazy", but as far as I can tell the only place that would generate the messa
Author: "Dustin J. Mitchell" <dustin AT zmanda DOT com>
Date: Fri, 4 Sep 2009 17:13:19 -0400
Looks just like it does locally. If the test had managed to reproduce this failure, then I would have expected to see write: Resource temporarily unavailable That means something deeper is going on h
Hi Dustin, Thanks again for your help. Here's the output of the test prog: burkm@selenium$ ./test parent pipe = r@3 w@4 parent closing parent sleeping child closing p[0] child exec'ing child duping c
Author: Nathan Stratton Treadway <nathanst AT ontko DOT com>
Date: Fri, 4 Sep 2009 18:40:55 -0400
I wonder if more data needs to get sent down the pipe before the unexpected "Resource temporarily unavailable" result is returned? If I have followed this thread correctly, it seems like the tests sh