Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-11 Thread Greg Smith
On Sun, 10 Aug 2008, Scott Marlowe wrote: The good news is that both Centos 5.2 and Ubuntu 7.10 seem immune to this particular bug, and have been running 13 hours now without a hitch. Not sure if it's relevant here, but you do know that I've been kicking back to lkml that pgbench has issues o

Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-10 Thread Scott Marlowe
On Sat, Aug 9, 2008 at 2:54 PM, Scott Marlowe <[EMAIL PROTECTED]> wrote: > On Sat, Aug 9, 2008 at 2:51 PM, Tom Lane <[EMAIL PROTECTED]> wrote: >> "Scott Marlowe" <[EMAIL PROTECTED]> writes: >>> I'm load testing a machine, and i'm seeing idle in transaction >>> processes that are no longer hooked to

Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-09 Thread Scott Marlowe
On Sat, Aug 9, 2008 at 2:51 PM, Tom Lane <[EMAIL PROTECTED]> wrote: > "Scott Marlowe" <[EMAIL PROTECTED]> writes: >> I'm load testing a machine, and i'm seeing idle in transaction >> processes that are no longer hooked to any outside client, that pull >> 100% CPU and can't be kill -9ed. > > To my k

Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-09 Thread Tom Lane
"Scott Marlowe" <[EMAIL PROTECTED]> writes: > I'm load testing a machine, and i'm seeing idle in transaction > processes that are no longer hooked to any outside client, that pull > 100% CPU and can't be kill -9ed. To my knowledge, the only way a process can't be kill -9'd is if it's stuck inside

Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-09 Thread Scott Marlowe
On Sat, Aug 9, 2008 at 1:47 PM, Stephen Frost <[EMAIL PROTECTED]> wrote: > * Scott Marlowe ([EMAIL PROTECTED]) wrote: >> I take that back. This problem followed the RAID card from one >> machine to another. > > That's certainly curious. The kernel backtrace didn't seem to have > anything terribly

Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-09 Thread Stephen Frost
* Scott Marlowe ([EMAIL PROTECTED]) wrote: > I take that back. This problem followed the RAID card from one > machine to another. That's certainly curious. The kernel backtrace didn't seem to have anything terribly interesting in it (at least to me). Sure there aren't more detailed logs? With

Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-09 Thread Scott Marlowe
On Sat, Aug 9, 2008 at 1:38 PM, Scott Marlowe <[EMAIL PROTECTED]> wrote: > On Sat, Aug 9, 2008 at 1:31 PM, Stephen Frost <[EMAIL PROTECTED]> wrote: >> * Scott Marlowe ([EMAIL PROTECTED]) wrote: >>> Aug 9 13:13:21 engelberg kernel: [71242.735046] >>> >>> Does this look like a kernel bug or a pgsql

Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-09 Thread Scott Marlowe
On Sat, Aug 9, 2008 at 1:31 PM, Stephen Frost <[EMAIL PROTECTED]> wrote: > * Scott Marlowe ([EMAIL PROTECTED]) wrote: >> Aug 9 13:13:21 engelberg kernel: [71242.735046] >> >> Does this look like a kernel bug or a pgsql bug to most people? > > It's certainly something kernel-related. It might be t

Re: [GENERAL] 100% CPU pg processes that don't die.

2008-08-09 Thread Stephen Frost
* Scott Marlowe ([EMAIL PROTECTED]) wrote: > Aug 9 13:13:21 engelberg kernel: [71242.735046] > > Does this look like a kernel bug or a pgsql bug to most people? It's certainly something kernel-related. It might be the OOM killer though.. You might want to disable that. Is the box running out

[GENERAL] 100% CPU pg processes that don't die.

2008-08-09 Thread Scott Marlowe
I'm load testing a machine, and i'm seeing idle in transaction processes that are no longer hooked to any outside client, that pull 100% CPU and can't be kill -9ed. I'm using pgbench -c 1000 -t 1000. postgresql.conf attached. This is on a 8 CPU AMD box with hardware RAID. I'll likely never see t