Jeff Roberson jroberson at chesapeake.net
Fri Nov 2 12:00:48 PDT 2007

On Fri, 2 Nov 2007, Michael wrote:

> Hello, i know that you are working on freebsd scheduler, probably this
> bug have relation to your work:
> After upgrading to Freebsd 7 (from RELENG_6), postgresql coredumps several
> times per day. I asked mail list pgsql-bugs and got answer that this
> is probably due to changes in scheduler (i tried both SCHED_ULE and
> http://www.nabble.com/PostgreSQL-crash-on-Freebsd-7-t4693969.html

I hope you don't mind me forwarding this to stable.  Does this problem 
occur with the 4BSD scheduler?  Is your system swapping?

>> What seems like a reasonable bet at this point is that the FBSD7
>> kernel's scheduler has been changed in a way that makes it possible for
>> it to sometimes not schedule a process for a very long time (order of a
>> couple minutes).  If that happened while the process was holding a
>> spinlock then other processes waiting to get the spinlock would fail
>> like this
> Can you dig into this problem?
> If this is incorrect email to ask for help, then sorry for
> disturbing.
> Michael

More information about the freebsd-stable mailing list