pg.ddx.io  pgsql-performance@postgresql.org mailing list archive  
help / color / mirror / Atom feed
Tanking a server with shared memory
2+ messages / 2 participants
[nested] [flat]

* Tanking a server with shared memory
@ 2004-09-06 01:15 Martin Foster <martin@ethereal-realms.org>
  2004-09-06 02:18 ` Re: Tanking a server with shared memory Tom Lane <tgl@sss.pgh.pa.us>
  0 siblings, 1 reply; 2+ messages in thread

From: Martin Foster @ 2004-09-06 01:15 UTC (permalink / raw)
  To: pgsql-performance

I have been experimenting with the  'IPC::Shareable' module under the 
native implementation of Perl 5 for OpenBSD 3.5.  While it is not loaded 
by default it is a pure pure implementation.

I have tested this module under two machines, one which used to run 
PostgreSQL and has a higher then normal amount of SYSV semaphores.  The 
other has a normal amount, when testing under the former database server 
things load up fine, clients can connect and all information is as it 
should.

When I test under the normal setup the machine tanks.    No core dumps, 
no errors produced, just a near instant lock-up of the server itself and 
that is with a non-privileged user.

While I know this is a Perl issue, but figured I might be able to gain 
some insight on how a server could drop without at least generating a 
panic.  Any ideas?

	Martin Foster
	Creator/Designer Ethereal Realms
	martin@ethereal-realms.org



^ permalink  raw  reply  [nested|flat] 2+ messages in thread

* Re: Tanking a server with shared memory
  2004-09-06 01:15 Tanking a server with shared memory Martin Foster <martin@ethereal-realms.org>
@ 2004-09-06 02:18 ` Tom Lane <tgl@sss.pgh.pa.us>
  0 siblings, 0 replies; 2+ messages in thread

From: Tom Lane @ 2004-09-06 02:18 UTC (permalink / raw)
  To: Martin Foster <martin@ethereal-realms.org>; +Cc: pgsql-performance

Martin Foster <martin@ethereal-realms.org> writes:
> While I know this is a Perl issue, but figured I might be able to gain 
> some insight on how a server could drop without at least generating a 
> panic.  Any ideas?

The standard spelling for this is "kernel bug".  Send a reproducible
example to the OpenBSD kernel maintainers.

			regards, tom lane



^ permalink  raw  reply  [nested|flat] 2+ messages in thread


end of thread, other threads:[~2004-09-06 02:18 UTC | newest]

Thread overview: 2+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2004-09-06 01:15 Tanking a server with shared memory Martin Foster <martin@ethereal-realms.org>
2004-09-06 02:18 ` Tom Lane <tgl@sss.pgh.pa.us>

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox