proclock table corrupted

Harshitha S <hershetha@gmail.com> writes:

We are encoutering the following error during normal operation of postgres.
postgres[10982]: [2-1] PANIC: proclock table corrupted

Ugh. Can you provide a reproducible test case?

Version of Postgres : 9.0.3
Architecture : mips
OS: RedHat Linux

[ raised eyebrow... ] I've been working at Red Hat for ten years, and
I'm pretty sure they have never shipped a MIPS-based distro in that time.
So what is that OS really?

Can you please let me know if 'fix-strong-lock-cleanup.patch' and this
error are related?

This is not an adequate identification of what patch you are talking
about; but if you are speaking of something related to Robert Haas'
fast-path locking code, that's not in 9.0.x.

regards, tom lane

Sorry, the OS is WindRiver Linux.
Yes , I am taking of the fast path locking patch discussed in the link
below.

http://postgresql.1045698.n5.nabble.com/bug-in-fast-path-locking-td5626629.html

Regards,
Harshitha

On Fri, May 25, 2012 at 7:09 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:

Harshitha S <hershetha@gmail.com> writes:

Sorry, the OS is WindRiver Linux.

What I would suspect first is issues with the MIPS spinlock assembly
code (look into s_lock.h) not being portable to your platform. That
code hasn't been tested on very many machines, I suspect. It's not
impossible that it doesn't work at all on multiprocessor machines;
is yours one?

Yes , I am taking of the fast path locking patch discussed in the link
below.
http://postgresql.1045698.n5.nabble.com/bug-in-fast-path-locking-td5626629.html

Not relevant to 9.0.x, for sure.

regards, tom lane