Re: Problem with locks

From: Alvaro Herrera <alvherre(at)commandprompt(dot)com>
To: Gregory Stark <stark(at)enterprisedb(dot)com>
Cc: PostgreSQL-development Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Problem with locks
Date: 2007-08-20 20:41:57
Message-ID: 20070820204156.GV27916@alvh.no-ip.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Gregory Stark wrote:
>
> I'm debugging this problem with stalled transactions waiting on locks which
> have already been granted and I'm puzzled by something. What is this PGPROC
> entry from?
>
> It's not a real process (pid is 0) and it's not garbage either (prev and next
> both point to a real PGPROC entry, namely MyProc which as you can see is
> waiting for this PGPROC).

Doesn't this look like your pointer arithmetic being borked? There are
too many values that look invalid.

> (gdb) p *(PGPROC*)(ShmemBase + MyProc->links.prev)
> $2 = {links = {prev = 2520941752, next = 2520941752}, sem = {semId = 1, semNum = 0}, waitStatus = 0, xid = 0, xmin = 0, pid = 0,
> databaseId = 0, roleId = 0, inCommit = 0 '\0', inVacuum = 0 '\0', isAutovacuum = 0 '\0', lwWaiting = 0 '\0', lwExclusive = 2 '\002',
> lwWaitLink = 0x0, waitLock = 0x2, waitProcLock = 0x0, waitLockMode = 0, heldLocks = 0, myProcLocks = {{prev = 0, next = 1}, {
> prev = 4294967296, next = 0}, {prev = 3028785127, next = 360654}, {prev = 73183493944770560, next = 137438953600}, {
> prev = 2520929504, next = 2520929072}, {prev = 2520940536, next = 2520940536}, {prev = 1, next = 0}, {prev = 0,
> next = 4294967296}, {prev = 4294967296, next = 0}, {prev = 2, next = 0}, {prev = 0, next = 0}, {prev = 1, next = 4294967296}, {
> prev = 0, next = 4000475635}, {prev = 360707, next = 73183493944770560}, {prev = 128, next = 2520933680}, {prev = 2520933680,
> next = 2520914616}}, subxids = {overflowed = -72 '�', nxids = 0, xids = {0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0,
> 0, 0, 1, 0, 0, 1, 0, 0, 1626942462, 0, 16384, 16430, 0, 16777216, 2, 0, 2520930944, 0, 2520930944, 0, 2520914784, 0, 2520914784,
> 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0}}}
> (gdb) p MyProc-ShmemBase
> $3 = (PGPROC *) 0xff9ac36e36e264b8
> (gdb) p (unsigned long)MyProc-ShmemBase
> $4 = 2520941752
>
>
> --
> Gregory Stark
> EnterpriseDB http://www.enterprisedb.com
>
> ---------------------------(end of broadcast)---------------------------
> TIP 4: Have you searched our list archives?
>
> http://archives.postgresql.org

--
Alvaro Herrera http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Heikki Linnakangas 2007-08-20 20:43:12 Re: Status of 8.3 patches
Previous Message Joshua D. Drake 2007-08-20 20:37:04 Re: Status of 8.3 patches