Re: Problem with PITR Past Particular WAL File

From: Craig McElroy <craig(dot)mcelroy(at)contegix(dot)com>
To: Teodor Sigaev <teodor(at)sigaev(dot)ru>
Cc: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, PostgreSQL Admin List <pgsql-admin(at)postgresql(dot)org>
Subject: Re: Problem with PITR Past Particular WAL File
Date: 2007-10-25 19:28:01
Message-ID: 4E5ADC27-53C9-4A63-8BE2-7563E2EC89EB@contegix.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin

Teodor:
Thanks for the followup. We will be upgrading this system to
8.2.5 in the next day or so and will test it on that anyways just to
be sure it isn't related. Let me know if there is anything else we
can provide to help track down the bug. We will be keeping this base
backup and WAL files around so that we can use them for testing.

Cheers,
-craig

---
Craig A. McElroy
Contegix
Beyond Managed Hosting(r) for Your Enterprise

On Oct 25, 2007, at 1:05 PM, Teodor Sigaev wrote:

>>> #0 0x080b8ee0 in entrySplitPage ()
>>> #1 0x080baccf in ginInsertValue ()
> Corrupted stack? gin_xlog_cleanup() doesn't call ginInsertValue()
> directly.
>
>>> #2 0x080b81b7 in gin_xlog_cleanup ()
>>> #3 0x080af4ce in StartupXLOG ()
>>> #4 0x080c04ca in BootstrapMain ()
>>> #5 0x08186b2f in StartChildProcess ()
>>> #6 0x081889eb in PostmasterMain ()
>>> #7 0x0814ee9e in main ()
>
>> Teodor, can you comment on whether this stack trace looks like it
>> could
>> be related to that fix?
> No, I suppose. That changes was about deadlock prevention,
> recognize splits and
> fixes in deletion of page in a posting tree. entrySplitPage() is a
> split of page of indexed values.
>
> Hmm. I suppose, I found a reason. Real backtrace should look like:
> gin_xlog_cleanup()
> ginContinueSplit()
> prepareEntryScan(,,,NULL) // makes btree.ginstate = NULL
> ginInsertValue()
> entrySplitPage() // tries to access
> // btree->ginstate->tupdesc in
> // 497 line
>
>
> That piece of code:
> value = index_getattr(leftrightmost, FirstOffsetNumber, btree-
> >ginstate->tupdesc, &isnull);
> btree->entry = GinFormTuple(btree->ginstate, value, NULL, 0);
> ItemPointerSet(&(btree->entry)->t_tid, BufferGetBlockNumber
> (lbuf), InvalidOffsetNumber);
>
> It just makes new tuple with the same value and another
> ItemPointer. Some later I'll make a patch which will not call
> index_getattr() and GinFormTuple() - now I know how make a test suite.
>
> --
> Teodor Sigaev E-mail:
> teodor(at)sigaev(dot)ru
> WWW: http://
> www.sigaev.ru/
>
> ---------------------------(end of
> broadcast)---------------------------
> TIP 6: explain analyze is your friend

In response to

Responses

Browse pgsql-admin by date

  From Date Subject
Next Message David Young 2007-10-26 05:52:06 32-bit to 64-bit migration screwup
Previous Message smiley2211 2007-10-25 18:31:46 Postgresql New login - password expiration???