Re: [COMMITTERS] pgsql: Rework subtransaction commit protocol for hot standby. - Mailing list pgsql-hackers

From Simon Riggs
Subject Re: [COMMITTERS] pgsql: Rework subtransaction commit protocol for hot standby.
Date
Msg-id 1224698262.27145.390.camel@ebony.2ndQuadrant
Whole thread Raw
In response to Re: [COMMITTERS] pgsql: Rework subtransaction commit protocol for hot standby.  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On Wed, 2008-10-22 at 13:25 -0400, Tom Lane wrote:
> conditionName=0x6deb8 "!(((*byteptr >> bshift) & ((1 << 2) - 1)) == 0
> || ((*byteptr >> bshift) & ((1 << 2) - 1)) == 0x03 || ((*byteptr >>
> bshift) & ((1 << 2) - 1)) == status)", errorType=0x6dde4
> "FailedAssertion", 
>     fileName=0x6ddf4 "clog.c", lineNumber=330) at assert.c:57

This implies we are setting commit on an already aborted xid.

Either we are incorrectly setting some commits as aborts, or more likely
that we are updating the wrong page with an xid.

I'm continuing to look.

-- Simon Riggs           www.2ndQuadrant.comPostgreSQL Training, Services and Support



pgsql-hackers by date:

Previous
From: "Dave Gudeman"
Date:
Subject: Re: PGDay.it collation discussion notes
Next
From: Simon Riggs
Date:
Subject: Re: [COMMITTERS] pgsql: Rework subtransaction commit protocol for hot standby.