On 2014-02-24 22:17:31 +0100, Andres Freund wrote:
> Those together explain the story. Note this bit:
>
> static void
> heap_xlog_lock(XLogRecPtr lsn, XLogRecord *record)
> {
> ...
> HeapTupleHeaderClearHotUpdated(htup);
> HeapTupleHeaderSetXmax(htup, xlrec->locking_xid);
> HeapTupleHeaderSetCmax(htup, FirstCommandId, false);
> /* Make sure there is no forward chain link in t_ctid */
> htup->t_ctid = xlrec->target.tid;
> ...
> }
>
> So, the replay of FD/2F0AE4D0 breaks the ctid chain *and* unsets the
> HOT_UPDATED flag.
Some quick archeology shows that the HeapTupleHeaderClearHotUpdated()
was in the original HOT commit (282d2a03d) and clearing of t_ctid was in
the original commit implementing FOR SHARE (bedb78d38). Both look like
they are copied from other places, I don't immediately see any need for
them here...
Greetings,
Andres Freund
-- Andres Freund http://www.2ndQuadrant.com/PostgreSQL Development, 24x7 Support, Training &
Services
