<p><br /> On Sep 23, 2011 5:59 PM, "Alvaro Herrera" <<a
href="mailto:alvherre@commandprompt.com">alvherre@commandprompt.com</a>>wrote:<br /> ><br /> ><br /> >
Excerptsfrom Linas Virbalas's message of vie sep 23 09:47:20 -0300 2011:<br /> > > On 9/23/11 12:05 PM, "Heikki
Linnakangas"<br/> > > <<a
href="mailto:heikki.linnakangas@enterprisedb.com">heikki.linnakangas@enterprisedb.com</a>>wrote:<br /> ><br />
>> But on the standby its size is the old one (thus, it seems, that the size<br /> > > changed after the
rsynctransfer and before the pg_stop_backup() was<br /> > > called):<br /> > ><br /> > > ls -l
pg_clog/<br/> > > total 8<br /> > > -rw------- 1 postgres postgres 8192 Sep 23 14:31 0000<br /> ><br />
>Sounds like rsync is caching the file size at the start of the run, and<br /> > then copying that many bytes,
ignoringthe growth that occurred after it<br /> > started.<br /> ><p>That pretty much matches what Daniel does
whenhe got the same failure case. <p>Is this not allowed? Shouldn't wal replay fix this? <p>/Magnus