8.2.9
I have a hunch what went wrong. We tested a warm standby failover
while the 'master' was still running - our warm standby includes a
process that grabs the final unarchived WAL segment from the master
pg_xlog directory. I have a hunch that this is where the disconnect
happened, I think the rule needs to be that we only execute a failover
based on bringing the master down as opposed to simply creating the
trigger file. I'm trying to test this theory out now - I'll update
the list with my findings.
/Kevin
On Jul 8, 2008, at 10:52 AM, Alvaro Herrera wrote:
kevin kempter wrote:
We suspected the index was corrupt but a reindex fails for obvious
reasons.
What version is this? We're investigating a similar problem on
8.1.11.
--
Alvaro Herrera http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support
