Re: Streaming replication on 9.1-beta2 after pg_restore is very slow - Mailing list pgsql-general

From Tom Lane
Subject Re: Streaming replication on 9.1-beta2 after pg_restore is very slow
Date
Msg-id 14178.1310051645@sss.pgh.pa.us
Whole thread Raw
In response to Re: Streaming replication on 9.1-beta2 after pg_restore is very slow  (David Hartveld <David.Hartveld@mendix.com>)
Responses Re: Streaming replication on 9.1-beta2 after pg_restore is very slow
List pgsql-general
Looking back, I notice that you built with gcc 4.6.0.  At least on Red
Hat machines, that gcc has a rather nasty optimization bug that breaks
WAL replay, with symptoms that seem to match what you have here ---
namely, the replay process quits and has to be restarted every few pages.
I'm betting Debian hasn't fixed that bug yet either and so you need this
post-beta2 patch:

http://git.postgresql.org/gitweb/?p=postgresql.git&a=commitdiff&h=c2ba0121c73b7461331104a46d140156e847572a

            regards, tom lane

pgsql-general by date:

Previous
From: Simon Riggs
Date:
Subject: Re: Streaming replication on 9.1-beta2 after pg_restore is very slow
Next
From: Karsten Hilbert
Date:
Subject: Re: Streaming replication on 9.1-beta2 after pg_restore is very slow