My dear Tom Lane && pgsql-hackers
Thanks very very much! I have modified the postgres source code that Tom lane fixed, and the origin phenomenon of that bug no longer appears in my old test case. Thanks again! Best Regards! at 2013-07-24 19:08, Andres Freund wrote: Hi, On 2013-07-24 10:54:07 +0000, hailong...@qunar.com<mailto:hailong...@qunar.com> wrote: The following bug has been logged on the website: Bug reference: 8327 Logged by: hailong.li Email address: hailong...@qunar.com<mailto:hailong...@qunar.com> PostgreSQL version: 9.2.4 Operating system: CentOS release 5.5 (Final) Description: http://www.postgresql.org/message-id/519a5917.40...@qunar.com There's a thread that discussed the issue on the -hackers list: http://archives.postgresql.org/message-id/5829.1370555205%40sss.pgh.pa.us This resulted in a fix: http://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=e472b921406407794bab911c64655b8b82375196 Which has also been backported to the 9.2 branch. There hasn't been a release since the commit though. Greetings, Andres Freund
