I have two questions about the stats collector during recovery.
Why does the stats collector need to wait for consistent recovery
mode? The activity statistics which bgwriter may send before
reaching the mode should be ignored?
Why doesn't ServerLoop() try to restart the stats collector
when it's not in progress?
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center
In response to
pgsql-hackers by date
|Next:||From: Kevin Grittner||Date: 2009-02-25 23:21:20|
|Subject: effective_cache_size less than shared_buffers|
|Previous:||From: Bruce Momjian||Date: 2009-02-25 22:50:07|
|Subject: Re: Synchronous replication & Hot standby patches|
pgsql-committers by date
|Next:||From: User Fxjr||Date: 2009-02-26 02:06:58|
|Subject: npgsql - Npgsql2: Really fixed culture dependent representation of |
|Previous:||From: Tom Lane||Date: 2009-02-25 18:00:22|
|Subject: pgsql: Fix an old problem in decompilation of CASE constructs: the |