[Pgpool-general] one of backends kicked after "VACUUM ANALYZE"

[Tatsuo Ishii]

>>> Putting the question the other way: is REINDEX TABLE safe, if made
>>> directly on database backends, without going through pgpool? As I
>>> understand, it does not change database data visible to the clients,
>>> so should be safe.
>>
>> Yes, doing REINDEX directly is safe.
> 
> However, if:
> 
> - we do REINDEX TABLE locally (i.e. not going through pgpool),
> 
> - the database is heavily used, and above reindexing contributes to
> - "deadlock detected".
> 
> 
> It means pgpool may still hit "deadlock detected" only on one of the
> backends. And kick one of the backends.
> 
> In other words, I think reindexing may be not safe when pgpool is
> used; no matter if REINDEX commands are issued through pgpool or
> not. The more busy the database is, the more likely it is to happen
> (of course will depend on usage patterns, database construction etc.).
> 
> Please correct me if I'm wrong.

I thought REINDEX is not performed under heavy load because it holds
strong lock.
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp