[Pgpool-general] Replication conncetions failing; help with debugging?

[Tatsuo Ishii]

> >> On the client:
> >> Connection for session 0x4bf5867000003b70 failed: server closed the
> >> connection unexpectedly
> >>          This probably means the server terminated abnormally
> >>          before or while processing the request.
> >
> > I'm not sure what kind of client you are using but did it happend at
> > the same time below?
> 
> I can't establish precise to-the-second timing.
> 
> >> 2010-08-27 21:00:16 LOG:   pid 30231: ProcessFrontendResponse: failed to
> >> read kind from frontend. frontend abnormally exited
> >
> > This message indicates that your client disconnected the connection to
> > pgpool, at least from the pgpool's point of view.
> 
> It's pgreplay on the front-end, which uses libpq and the asynchronous 
> protocol.  I'd just assume that async connections were incompatible with 
> pgpool,

pgbench uses "asynchronous protocol". BTW there's no such
"asynchronous" protocol in PostgreSQL. So from the pgpool's point of
view, there's no difference.

>  execept that the same client worked in load-balancing, 
> master-slave mode without errors.  It's just in pgpool replication mode 
> that it fails.
> 
> There also seems to be a variable amount of time before the error is 
> encountered; anywhere from 1 and 60 seconds (and since this is pgreplay, 
> it's running the exact same queries each time).  So it's not on the 
> first statement.
> 
> What does "detect error: kind: S" mean?

It's normal. pgpool just finds that backend returns "S"(parameter
status).
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp