Re: BUG #7500: hot-standby replica crash after an initial rsync - Mailing list pgsql-bugs

From Andres Freund
Subject Re: BUG #7500: hot-standby replica crash after an initial rsync
Date
Msg-id 201208212208.06042.andres@2ndquadrant.com
Whole thread Raw
In response to Re: BUG #7500: hot-standby replica crash after an initial rsync  (Maxim Boguk <maxim.boguk@gmail.com>)
Responses Re: BUG #7500: hot-standby replica crash after an initial rsync  (Maxim Boguk <maxim.boguk@gmail.com>)
List pgsql-bugs
On Tuesday, August 21, 2012 03:30:44 PM Maxim Boguk wrote:
> Hi Andres,
>
> > > I have some problems with debug startup process with gdb...
> >
> > > I following next sequence of commands (and got no useful results):
> > Youre debugging the postmaster that way. The easiest way would be to just
> > attach to the startup process with gdb -p. Not sure if you can manage
> > that timingwise without manually adding sleeps.
>
> It's imposible without added sleep seems. Too fast for me.
> Where I should add sleep call in PostgreSQL source? (process managment part
> of PostgreSQL isn't known for me at all).
I would suggest adding it to StartupXLOG() in xlog.c just before the /* REDO
*/ comment or similar.

I would add something akin to

elog(WARNING, "pid of startup is: %d, sleeping for 10s", getpid());
sleep(10);

Greetings,

Andres
--
 Andres Freund                       http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services

pgsql-bugs by date:

Previous
From: Alvaro Herrera
Date:
Subject: Re: BUG #4958: Stats collector hung on WaitForMultipleObjectsEx while attempting to recv a datagram
Next
From: Tom Lane
Date:
Subject: Re: BUG #4958: Stats collector hung on WaitForMultipleObjectsEx while attempting to recv a datagram