Re: (Never?) Kill Postmaster?

From: Bill Moran <wmoran(at)potentialtech(dot)com>
To: Stefan Schwarzer <stefan(dot)schwarzer(at)grid(dot)unep(dot)ch>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: (Never?) Kill Postmaster?
Date: 2007-10-24 12:05:47
Message-ID: 20071024080547.3745a3fb.wmoran@potentialtech.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

In response to Stefan Schwarzer <stefan(dot)schwarzer(at)grid(dot)unep(dot)ch>:

> Hi there,
>
> I read dozens of times the "TIP 2: Don't 'kill -9' the postmaster"...
>
> Now, what am I supposed to do if I launched a query which takes ages,
> and which I want to interrupt?
>
> Thanks for any advice,

Tracing through the other threads, this is obviously not the regular
wayward query, but one that won't die by the normal methods.

Unfortunately, I came across this recently, and the only solution I
found was to do a "pg_ctl restart -m i" (Yes, I tried -m f first).

Luckily, the db in question was such that the front ends didn't
suffer horribly from this and reconnected, and that the database
finished up its recovery in a timely manner.

Hopefully, I can generate a reproducible example so I can file a
bug, but haven't gotten that far with it yet.

--
Bill Moran
http://www.potentialtech.com

In response to

Browse pgsql-general by date

  From Date Subject
Next Message luca.ciciriello 2007-10-24 13:06:03 using libpq.lib in Microsoft C++ (managed)
Previous Message Tom Lane 2007-10-24 12:01:13 Re: (Never?) Kill Postmaster?