Re: amcheck assert failure - Mailing list pgsql-bugs

From Peter Geoghegan
Subject Re: amcheck assert failure
Date
Msg-id CAH2-Wz=r2g0pet4hVPi88Yh+uqRmMphEON0U9Yd28HSW+ac7vQ@mail.gmail.com
Whole thread Raw
In response to Re: amcheck assert failure  (Peter Geoghegan <pg@bowt.ie>)
Responses Re: amcheck assert failure  (Peter Geoghegan <pg@bowt.ie>)
List pgsql-bugs
On Sat, Apr 20, 2019 at 2:00 PM Peter Geoghegan <pg@bowt.ie> wrote:
> Obviously you'll need to figure out which block number the problem
> page is located in before you do this, but that shouldn't be hard --
> get a coredump and inspect the state from amcheck a few frames down.

Alternatively, you could dump the page image from within gdb. Go to
the frame for _bt_check_natts() (e.g., enter "f 1"), and then dump the
contents of the page to file:

(gdb) dump binary memory /tmp/dump_block.page page (page + 8192)

-- 
Peter Geoghegan



pgsql-bugs by date:

Previous
From: Peter Geoghegan
Date:
Subject: Re: amcheck assert failure
Next
From: Peter Geoghegan
Date:
Subject: Re: amcheck assert failure