Postgres Pro
Downloads
Home   >   mailing lists

Re: PANIC in GIN code - Mailing list pgsql-hackers

From Jeff Janes
Subject Re: PANIC in GIN code
Date
Msg-id CAMkU=1yPy2B9Cz5az5AMcw1i+0++cYvnrpgcr3b2-j-XHM9dnA@mail.gmail.com
Whole thread Raw
In response to Re: PANIC in GIN code  (Heikki Linnakangas <hlinnaka@iki.fi>)
Responses Re: PANIC in GIN code
List pgsql-hackers
Tree view
On Mon, Jun 29, 2015 at 2:08 PM, Heikki Linnakangas <hlinnaka@iki.fi> wrote:
 

Now kill -9 postmaster, and restart. Voila, the page headers are all zeros:

postgres=# select * from page_header(get_raw_page('i_foo', 0));
    lsn    | checksum | flags | lower | upper | special | pagesize | version |
prune_xid
-----------+----------+-------+-------+-------+---------+----------+---------+-
----------
 0/1891270 |        0 |     0 |     0 |     0 |       0 |        0 |     0 |
        0
(1 row)

postgres=# select * from gin_metapage_info(get_raw_page('i_foo', 0));ERROR:  input page is not a GIN metapage
DETAIL:  Flags 0189, expected 0008

I just pushed a fix for this, but unfortunately it didn't make it 9.5alpha1.

Thanks.  I think that that fixed it.  It survived for over an hour this time.  Then it hit a different problem (apparent deadlock in the LWLock code, but that is probably a topic for another thread, I'm working on reproducing it.)

Cheers,

Jeff

pgsql-hackers by date:

Previous
From: Sawada Masahiko
Date:
Subject: Re: pg_file_settings view vs. Windows
Next
From: Peter Geoghegan
Date:
Subject: Bug in bttext_abbrev_convert()