BUG #2712: could not fsync segment: Permission denied

From: "Thomas H" <me(at)alternize(dot)com>
To: pgsql-bugs(at)postgresql(dot)org
Subject: BUG #2712: could not fsync segment: Permission denied
Date: 2006-10-22 23:28:32
Message-ID: 200610222328.k9MNSWTP030133@wwwmaster.postgresql.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs pgsql-patches


The following bug has been logged online:

Bug reference: 2712
Logged by: Thomas H
Email address: me(at)alternize(dot)com
PostgreSQL version: 8.2b1
Operating system: windows 2003 standard
Description: could not fsync segment: Permission denied
Details:

sometimes we're seeing loads of errors in the log:

2006-10-22 23:48:50 LOG: could not fsync segment 0 of relation
1663/3964774/6409340: Permission denied
2006-10-22 23:48:50 ERROR: storage sync failed on magnetic disk: Permission
denied
2006-10-22 23:48:51 LOG: could not fsync segment 0 of relation
1663/3964774/6409340: Permission denied
2006-10-22 23:48:51 ERROR: storage sync failed on magnetic disk: Permission
denied
2006-10-22 23:48:52 LOG: could not fsync segment 0 of relation
1663/3964774/6409340: Permission denied
2006-10-22 23:48:52 ERROR: storage sync failed on magnetic disk: Permission
denied
2006-10-22 23:48:53 LOG: could not fsync segment 0 of relation
1663/3964774/6409340: Permission denied
2006-10-22 23:48:53 ERROR: storage sync failed on magnetic disk: Permission
denied
{...}

when this happens, there are also files locked within the data\base\{dbid}\.
access to those files are denied by the os - the files vanish as soon as
postmaster ist stopped & restarted.

i haven't yet found a possible reason - i suspect the error to appear
*sometimes* after issuing a "VACUUM FULL ANALYZE {tablename}" / "REINDEX
TABLE {tablename}".

the hardware is checked and ok.

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message Thomas H. 2006-10-23 01:31:50 Re: BUG #2712: could not fsync segment: Permission denied
Previous Message Yang Zhang 2006-10-22 22:54:49 BUG #2711: PGUSER affects "pg_ctl -w start" behavior

Browse pgsql-patches by date

  From Date Subject
Next Message Hitoshi Harada 2006-10-23 01:02:49 Re: [PATCHES] smartvacuum() instead of autovacuum
Previous Message Tom Lane 2006-10-22 17:52:34 Re: pg_buffercache tidyup