This is a discussion on postgres 8.1 crashing within the pgsql Bugs forums, part of the PostgreSQL category; --> my previous post just vanished... Hi! I just upgraded pg 8.0.3 to 8.1 (dump and reload) and outcome was ...
| |||||||
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| my previous post just vanished... Hi! I just upgraded pg 8.0.3 to 8.1 (dump and reload) and outcome was a corrupted database. pg 8.1 seemed to fly but backends were crashing :-/ While tailing the log file I noticed a couple of: ERROR: invalid memory alloc request size 1101389186 Hardware is Sun v40z with 64bit Debian Sarge, Linux 2.6.13.2. Size of database is somewhere around 500M rows and it's loaded by a website with a lot of traffic. Configuration: http://tuner.bdb.fi/postgresql.conf Some backtraces (update queries are included in the bt): http://tuner.bdb.fi/pg_crashing.txt http://tuner.bdb.fi/pg_crashing2.txt http://tuner.bdb.fi/pg_crashing3.txt And users-table which is involved in all the crashes is described here (contains ~700000 rows): http://tuner.bdb.fi/pg_crashing_table.txt I tried to reindex the table but it didn't help. PG went corrupted after a while: http://tuner.bdb.fi/pg_crashing4.txt Any ideas? |\__/| ( oo ) Kari Lavikka - tuner@bdb.fi - (050) 380 3808 __ooO( )Ooo_______ _____ ___ _ _ _ _ _ _ _ "" ---------------------------(end of broadcast)--------------------------- TIP 2: Don't 'kill -9' the postmaster |
| |||
| Kari Lavikka <tuner@bdb.fi> writes: > Some backtraces (update queries are included in the bt): > http://tuner.bdb.fi/pg_crashing.txt > http://tuner.bdb.fi/pg_crashing2.txt > http://tuner.bdb.fi/pg_crashing3.txt This looks to be the toast-vs-index bug already identified. Patch is here: http://archives.postgresql.org/pgsql...1/msg00439.php regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 3: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faq |
| |||
| Tom Lane wrote: > Kari Lavikka <tuner@bdb.fi> writes: > > Some backtraces (update queries are included in the bt): > > http://tuner.bdb.fi/pg_crashing.txt > > http://tuner.bdb.fi/pg_crashing2.txt > > http://tuner.bdb.fi/pg_crashing3.txt > > This looks to be the toast-vs-index bug already identified. > Patch is here: > http://archives.postgresql.org/pgsql...1/msg00439.php Should we start planning for an 8.1.1? -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania 19073 ---------------------------(end of broadcast)--------------------------- TIP 4: Have you searched our list archives? http://archives.postgresql.org |
| ||||
| Bruce Momjian <pgman@candle.pha.pa.us> writes: > Should we start planning for an 8.1.1? Yeah, it's getting to be that time. regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match |