Skip site navigation (1) Skip section navigation (2)

Less available diskspace after crashed CLUSTER

From: "Sander Steffann" <sander(at)steffann(dot)nl>
To: <pgsql-admin(at)postgresql(dot)org>
Subject: Less available diskspace after crashed CLUSTER
Date: 2005-01-15 15:13:05
Message-ID: 003201c4fb14$b7377840$64c8a8c0@balefirehome (view raw or flat)
Thread:
Lists: pgsql-admin
Hi,

I was running a CLUSTER on a big database (approx. 8G) and I didn't 
anticipate the diskspace usage, so I ran out of diskspace. This is on 
PostgreSQL 7.4.6. The backend crashed with the following log messages:

2005-01-15 15:17:46 [30605] PANIC:  PANIC:  could not write to file 
"/var/lib/pgsql/data/pg_xlog/xlogtemp.30605": Geen ruimte over op apparaat
STATEMENT:  CLUSTER;
2005-01-15 15:17:46 [11245] LOG:  server process (PID 30605) was terminated 
by signal 6
2005-01-15 15:17:46 [11245] LOG:  terminating any other active server 
processes
2005-01-15 15:17:46 [3378] WARNING:  terminating connection because of crash 
of another server process
DETAIL:  The postmaster has commanded this server process to roll back the 
current transaction and exit, because another server process exited 
abnormally and possibly corrupted shared memory.
HINT:  In a moment you should be able to reconnect to the database and 
repeat your command.
<... a lot more of the same messages from other backends ...>
2005-01-15 15:17:46 [11245] LOG:  all server processes terminated; 
reinitializing
2005-01-15 15:17:46 [3389] LOG:  database system was interrupted at 
2005-01-15 15:17:44 CET
2005-01-15 15:17:46 [3389] LOG:  checkpoint record is at 12/9E2976B0
2005-01-15 15:17:46 [3389] LOG:  redo record is at 12/9E00A790; undo record 
is at 0/0; shutdown FALSE
2005-01-15 15:17:46 [3389] LOG:  next transaction ID: 13482165; next OID: 
1694415
2005-01-15 15:17:46 [3389] LOG:  database system was not properly shut down; 
automatic recovery in progress
2005-01-15 15:17:46 [3389] LOG:  redo starts at 12/9E00A790
2005-01-15 15:18:03 [3389] LOG:  could not open file 
"/var/lib/pgsql/data/pg_xlog/00000012000000BF" (log file 18, segment 191): 
Onbekend bestand of map
2005-01-15 15:18:03 [3389] LOG:  redo done at 12/BEFFD810
2005-01-15 15:18:13 [3389] LOG:  recycled transaction log file 
"000000120000009B"
2005-01-15 15:18:13 [3389] LOG:  recycled transaction log file 
"000000120000009C"
2005-01-15 15:18:13 [3389] LOG:  recycled transaction log file 
"000000120000009D"
2005-01-15 15:18:13 [3389] LOG:  database system is ready

Translations:
 Geen ruimte over op apparaat -> No space left on device
 Onbekend bestand of map -> Unknown file or directory

The database is running nice again, but I have the feeling I lost some 
diskspace... I suspected a temp file, so I stopped the postmaster, looked in 
/var/lib/pgsql/data/base/17142/pgsql_tmp, found one file and deleted it. But 
now I still have 1G less available diskspace than before the CLUSTER.

Where should I look next? I would like the diskspace back :-)

Thanks,
Sander.



Responses

pgsql-admin by date

Next:From: Tom LaneDate: 2005-01-15 18:01:17
Subject: Re: Less available diskspace after crashed CLUSTER
Previous:From: Bradley KieserDate: 2005-01-15 10:41:40
Subject: Re: Script terminates even though ON_ERROR_STOP not set

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group