Date: Tue, 15 Dec 2015 20:34:28 +0100
From: =?ISO-8859-2?Q?Micha=B3_Nowak?= <minowack@wp.pl>
To: pgsql-admin@postgresql.org
Subject: semop hanging - Postgres 9.4.4
Message-ID: <56706b44002bc1.54511812@wp.pl>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-2
Content-Transfer-Encoding: 8bit
Content-Disposition: inline
Organization: Poznaj Poczte WP http://poczta.wp.pl/info-start.html
Precedence: bulk
Sender: pgsql-admin-owner@postgresql.org

Hello,

Since couple of days on my Postgres cluster i've seen few times a hanged process. It looks like one of postgres workers stuck on simple INSERT query.

  PID USERNAME PRI NICE  SIZE   RES STATE   TIME   WCPU    CPU COMMAND
12173 postgres  20    0 1372M  345M run   166:18 24.47% 99.90% postgres: db01 db01 10.0.0.101(58713) INSERT


It utilizes 100% of one CPU and after 1-2 hours it starts to utilize memory. After another 1-2 hours OOM-Killer have to take action. At the same time other postgres workers are properly processing queries. 

There's no error on syslog/dmesg. Strace:


semop(58261738, {{12, -1, 0}}, 1)       = 0
semop(58261738, {{12, -1, 0}}, 1)       = 0
semop(58261738, {{12, -1, 0}}, 1)       = 0
semop(58851580, {{0, 1, 0}}, 1)         = 0
semop(58261738, {{12, -1, 0}}, 1)       = 0
semop(58261738, {{12, -1, 0}}, 1)       = 0
semop(58261738, {{12, -1, 0}}, 1)       = 0
semop(57802972, {{1, 1, 0}}, 1)         = 0
select(0, NULL, NULL, NULL, {0, 1000})  = 0 (Timeout)
semop(58261738, {{12, -1, 0}}, 1)       = 0
semop(58261738, {{12, -1, 0}}, 1)       = 0
semop(58261738, {{12, -1, 0}}, 1)       = 0


Is this related to incorrect semaphores configuration?


Here's my current configuration:

sql01:~# ipcs -l

------ Shared Memory Limits --------
max number of segments = 4096
max seg size (kbytes) = 18014398509465599
max total shared memory (kbytes) = 18446744073642442748
min seg size (bytes) = 1

------ Semaphore Limits --------
max number of arrays = 2048
max semaphores per array = 250
max semaphores system wide = 512000
max ops per semop call = 100
semaphore max value = 32767

------ Messages Limits --------
max queues system wide = 32000
max size of message (bytes) = 8192
default max size of queue (bytes) = 16384


sql01:~# ipcs -u

------ Shared Memory Status --------
segments allocated 1
pages allocated 1
pages resident  1
pages swapped   0
Swap performance: 0 attempts     0 successes

------ Semaphore Status --------
used arrays = 257
allocated semaphores = 4369

------ Messages Status --------
allocated queues = 0
used headers = 0
used space = 0 bytes


Some values from postgresql.conf:
max_connections = 4096
effective_cache_size = 12GB
work_mem = 16MB
shared_buffers = 1GB


postgres=# select version();
                                               version
------------------------------------------------------------------------------------------------------
 PostgreSQL 9.4.4 on x86_64-unknown-linux-gnu, compiled by gcc (Ubuntu 4.8.2-19ubuntu1) 4.8.2, 64-bit


I would appreciate any help.

Best Regards,

Marcin


-- 
Sent via pgsql-admin mailing list (pgsql-admin@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin