From: Rural Hunter <ruralhunter@gmail.com>
Message-ID: <53D51E49.102@gmail.com>
Date: Sun, 27 Jul 2014 23:44:09 +0800
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64;
 rv:31.0) Gecko/20100101 Thunderbird/31.0
MIME-Version: 1.0
To: "pgsql-performance@postgresql.org" <pgsql-performance@postgresql.org>
Subject: Re: Very slow planning performance on partition table
References: <53CFB6BC.4050606@gmail.com> <25010.1406253203@sss.pgh.pa.us>
 <53D26875.2080604@gmail.com>
In-Reply-To: <53D26875.2080604@gmail.com>
Content-Type: text/plain; charset=gb18030; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Sender: pgsql-performance-owner@postgresql.org

Anyone? I can see many pg processes are in BIND status with htop. Some 
of them could be hanging like 30 mins. I tried gdb on the same process 
many times and the trace shows same as my previous post. This happened 
after I partitioned my main tables to 60 children tables. And also, I'm 
experiecing a cpu peak around 30-60 mins every 1-2 days. During the 
peak, all my cpus(32 cores) are full utilized while there is no special 
load and the memory and io are fine. Sometimes I had to kill the db 
process and restart the db to escape the situation. I tried to upgrade 
to the latest 9.2.9 but it didn't help.

在 2014/7/25 22:23, Rural Hunter 写道:
> I run dbg on the backend process and got this:
> (gdb) bt
> #0  0x00007fc4a1b6cdb7 in semop () from /lib/x86_64-linux-gnu/libc.so.6
> #1  0x00000000005f8703 in PGSemaphoreLock ()
> #2  0x0000000000636703 in LWLockAcquire ()
> #3  0x0000000000632eb3 in LockAcquireExtended ()
> #4  0x000000000062fdfb in LockRelationOid ()
> #5  0x0000000000474e55 in relation_open ()
> #6  0x000000000047b39b in index_open ()
> #7  0x00000000005f3c22 in get_relation_info ()
> #8  0x00000000005f6590 in build_simple_rel ()
> #9  0x00000000005f65db in build_simple_rel ()
> #10 0x00000000005de8c0 in add_base_rels_to_query ()
> #11 0x00000000005df352 in query_planner ()
> #12 0x00000000005e0d51 in grouping_planner ()
> #13 0x00000000005e2bbe in subquery_planner ()
> #14 0x00000000005e2ef9 in standard_planner ()
> #15 0x00000000006426e1 in pg_plan_query ()
> #16 0x000000000064279e in pg_plan_queries ()
> #17 0x00000000006f4b7a in BuildCachedPlan ()
> #18 0x00000000006f4e1e in GetCachedPlan ()
> #19 0x0000000000642259 in exec_bind_message ()
> #20 0x0000000000643561 in PostgresMain ()
> #21 0x000000000060347f in ServerLoop ()
> #22 0x0000000000604121 in PostmasterMain ()
> #23 0x00000000005a5ade in main ()
>
> Does that indicate something? seems it's waiting for some lock.
>


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