[pgpool-general: 3396] Re: v3.4.0.(3)? - memory issue and connection hangs

[Pablo Sanchez]

[ Comments below, in-line ]

On 01/02/2015 07:56 PM, Tatsuo Ishii wrote:

Hi Tatsuo,

Thank you for the speedy response.

> I thought we fixed memory leak issue in this commit (which is included
> in the RPM you installed).

Ah yes, the reason we pulled the latest RPM was because (I believe)
the memory leak mentioned.  As you probably know, the leak was far
worst.

> There may be another code path which trigger another pattern of memory
> leak. Let me invest this...

Thank you.  If there's anything I can provide please let me know.

> Ok, this definitely a different pattern of memory leak. Let me
> invest...

Again, thank you.

>> :: Hang on in-rush :::
>>
>> We've been using PGBench to stress test the environment.  One test in
>> particular is how our environment handles a sudden burst of
>> connections.
>
> Ok, I will try to reproduce the problem using your test case.

... and again:  thank you.

> In the mean time I would suggest increasing
> listen_backlog_multiplier might help.

Yes, we have increased this parameter to 107:

    listen_backlog_multiplier = 107

and I've set the respective kernel parameter to 10,240:

    # Default:  128
    net.core.somaxconn = 10240

Cheers,
--
Pablo Sanchez - Blueoak Database Engineering, Inc
Ph:    819.459.1926         Blog:  http://pablo-blog.blueoakdb.com
iNum:  883.5100.0990.1054