Re: Memory leak (possibly connected to postgis) leading to server crash

From: Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>
To: Roman Cervenak <roman(at)cervenak(dot)info>
Cc: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, pgsql-bugs(at)lists(dot)postgresql(dot)org
Subject: Re: Memory leak (possibly connected to postgis) leading to server crash
Date: 2019-12-18 14:55:21
Message-ID: 20191218145521.oyhvjn5ptf3a6ej4@development
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

On Wed, Dec 18, 2019 at 09:49:22AM +0100, Roman Cervenak wrote:
>Sure, here it is, 10k queries generated by this tool:
>https://sygicpublic.blob.core.windows.net/postgres-bug-data/postgis_memory_leak_queries.txt
>

Thanks. I've tried running this on current branches (so PostgreSQL 13dev
and PostGIS 3.1.0dev), but that does not seem to be leaking so far. I'll
leave running it for a bit longer and then I'll try running it on 12.1
and 3.0.0 which I think are the versions you reported. I don't have an
Ubuntu box at hand, so I'll have to build it locally, though.

BTW it's customary to reply in-line here, top-posting makes it harder to
follow the discussion etc.

regards

>RC
>
>On Tue, Dec 17, 2019 at 7:10 PM Tomas Vondra <tomas(dot)vondra(at)2ndquadrant(dot)com>
>wrote:
>
>> On Tue, Dec 17, 2019 at 02:20:41PM +0100, Roman Cervenak wrote:
>> >.NET core is cross platform, you can run it under windows or linux, just
>> >install .net core runtime.
>>
>> That's possible, but I have zero experience with .net core, and
>> unfortunately I have no system with it installed. I've tried running it
>> under mono I just installed, but quickly ran into issue that I don't
>> know how to solve. I'm willing to spend time investigating the issue,
>> but this seems a bit too much ...
>>
>> >I did not try to run loop with fixed params, I tried to mimic my
>> production
>> >workload (which is not random of course, but highly variable by location).
>> >You can certainly try.
>> >
>>
>> Can you capture a couple of queries, and share those? I'll run that
>> under valgrind and see if that produces something interesting.
>>
>>
>> regards
>>
>> --
>> Tomas Vondra http://www.2ndQuadrant.com
>> PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
>>

--
Tomas Vondra http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Daniel Gustafsson 2019-12-18 15:15:29 Re: BUG #16171: Potential malformed JSON in explain output
Previous Message hubert depesz lubaczewski 2019-12-18 14:06:31 Re: Fwd: weird long time query