On 2013-12-10 11:12:03 -0800, Josh Berkus wrote:
> On 12/10/2013 10:48 AM, Andres Freund wrote:
> > On 2013-12-10 10:44:30 -0800, Josh Berkus wrote:
> >> On 12/10/2013 10:39 AM, Andres Freund wrote:
> >>> Hi,
> >>>
> >>> On 2013-12-10 10:38:32 -0800, Josh Berkus wrote:
> We've just run across
On 12/10/2013 10:48 AM, Andres Freund wrote:
> On 2013-12-10 10:44:30 -0800, Josh Berkus wrote:
>> On 12/10/2013 10:39 AM, Andres Freund wrote:
>>> Hi,
>>>
>>> On 2013-12-10 10:38:32 -0800, Josh Berkus wrote:
We've just run across a case of this exact issue on 9.2.4. I thought it
was sup
On 2013-12-10 10:44:30 -0800, Josh Berkus wrote:
> On 12/10/2013 10:39 AM, Andres Freund wrote:
> > Hi,
> >
> > On 2013-12-10 10:38:32 -0800, Josh Berkus wrote:
> >> We've just run across a case of this exact issue on 9.2.4. I thought it
> >> was supposed to be 9.3-only?
> >
> > Could you please
On 12/10/2013 10:39 AM, Andres Freund wrote:
> Hi,
>
> On 2013-12-10 10:38:32 -0800, Josh Berkus wrote:
>> We've just run across a case of this exact issue on 9.2.4. I thought it
>> was supposed to be 9.3-only?
>
> Could you please describe "this exact issue"?
Fatal errors due to missing pg_sub
Hi,
On 2013-12-10 10:38:32 -0800, Josh Berkus wrote:
> We've just run across a case of this exact issue on 9.2.4. I thought it
> was supposed to be 9.3-only?
Could you please describe "this exact issue"?
Greetings,
Andres Freund
--
Andres Freund http://www.2ndQuadrant.co
Andres, all:
We've just run across a case of this exact issue on 9.2.4. I thought it
was supposed to be 9.3-only?
--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www
On Thu, Nov 28, 2013 at 5:15 AM, Andres Freund wrote:
> Hi,
>
> Do you still have the core file around? If so could you 'p
> *ShmemVariableCache' and 'p *ControlFile'?
>
So sorry, I didn't see this message until just today. Seems it was
accidentally archived before hitting my eyeballs.
I see tha
Hi,
On 2013-11-24 16:56:26 -0500, J Smith wrote:
> coredumper worked like a charm. Useful tool, that is... although as a
> bit of advice, I'd try not to run it on Postgres if your various
> memory settings are tweaked towards production use -- the core dump
> that was captured on my server weighed
On 2013-11-27 13:57:52 -0300, Alvaro Herrera wrote:
> Per bug report by J Smith in
> cadfupgc5bmtv-yg9znxv-vcfkb+jprqs7m2oesqxam_4z1j...@mail.gmail.com
> diagnosed by Andres Freund.
Alvaro, do you see a way this could actually have caused J.'s problems?
I thought about a few, but each
Alvaro Herrera escribió:
> Andres Freund escribió:
> This seems simple to handle by adding the check you propose to the loop.
> Basically if the xmax doesn't match the xmin, we reached the end,
> there's nothing more to lock and we can return success without any
> further work:
As mentioned in th
On 2013-11-25 18:06:30 -0300, Alvaro Herrera wrote:
> > I mean that in the !KEYS_UPDATED case we don't need to abort if we're
> > only acquiring a key share...
>
> Hm, I think that's correct -- we don't need to abort. But we still need
> to wait until the updater completes. So this proposed patc
Andres Freund escribió:
> On 2013-11-25 17:10:39 -0300, Alvaro Herrera wrote:
> > Let me point out that this is exactly the same code that would be
> > affected by my proposed fix for #8434, which would have this check the
> > updateXid in all cases, not only in KEYS_UPDATED as currently.
>
> Hm.
On 2013-11-25 17:10:39 -0300, Alvaro Herrera wrote:
> > I am not sure whether that's the origin of the problem but at the very
> > least it seems to me that heap_lock_updated_tuple_rec() is missing
> > several important pieces:
> > a) do the priorXmax==xmin dance to check we're still following the
Andres Freund escribió:
> Ok, this is helpful. Do you rather longrunning transactions? The
> transaction that does foreign key checks has an xid of 10260613, while
> the row that's getting checked has 13514992.
Thanks for the analysis.
> #4 0x00635dc7 in XactLockTableWait (xid=13514992)
On Mon, Nov 25, 2013 at 11:46 AM, Alvaro Herrera
wrote:
> J Smith escribió:
>
>> We did have some long-running transactions, yes. We refactored a bit
>> and removed them and the problem ceased on our end. We ended up
>> reverting our changes for the sake of running this experiment over the
>> week
J Smith escribió:
> We did have some long-running transactions, yes. We refactored a bit
> and removed them and the problem ceased on our end. We ended up
> reverting our changes for the sake of running this experiment over the
> weekend and the errors returned. We've since restored our fix and
>
On Mon, Nov 25, 2013 at 6:47 AM, Andres Freund wrote:
> Hi,
>
> On 2013-11-24 16:56:26 -0500, J Smith wrote:
>
>> Nov 23 14:38:32 dev postgres[23810]: [4-1] user=dev,db=dev ERROR: could not
>> access status of transaction 13514992
>> Nov 23 14:38:32 dev postgres[23810]: [4-2] user=dev,db=dev DET
Hi,
On 2013-11-24 16:56:26 -0500, J Smith wrote:
> coredumper worked like a charm. Useful tool, that is... although as a
> bit of advice, I'd try not to run it on Postgres if your various
> memory settings are tweaked towards production use -- the core dump
> that was captured on my server weighed
coredumper worked like a charm. Useful tool, that is... although as a
bit of advice, I'd try not to run it on Postgres if your various
memory settings are tweaked towards production use -- the core dump
that was captured on my server weighed in at 16 GB.
Anyways, I've attached both the log entries
On Tue, Nov 19, 2013 at 10:16 AM, J Smith wrote:
> Alright, we'll look into doing that heading into the weekend.
> Interestingly, we haven't experienced the issue since our main Java
> developer made some modifications to our backend system. I'm not
> entirely sure what the changes entail except t
Alright, we'll look into doing that heading into the weekend.
Interestingly, we haven't experienced the issue since our main Java
developer made some modifications to our backend system. I'm not
entirely sure what the changes entail except that it's a one-liner
that involves re-SELECTing a table du
On Fri, Nov 15, 2013 at 4:01 PM, J Smith wrote:
> On Fri, Nov 15, 2013 at 3:21 PM, Robert Haas wrote:
>> I think what would help the most is if you could arrange to obtain a
>> stack backtrace at the point when the error is thrown. Maybe put a
>> long sleep call in just before the error happens,
On Fri, Nov 15, 2013 at 3:21 PM, Robert Haas wrote:
>
> I think what would help the most is if you could arrange to obtain a
> stack backtrace at the point when the error is thrown. Maybe put a
> long sleep call in just before the error happens, and when it gets
> stuck there, attach gdb and run
On Wed, Nov 13, 2013 at 12:29 PM, J Smith wrote:
> Looks like we got another set of errors overnight. Here's the log file
> from the errors. (Log file scrubbed slightly to remove private data,
> but still representative of the problem I believe.)
>
> Nov 13 05:34:34 dev postgres[6084]: [4-1] user=
Looks like we got another set of errors overnight. Here's the log file
from the errors. (Log file scrubbed slightly to remove private data,
but still representative of the problem I believe.)
Nov 13 05:34:34 dev postgres[6084]: [4-1] user=dev,db=dev ERROR:
could not access status of transaction 63
On Tue, Nov 12, 2013 at 11:55 AM, Andres Freund wrote:
> Hi,
>
> On 2013-11-12 11:46:19 -0500, J Smith wrote:
>> > * Does SELECT count(*) FROM pg_prepared_xacts; return 0?
>>
>> Yes it does.
>
> Could you show the output? Do you actually use prepared xacts actively?
jay:dev@jagger=# select * from
On Tue, Nov 12, 2013 at 11:54 AM, Stephen Frost wrote:
>
> Did you also upgrade to PostGIS 2.x as part of this..? Seems like it'd
> be unrelated, but one never knows. Any chance you could distill this
> down into a small test case which exhibits the problem? I'm guessing
> 'no', but figured I'd
Hi,
On 2013-11-12 11:46:19 -0500, J Smith wrote:
> > * Does SELECT count(*) FROM pg_prepared_xacts; return 0?
>
> Yes it does.
Could you show the output? Do you actually use prepared xacts actively?
Do you actively use row level locking? Is there high concurrency in that
environment? In short,
* J Smith (dark.panda+li...@gmail.com) wrote:
> We haven't been able to use pg_upgrade as we rely heavily on PostGIS
> and do hard upgrades via pg_dump and postgis_restore.pl when we
> upgrade.
Did you also upgrade to PostGIS 2.x as part of this..? Seems like it'd
be unrelated, but one never know
* Andres Freund (and...@2ndquadrant.com) wrote:
> He referred to using pg_dumpall/pg_dump. But that bug was erroring out
> on pg_clog, not pg_subtrans, right?
Yeah, that was pg_clog. Obviously responded before really looking at
it. :)
> My gut feeling is thats it's related to foreign key locks d
G'day Andres.
On Tue, Nov 12, 2013 at 11:13 AM, Andres Freund wrote:
> Hi,
>
> On 2013-11-12 10:56:55 -0500, J Smith wrote:
>> G'day list. Didn't get any interest in pgsql-general, thought I'd try
>> my luck here, which perhaps would be more fitting in case I've
>> stumbled upon an edge case issu
On Tue, Nov 12, 2013 at 11:25 AM, Stephen Frost wrote:
>
> How was this upgrade done? If you used pg_upgrade, what version of the
> pg_upgrade code did you use? As I recall, there was a bug in older
> versions which could exhibit in this way..
>
> http://wiki.postgresql.org/wiki/20110408pg_upgra
On 2013-11-12 11:25:03 -0500, Stephen Frost wrote:
> * J Smith (dark.panda+li...@gmail.com) wrote:
> > I've recently upgraded a number of servers from PostgreSQL 9.2.5 to
> > 9.3.1 and have started getting the following errors every couple of
> > hours along with some failed transactions.
>
> How
* J Smith (dark.panda+li...@gmail.com) wrote:
> I've recently upgraded a number of servers from PostgreSQL 9.2.5 to
> 9.3.1 and have started getting the following errors every couple of
> hours along with some failed transactions.
How was this upgrade done? If you used pg_upgrade, what version of
Hi,
On 2013-11-12 10:56:55 -0500, J Smith wrote:
> G'day list. Didn't get any interest in pgsql-general, thought I'd try
> my luck here, which perhaps would be more fitting in case I've
> stumbled upon an edge case issue or something...
Normally the bug report for/the -bugs mailing list is the ri
35 matches
Mail list logo
