Re: streaming replication and archive_status

2017-04-17 17:08 GMT-03:00 Jeff Janes <jeff(dot)janes(at)gmail(dot)com>:

> On Mon, Apr 17, 2017 at 12:22 PM, Luciano Mittmann <mittmann(at)gmail(dot)com>
> wrote:
>
>> Hi All,
>>
>> anyone knows why there are so many files in the directory
>> pg_xlog/archive_status/ in replication server?
>>
>> # pg_xlog/archive_status/ | wc -l
>>
>> 75217
>>
>> Is possible to clean this .done files or just don't need to worry ?
>>
>> It's not occurs on primary or standalone servers, just on replication.
>>
>
> What version? Are you logging checkpoints, and if so what do they say?
>
>
On Mon, Apr 17, 2017 at 1:24 PM, Luciano Mittmann <mittmann(at)gmail(dot)com>
wrote:

> Hi Jeff,
>
> checkpoint message on standby node:
>
> [ 2017-04-17 17:21:56 BRT] @ LOG: restartpoint complete: wrote 21475
> buffers (2.6%); 0 transaction log file(s) added, 0 removed, 0 recycled;
> write=149.816 s, sync=0.064 s, total=149.890 s; sync files=314,
> longest=0.002 s, average=0.000 s; distance=145449 kB, estimate=236346 kB
> [ 2017-04-17 17:21:56 BRT] @ LOG: recovery restart point at 126/A7072A88
> [ 2017-04-17 17:21:56 BRT] @ DETAIL: last completed transaction was at
> log time 2017-04-17 17:21:02.289164-03
>

I wish the checkpoint logging code would tell you how many transaction log
file it intentionally retained as well. I've not seen it "accidentally"
retain files, but seeing the number logged would help simplify
troubleshooting.

Does each file in pg_xlog/archive_status/ have a corresponding file one
directory up?

Cheers,

Jeff