Horton

Re: Add jsonlog log_destination for JSON server logs

Andrew Dunstan <andrew@dunslane.net>

View thread
From: Andrew Dunstan <andrew@dunslane.net>
To: Michael Paquier <michael@paquier.xyz>, Andres Freund <andres@anarazel.de>
Cc: Sehrope Sarkuni <sehrope@jackdb.com>, Pg Hackers <pgsql-hackers@postgresql.org>, david@fetter.org
Date: 2022-01-06T23:28:26Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Introduce log_destination=jsonlog

  2. Refactor set of routines specific to elog.c

  3. Fix compilation warning in syslogger.c

  4. Refactor per-destination file rotation in logging collector

  5. Refactor output file handling when forking syslogger under EXEC_BACKEND

  6. Add regression tests for csvlog with the logging collector

  7. Refactor the syslogger pipe protocol to use a bitmask for its options

On 1/6/22 13:06, Andrew Dunstan wrote:
> On 1/5/22 02:32, Michael Paquier wrote:
>> On Sun, Jan 02, 2022 at 01:34:45PM -0800, Andres Freund wrote:
>>> The tests don't seem to pass on windows:
>>> https://cirrus-ci.com/task/5412456754315264?logs=test_bin#L47
>>> https://api.cirrus-ci.com/v1/artifact/task/5412456754315264/tap/src/bin/pg_ctl/tmp_check/log/regress_log_004_logrotate
>>>
>>> psql:<stdin>:1: ERROR:  division by zero
>>> could not open "c:/cirrus/src/bin/pg_ctl/tmp_check/t_004_logrotate_primary_data/pgdata/current_logfiles": The system cannot find the file specified at t/004_logrotate.pl line 87.
>> This seems to point out that the syslogger is too slow to capture the
>> logrotate signal, and the patch set is introducing nothing new in
>> terms of infrastructure, just an extra value for log_destination.
>> This stuff passes here, and I am not spotting something amiss after an
>> extra close read.
>>
>> Attached is an updated patch set that increases the test timeout (5min
>> -> 10min).  That should help, I assume.
>
> ITYM 3 min -> 6  min. But in any case, is that really going to solve
> this? The file should exist, even if its contents are not up to date, AIUI.



I have tested on an msys2 setup with your v8 patches and I am getting this:


#   Failed test 'current_logfiles is sane'
#   at t/004_logrotate.pl line 96.
#                   'stderr log/postgresql-2022-01-06_222419.log
# csvlog log/postgresql-2022-01-06_222419.csv
# '
#     doesn't match '(?^:^stderr log/postgresql-.*log
# csvlog log/postgresql-.*csv
# jsonlog log/postgresql-.*json$)'

#   Failed test 'found expected log file content for stderr'
#   at t/004_logrotate.pl line 103.
#                   ''
#     doesn't match '(?^:division by zero)'

#   Failed test 'found expected log file content for jsonlog'
#   at t/004_logrotate.pl line 105.
#                   undef
#     doesn't match '(?^:division by zero)'

#   Failed test 'pg_current_logfile() gives correct answer with jsonlog'
#   at t/004_logrotate.pl line 105.
#          got: ''
#     expected: undef
# Looks like you failed 4 tests of 14.
[22:37:31] t/004_logrotate.pl ...
Dubious, test returned 4 (wstat 1024, 0x400)
Failed 4/14 subtests
[22:37:31]


cheers


andrew


--
Andrew Dunstan
EDB: https://www.enterprisedb.com