Re: [PATCHES] log_line_info

Bruce Momjian wrote:

>Andrew Dunstan wrote:
>
>
>>>Nice. Only two comments --- does this mean we should remove log_pid?
>>>Seems it is now duplicate functionality. Is that the only duplication?
>>>Also, I don't see any documention changes in the patch, but I assume you
>>>will work on that before final.
>>>
>>>
>>>
>>I will do docs. We could sensibly get rid of log_pid and log_timestamp
>>with my latest patch. I would also suggest getting rid of
>>log_source_port, since there really isn't any reason *not* to log the
>>source port. Do you want me to make those changes in my patch? Or I can
>>leave them for now and we can get rid of them when everyone is happy.
>>
>>
>
>I agree, but let's make it a separate patch.
>
>Oh, I think we still need log_timestamp for postmaster-generated lines,
>no? What does log_line_info output for postmaster-generated logs?
>

I have fixed the postmaster lines issue. There is a new escape %X that
says "postmaster and friends stop here".

Example, with log_timestamp = false and log_pid = false and
log_line_info = '%T [%P] %X %U(at)%D(%C:%S) %R %I line:%L ' :

2004-02-28 20:24:49 [11154] LOG: database system was shut down at
2004-02-28 20:24:20 EST
2004-02-28 20:24:49 [11154] LOG: checkpoint record is at 0/9D1874
2004-02-28 20:24:49 [11154] LOG: redo record is at 0/9D1874; undo
record is at 0/0; shutdown TRUE
2004-02-28 20:24:49 [11154] LOG: next transaction ID: 467; next OID: 17145
2004-02-28 20:24:49 [11154] LOG: database system is ready
2004-02-28 20:25:30 [11158]
[unknown](at)[unknown](40413f8a.2b96:2004-02-28 20:25:30) line:1 LOG:
connection received: host=alphonso port=45621
2004-02-28 20:25:30 [11158] andrew(at)blurflx(40413f8a.2b96:2004-02-28
20:25:30) alphonso:45621 authentication line:2 LOG: connection
authorized: user=andrew database=blurflx
2004-02-28 20:25:34 [11158] andrew(at)blurflx(40413f8a.2b96:2004-02-28
20:25:30) alphonso:45621 idle line:3 LOG: statement: SELECT n.nspname
as "Schema",
c.relname as "Name",
CASE c.relkind WHEN 'r' THEN 'table' WHEN 'v' THEN 'view' WHEN
'i' THEN 'index' WHEN 'S' THEN 'sequence' WHEN 's' THEN 'special' END as
"Type",
u.usename as "Owner",
c2.relname as "Table"
FROM pg_catalog.pg_class c
JOIN pg_catalog.pg_index i ON i.indexrelid = c.oid
JOIN pg_catalog.pg_class c2 ON i.indrelid = c2.oid
LEFT JOIN pg_catalog.pg_user u ON u.usesysid = c.relowner
LEFT JOIN pg_catalog.pg_namespace n ON n.oid = c.relnamespace
WHERE c.relkind IN ('i','')
AND n.nspname NOT IN ('pg_catalog', 'pg_toast')
AND pg_catalog.pg_table_is_visible(c.oid)
ORDER BY 1,2;
2004-02-28 20:25:38 [11158] andrew(at)blurflx(40413f8a.2b96:2004-02-28
20:25:30) alphonso:45621 idle line:4 LOG: disconnection: session time:
0:00:08.50 user=andrew database=blurflx host=alphonso:45621
2004-02-28 20:25:44 [11149] LOG: received smart shutdown request
2004-02-28 20:25:44 [11170] LOG: shutting down
2004-02-28 20:25:46 [11170] LOG: database system is shut down

>
>Also, should we call the option just log_line? Is that clearer, or
>log_line_prefix?
>
>

I floated the name log_line_info a while back and noone objected. I
don't think "log_line" is any clearer - quite the contrary IMNSHO.

cheers

andrew