Hi there,
I created two patches improving the log messages generated by
log_lock_waits. The first patch shows the process IDs holding a lock
we try to acquire (show_pids_in_lock_log.patch), sample output
(log_lock_waits=on required):
session 1: BEGIN; LOCK TABLE foo IN SHARE MODE;
session 2: BEGIN; LOCK TABLE foo IN SHARE MODE;
session 3: BEGIN; LOCK TABLE foo IN EXCLUSIVE MODE;
Output w/o patch:
LOG: process 13777 still waiting for ExclusiveLock on relation 16385 of database 16384 after 1000.030 ms
Output with patch:
LOG: process 13777 still waiting for ExclusiveLock on relation 16385 of database 16384 after 1000.030 ms
CONTEXT: processes owning lock: 13775, 13776
The second patch (show_table_name_and_tuple_in_lock_log.patch)
includes relation info (table name and OID) as well as some tuple
information (if available). Sample output (log_lock_waits=on required):
session 1:
CREATE TABLE foo (val integer);
INSERT INTO foo (val) VALUES (1);
BEGIN;
UPDATE foo SET val = 3;
session 2:
BEGIN;
UPDATE TABLE foo SET val = 2;
Output w/o patch:
LOG: process 24774 acquired ShareLock on transaction 696 after 11688.720 ms
Output with patch:
LOG: process 24774 acquired ShareLock on transaction 696 after 11688.720 ms
CONTEXT: relation name: foo (OID 16385)
tuple (ctid (0,1)): (1)
Regarding this patch I am not really sure where to put the
functions. Currently they are located in backend/storage/lmgr/lmgr.c
because XactLockTableWait() is located there, too. What do you think?
I also created two test specs for easy creation of the log output;
however, I was not able to provide an expected file since the process
IDs vary from test run to test run.
Regards,
--
Christian Kruse http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services