I_S.processlist TIME_MS occasionally shows a truncated BIGINT max value

Bug #1402954 reported by Peiran Song
12
This bug affects 2 people
Affects Status Importance Assigned to Milestone
Percona Server moved to https://jira.percona.com/projects/PS
New
Undecided
Unassigned
5.5
New
Undecided
Unassigned
5.6
New
Undecided
Unassigned

Bug Description

A customer observed this on PS 5.5.40 that, for slave SQL thread, the I_S.processlist TIME_MS sometimes has a value of 18446744073709552. Also TIME_MS and TIME often are not consistent with each other for the SQL thread.

If the first sample shows TIME and TIME_MS as expected, then the rest showed that the two are not consistent and the last two samples showed the huge number associated with different STATE.

mysql> select * from information_schema.processlist where id = 21588\G
--------------
select * from information_schema.processlist where id = 21588
--------------

*************************** 1. row ***************************
           ID: 21588
         USER: system user
         HOST:
           DB: NULL
      COMMAND: Connect
         TIME: 2
        STATE: Slave has read all relay log; waiting for the slave I/O thread t
         INFO: NULL
      TIME_MS: 1397
    ROWS_SENT: 0
ROWS_EXAMINED: 0
    ROWS_READ: 2
1 row in set (0.08 sec)

mysql> select * from information_schema.processlist where id = 21588\G
--------------
select * from information_schema.processlist where id = 21588
--------------

*************************** 1. row ***************************
           ID: 21588
         USER: system user
         HOST:
           DB: assistly
      COMMAND: Connect
         TIME: 8
        STATE: Searching rows for update
         INFO: UPDATE `ticket_attachments` SET `ticket_id` = 342046622 WHERE `ticket_attachments`.`ticket_id` = 342109693 AND `ticket_attachments`.`site_id` = 146598
      TIME_MS: 161
    ROWS_SENT: 0
ROWS_EXAMINED: 0
    ROWS_READ: 1
1 row in set (0.08 sec)

mysql> select * from information_schema.processlist where id = 21588\G
--------------
select * from information_schema.processlist where id = 21588
--------------

*************************** 1. row ***************************
           ID: 21588
         USER: system user
         HOST:
           DB: assistly
      COMMAND: Connect
         TIME: 11
        STATE: System lock
         INFO: UPDATE sites SET ticket_display_id = LAST_INSERT_ID(COALESCE(`ticket_display_id`,0) + 1) WHERE id = 200559
      TIME_MS: 18446744073709552
    ROWS_SENT: 0
ROWS_EXAMINED: 0
    ROWS_READ: 1
1 row in set (0.09 sec)

mysql> select * from information_schema.processlist where id = 21588\G
--------------
select * from information_schema.processlist where id = 21588
--------------

*************************** 1. row ***************************
           ID: 21588
         USER: system user
         HOST:
           DB: assistly
      COMMAND: Connect
         TIME: 9
        STATE: Updating
         INFO: UPDATE `user_loads` SET `agent_session_id` = 28767527, `updated_at` = '2014-12-15 19:24:11' WHERE `user_loads`.`id` = 145147
      TIME_MS: 18446744073709552
    ROWS_SENT: 0
ROWS_EXAMINED: 0
    ROWS_READ: 1
1 row in set (0.09 sec)

Tags: i48583 i66073
tags: added: i48583
Revision history for this message
Nilnandan Joshi (nilnandan-joshi) wrote :

Not getting BIGINT value in TIME_MS with PS 5.6.21 but its true that TIME_MS and TIME often are not consistent with each other for the SQL thread.

mysql> select * from information_schema.processlist where id = 1\G
*************************** 1. row ***************************
           ID: 1
         USER: system user
         HOST:
           DB: test
      COMMAND: Connect
         TIME: 106
        STATE: updating
         INFO: update nil set name = 'nilnandan' where name = 'njoshi'
      TIME_MS: 2175
    ROWS_SENT: 0
ROWS_EXAMINED: 167564
1 row in set (0.00 sec)
...
..mysql> select * from information_schema.processlist where id = 1\G
*************************** 1. row ***************************
           ID: 1
         USER: system user
         HOST:
           DB: test
      COMMAND: Connect
         TIME: 213
        STATE: updating
         INFO: update nil set name = 'nilnandan' where name = 'njoshi'
      TIME_MS: 108743
    ROWS_SENT: 0
ROWS_EXAMINED: 4171262
1 row in set (0.00 sec)

.

Revision history for this message
Przemek (pmalkowski) wrote :

Another example of time_ms showing bin int maxvalue:

mysql> SELECT host,db,time,time_ms,state,LEFT(info,100) FROM information_schema.processlist WHERE time=0 AND time_ms>1000 AND command !='Binlog Dump' ORDER BY time DESC LIMIT 100;
+------------------+----------------------+------+-------------------+-------+----------------+
| host | db | time | time_ms | state | LEFT(info,100) |
+------------------+----------------------+------+-------------------+-------+----------------+
| 10.10.1.53:50377 | beanstalk_production | 0 | 18446744073709552 | | NULL |
| 10.10.1.52:55013 | beanstalk_production | 0 | 18446744073709552 | | NULL |
| 10.10.1.73:40279 | dploy_production | 0 | 18446744073709544 | | NULL |
| 10.10.1.78:54975 | dploy_production | 0 | 18446744073709536 | | NULL |
| 10.10.1.78:54795 | dploy_production | 0 | 18446744073709544 | | NULL |
| 10.10.1.13:33716 | beanstalk_production | 0 | 18446744073709540 | | NULL |
+------------------+----------------------+------+-------------------+-------+----------------+
6 rows in set (0.01 sec)

Version | 5.5.44-37.3-log Percona Server (GPL), Release 37.3, Revision 729fbe2

tags: added: i66073
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.