mirror/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-01-18 04:53:01 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions

Very minor fixes for Seconds_Behind_Master column of SHOW SLAVE STATUS.

Browse source
This commit is contained in:
guilhem@mysql.com 2004-12-16 22:38:42 +01:00
parent cd4f53f244
commit 3287e82f09
2 changed files with 18 additions and 14 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
sql/log_event.cc
View file

@ -373,6 +373,9 @@ int Log_event::exec_event(struct st_relay_log_info* rli)
Note that Rotate_log_event::exec_event() does not call this function,
so there is no chance that a fake rotate event resets
last_master_timestamp.
Note that we update without mutex (probably ok - except in some very
rare cases, only consequence is that value may take some time to
display in Seconds_Behind_Master - not critical).
*/
rli->last_master_timestamp= when;
}

29
sql/slave.cc
View file

@ -4189,6 +4189,21 @@ Before assert, my_b_tell(cur_log)=%s rli->event_relay_log_pos=%s",
*/
if (hot_log)
{
/*
We say in Seconds_Behind_Master that we have "caught up". Note that
for example if network link is broken but I/O slave thread hasn't
noticed it (slave_net_timeout not elapsed), then we'll say "caught
up" whereas we're not really caught up. Fixing that would require
internally cutting timeout in smaller pieces in network read, no
thanks. Another example: SQL has caught up on I/O, now I/O has read
a new event and is queuing it; the false "0" will exist until SQL
finishes executing the new event; it will be look abnormal only if
the events have old timestamps (then you get "many", 0, "many").
Transient phases like this can't really be fixed.
*/
time_t save_timestamp= rli->last_master_timestamp;
rli->last_master_timestamp= 0;
DBUG_ASSERT(rli->relay_log.get_open_count() == rli->cur_log_old_open_count);
/*
We can, and should release data_lock while we are waiting for
@ -4231,20 +4246,6 @@ Before assert, my_b_tell(cur_log)=%s rli->event_relay_log_pos=%s",
*/
pthread_mutex_unlock(&rli->log_space_lock);
pthread_cond_broadcast(&rli->log_space_cond);
/*
We say in Seconds_Behind_Master that we have "caught up". Note that
for example if network link is broken but I/O slave thread hasn't
noticed it (slave_net_timeout not elapsed), then we'll say "caught
up" whereas we're not really caught up. Fixing that would require
internally cutting timeout in smaller pieces in network read, no
thanks. Another example: SQL has caught up on I/O, now I/O has read
a new event and is queuing it; the false "0" will exist until SQL
finishes executing the new event; it will be look abnormal only if
the events have old timestamps (then you get "many", 0, "many").
Transient phases like this can't really be fixed.
*/
time_t save_timestamp= rli->last_master_timestamp;
rli->last_master_timestamp= 0;
// Note that wait_for_update unlocks lock_log !
rli->relay_log.wait_for_update(rli->sql_thd, 1);
// re-acquire data lock since we released it earlier