Alex
- Total activity 15
- Last activity
- Member since
- Following 0 users
- Followed by 0 users
- Votes 0
- Subscriptions 2
Comments
Recent activity by Alex Sort by recent activity-
Sorry, links:Galera: https://launchpad.net/galera/1.x/22.1.1 MySQL 5.1: https://launchpad.net/codership-mysql/5.1/5.1.59-22.2 MySQL 5.5: https://launchpad.net/codership-mysql/5.5/5.5.17-22.3
-
Hi everybody, We have released new MySQL/Galera versions which should fix that OOM bug (certification index is now cleaned up as soon as it exceeds certain size) Although the bug is in Galera libra...
-
Mark, this is exactly the case of certification index buildup. This is fixed in trunk.
-
Mark, It depends on what you mean by "helping". Indeed, deleting a million rows from a table is a good example of a key-only transaction that would take a lot of RAM in Galera while containing no d...
-
Mark, I'm quite confident that OOM that you have had was due to certification index buildup. We have identified a test case where certification index could grow up to 1.2Gb. So it certainly can be ...
-
Hi everybody. I'd like to let you know that we have identified at least one source of trouble to be certification index being purged insufficiently frequently and implementing a fix for it to be in...
-
So, which table crashes the cluster, devices_i? Or any three of them can?
-
Hi, That's strange. Galera, or MySQL-wsrep patch for that matter does not create any temporary tables, at least intentionally. MySQL itself can in some circumstances, but this should not be happen...
-
Thanks for the logs, they tell a story of a tragic beauty: Everything went well until at around 14:40-14:50 slaves (db1, db3, db4) started to experience some performance problems, most likely IO st...
-
Looking at the log, this particular node lost connectivity to the rest of the cluster at around 2:42:00. Anything specific happened at that point? Also it looks like 10.176.215.163 was restarted th...