[Toolserver-l] Replication is back after master-switch and schema-update

[DaB.]

Hello all,

yesterday at highnoon the master-db-servers were switched because of an huge 
schema-update. Because we have only one cluster of db-servers at the moment we 
couldn't do the schema-update the same way like the foundation (the foundation 
stops one of their slaves, does the schema-update, waits, start it again, 
stops another slave, does the schema-update and so on until all slaves are 
updated, and declare one slave as new master, runs the schema-update on the 
old master = no downtime) and so it took several hours until the schema-update 
was done and we could switch to the new master-servers.
I deleted the view on the logging-table during the update to speed the update 
up and killed several long-running queries too tonight.
The update was finish before I wake-uped, and I just finished the 
masterswitch, restarted replication and recreated the logging-tables (see 
maintaince-log for details). So in a few hours everything should be normal 
again.

Happy Corpus Christi (it's an holiday here :)).

Sincerly,
DaB.

-- 
wp-blog.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
Url : http://lists.wikimedia.org/pipermail/toolserver-l/attachments/20090611/fcf2f8ea/attachment.pgp