Hello all, At Sunday 24 February 2013 15:28:18 DaB. wrote:
the movement is done (commons is missing at the moment at s5). s2 was also rebooted successful and we had a general downtime arround 21:20 UTC. More details about all 3 tomorrow.
ok, now with more details (sorry, but I really had to went for bed last night). The movement of the user-databases worked without problems. The dump&import was done within 30min. The general downtime happened shortly after so sql-s5 was read-only for longer because I had to fix the other problem first. Please check if everything was moved correct and message me if anything is missed (the old box will be there until Wednesday 12:00 UTC). There was a unplanned restart of mysql of sql-s5 after the movement which caused a downtime for 10 minutes – sorry for that. The commons-import was not done at that time and broke for that reason. I will restart it after the reboot tonight. Because z-dat-s5-b (the new host for sql-s5) shows similar performance- problems like z-dat-s2-b (the new host for s2) I have configured the cassia (the old sql-s5 host) for sql-s5-rr to lift a little load from z-dat-s5-b; I will also add the same config-changes of z-dat-s2-b on z-dat-s5-b tonight.
Sincerely, DaB.