Thursday 17th May 2018

Atlas MySQL failed ASR, scheduled 6 months ago

Failed automated system recovery detected for MySQL on Atlas. Investigating further.

1:46 PM EDT- localized InnoDB pagefile corruption. Issue addressed, another MySQL restart necessary and each restart, given the size of the database server, takes about 5-8 minutes to complete.

1:54 PM EDT- MySQL immediately shutting down after starting up. Investigating further.

1:58 PM EDT- MySQL service restored. Initial cause for startup -> shutdown cycle still under investigation; appears to be linked to systemd process manager.

2:05 PM EDT- systemd service definition configuration changed using Luna, another platform as reference. MySQL comes up without issue now.