Our dedicated servers provider OVH (ovh.com) has sent us an alert about a hardware (electrical reboot) replacement that will happen on June 14, at 6:00AM EDT and last about 1 to 1.5 hour:
http://status.ovh.net/?do=details&id=14704&edit=yep
This hardware replacement will affect one of our core servers which is relied by our other servers to perform EasyCron's service.
As a result, during the intervention, our service will be interrupted. And the service will recover automatically once the maintenance is finished and the server is booted again.
From information we got, the replacement of electric part could do no impact to the server:
http://status.ovh.net/?do=details&id=14521
or cause a variable length of downtime (from dozen of minutes to 1 hour):
http://status.ovh.net/?do=details&id=14437
http://status.ovh.net/?do=details&id=14530
There is nothing we can do now to 100% avoid this service interrupt. But we're working on a solution to have our HA (high availability) strategy cover this server, so that in the future if similar maintenance happen again, our service will not be affected.
As a remedy, EasyCron will fire those cron jobs that have missed executions during the maintenance for *one* time after the service back to normal. That means, your affected cron jobs will get triggered for one time after our server is up again.
We're really sorry for the inconvenience that will be caused by this server maintenance.
UPDATE:
Our server's downtime started from
June 14th, 6:39 AM EDT, and ended on
June 14th, 7:14 AM EDT, totally lasted
35 minutes.
Our service fully backed to normal at
June 14th, 7:20 AM EDT.
No comments:
Post a Comment