fremont86 ran into an issue with its RAID controller. We are rebooting the host, and Linodes will be coming back up shortly.

We apologize for the inconvenience.

_________________
Disclaimer: I am no longer employed by Linode; opinions are my own alone.

Another reboot. We think we've identified the problem being a faulty drive which is causing the RAID unit trouble. We'll be replacing this shortly. Linodes should be coming back on line shortly.

We apologize for the inconvenience.

-Chris

Additional problems have been encountered. We're working on a resolution. Updates in a bit.

-Chris

We're working on configuring migrations off this machine now, which we will execute one by one. Please stand by for further updates.

Again, we apologize for this inconvenience.

I'm getting some serious lag trying to SSH into my node on fremont87. Is this related?

No.

We're still working through the problems with fremont86. Hopefully I'll have some news shortly.

-Chris

Any new news on what is going on?

Thanks,
Blue

Fremont86 is having issues with its RAID unit. We're working on a solution as quickly as we can. Hopefully an announcement soon.

-Chris

All Linodes on fremont86 have been booted. The RAID unit is rebuilding.

This downtime stemmed from what appears to have been a faulty drive in the RAID10 unit, which somehow caused its pair (RAID10 is two RAID1s striped) to have errors, so the subunit started rebuilding itself, despite the source drive encountering ECC errors.

Things should continue along now, while the unit rebuilds itself. However, once that's done (in a day or so) we'll be rebooting this machine again. For that, we'll send out advanced notice via the ticket system.

Again, we're sorry for the downtime but unfortunately hardware problems happen. We're just happy that this resulted in some downtime instead of something catastrophic.

Regards,
-Chris

My Linode's back up. Thanks for the fast action, caker!

To keep the record straight, fremont86 was taken down for about 12 minutes again at around 01:04am eastern time. A message was broadcast to the tty right before the shutdown.

Would there be any more anticipated downtime due to this RAID issue? Or it is resolved?

Thanks.

We opened tickets with everyone on this host notifying them of an imminent migration. That downtime you saw was your Linode's turn in the migration queue, being moved to another host, so -- your Linode is no longer on fremont86...

-Chris