Vaibhav Singh

Blog - vaibhavsingh.com

Screwed my RAID configuration

A fine March Wednesday, I spun my home lab and loaded configuration into PNET labs, as I began working through a workbook, that is when I was reminded of a planned vacation that was coming up that very weekend.

As I really didn’t want to shut the entire thing down; it takes ages to reboot, set up and reload everything to continue from where I was. I have a good UPS, so, I left it running as is and went out of town!

Upon return, I found the lab session ended and there were alarms regarding disk failure lit up on ESXi as well as the DELL front panel.

Apparently, one out of a pair of IBM 10K SAS 1Tb disks (RAID-0) had crashed. This took down my entire COVID19 lock-down era lab work. I did have exported backups of my PNET lab projects, but the RTO of rebuilding the entire setup from scratch was huge – a weekend worth of effort, at least.

Lessons learnt

RAID-0 is good for speed and cache purpose, but don’t use it for anything that need retention or is important to you.

Conclusion

Ordered two additional SAS disks, and another USB3 external 1Tb HDD. Now the configuration is RAID-1 with 2 disks + 1 as hot spare.

The USB SSD is attached to a VM running Veeam Backup and Restore software to take regular backups.

Going another step, I’ve setup an automatic MEGA cloud backup off from the External HDD, as when it is idle.

Tags:

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top