Server 2003: "failed redundancy" on software mirror

[cloud32187]

So I have this customer running server 2003 with the raid 1 set up in windows. Not ideal I know. Went to the business today on an unrelated issue regarding network connectivity (no one could connect to server and couldn't RDP). Simply rebooted the server and all the workstations and that issue was resolved.

Later on I went to snoop around the server to see if I could find the cause. For shits and giggles i went to disk management (drives are barely a year old) only to see "failed redundancy" and one of the drives is "missing" status but still recognized.

When I try to "reactivate disk" the cursor is busy then stops a few seconds later, and no change to the status.

The BIOS picks up both drives no problem with no alerts and even gives you the option of what drive to boot from when starting up the server. Just checking to see if there is anyone out there who knows anything about this before I go and just replace the drive.

Here's a picture of diskmgmt:
http://i.imgur.com/sF0YqpP.png

Here's a link to the technet article with steps I have tried:
http://technet.microsoft.com/en-us/library/cc732026.aspx

 

[pceinc]

Have you verified the health of the drives using a disk utility?

 

[YeOldeStonecat]

Sounds like that second drive is shot....perhaps time to "break the mirror", swap the disk with a new one, and build a fresh mirror.

 

[cloud32187]

Well thank you for the input, however the issue is now solved. I had to wait till all the employees left to give the server a restart. Once I restarted, I tried to "reactivate disk" again, it instantly started resyncing the mirror. It is now fully synced and working properly.

I got worried cause I did try to check the health of the drives but one was not being picked up due to being "missing" in disk management. But now all drives show up and are very healthy.

 

[brandonkick]

Do yourself a favor a make a clone of that setup...

I think your going to find the drive is failing.

 

[YeOldeStonecat]

My hunch too....the drive fell off for a reason.
Sometimes you can massage things back into place....for a while....and then...

...it's a server, why risk it. By the description (software RAID) and the size of the drives (500 gigs) I'm guessing it's a budget server thus SATA disks...so, pair of new enterprise grade SATA disks of that capacity don't cost much...pretty cheap insurance.

 

[NETWizz]

I have seen a ton of times with a hardware RAID that a drive fell out of the array with a blinking, amber light.... Removing and re-inserting the drive generally re-built the array and restored redundancy often for years.

Not saying the drive won't fail again only that sometimes a drive falls out for no apparent reason.