D-Link Forums
The Graveyard - Products No Longer Supported => D-Link Storage => DNS-321 => Topic started by: rbeheshti on July 29, 2008, 08:36:48 AM
-
One of the drives in my DNS-323 failed (Raid 1) >:(. How do I safely recover from this failure?
I had two identical 750 Gig Seagate configured as Raid 1 in the unit, one of the drives has failed, I just bought another identical drive to replace it, but what is the right procedure to do it w/o losing any data?
If I just replace the bad drive, does the unit automatically re-mirror the new drive from the working one?
Will I be prompted to re-format and/or reconfigure the unit as RAID 1 again?
I just don't want to loss any data and want to make sure I follow the right procedure.
Can I start the unit with just the one good drive and copy my data before doing anything else first? how?
Please HELP.
THanks.
RB
-
I believe this is documented in the FAQ on the support website - anyway - with RAID1 and a single disk failure, your data should be accessible, I would start by backing it up - out of an abundance of caution - and then shut the system down and remove the failed drive and replace it with the new one.
When the system is powered up, log in through the web interface, it should prompt you to format the replacement drive after which I think it will prompt for a restart and it should then resynch or rebuild the array.
Theoretically this entire process can be done without loss of data.
-
Smallnetbuilder.com has also reported that they tested the RAID feature and it discovered a RAID1 failure and it was tricky process and difficult to recover so please investigate into this D-Link and if it’s a bug kill it man this is already starting to concern me so here is a link to the article:
???
http://www.smallnetbuilder.com/content/view/30521/75/1/2/
-
Smallnetbuilder.com has also reported that they tested the RAID feature and it discovered a RAID1 failure and it was tricky process and difficult to recover so please investigate into this D-Link and if it’s a bug kill it man this is already starting to concern me so here is a link to the article:
???
http://www.smallnetbuilder.com/content/view/30521/75/1/2/
You do realise that the reviewer read the FAQ and found the approriate instructions fter he had already done things incorrectly.
From my point of view - and this is after some 15 years of working with RAID - the only failure described there was a failure to follow instructions - for RAID1 to be considered functional, all that was required was the data to remain available after a drive had failed or in this case had been removed.
The mechanisms and procedures for drive replacement and rebuilding the array are dependent on the product and it's manufacturer, and whilst the reviewer may not have come across any other device that requires a fresh drive (and neither have I) - and the fact that the unit did not do what he expected that does not mean that the RAID failed or that there was a bug, it just means his expectations may have been incorrect.
I would suggest that you NEVER "reinsert a failed drive" into a RAID1 array without wiping it clean - whether it is this product or any other one, I learned the hard way many years ago that in doing this you run the risk of rebuilding the array from the wrong drive.