NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
Solved
RN102 fails to boot when one mirrored drive fails or is missing
Hello everyboby,
I have been looking for a solution or help but was not able to find any hints regarding the following problem.
I own a RN102 device equiped with two disk forming a X-RAID mirror. So far, so good. The RAID is healthy and looks fine. The last few days the device complained that disk2 is likely to fail. So I tried to replace the device by pulling the disk accordingto the handbook. At this moment the RN102 stucks an is not reachable anymore. When I reset the device, the poer LED keeps on blinking. RAIDAR is able to find the device but indicating the the is no configuration on it. The first disk is show as already used with data. When I put the second disk back an reboot the RN, it comes up fine and operates normally.
It seems that the OS an it's config is missing on the first disk.
Any idea how to get the corrected?
Thanks to everybody.
Greetings
Chris
Well, it's still unclear to me what the root cause was. Somehow the RAID superblock that gets generated for the newly-added disk kept getting rejected because the kernel thought it was invalid. However, I don't know which part it thought was invalid, and there's no easy way to remotely debug at that low level.
Anyway, I got your box fixed up by restoring a good superblock made on another machine to your disk, and it's resyncing now and looks normal. I also made sure that volume repair worked after simulating a failure, so I don't think you'll run into this again. If you do, please let me know.
19 Replies
Replies have been turned off for this discussion
Somehow your disk 1 does not have a root partition. It's not clear how this situation developed, but our current software does not automatically repair this situation. That will be added in 6.9.0.
So there are two options for you. You can update to 6.9.0 Beta 2 (most likely available later today) which should fix it up automatically, or you can enable remote access for one of us and we can manually fix it up for you.
Hello Skywalker,
thank you for the prompt feedback. I have choosen a third approach and replace the drive not booting with a new one. After rebuild, the system is now able to boot with the other disk pulled. So I guess the complete rebuild created the root partition too. THANKS for the support!
Now I am facing an other problem. I now tried to replace the now pulled disk with one I have used before in that enclosure and the system doen not start to rebuild but indicates that the inactive volumes should be removed. I have tried to diskpart clean in windows for that drive but still the same error. I see the new (reused) disk colored red and some kind of ghost volume appearing when it is insert.
Any ideas how to force the system to accep the old disk. Pushing destroy form the volume menue doesn't have any effect.
Thank you and best greeting.
Chris
Great, I'm glad option #3 worked for you. I was going to suggest that, but I was nervous since the bootable disk had some errors. But I'm happy to see it worked for you. :-)
When you click on the red (reused) disk, do you see a Format button on the right side of the GUI?
I think the first thing to do is make sure you have a current backup of the data.
At the moment does the volume tab show the RAID status as redundant or degraded?
If it is redundant, then download the log zip file. Look in disk_info.log at the SMART status for both disks. Maybe also look in mdstat.log, and see if the RAID array is using both disks.
If the disks are healthy and you have a backup, then you can do a factory reset (with the original disk 1 and the replacement disk 2). Then rebuild the NAS and restore the data. But Netgear might want to look at your logs before you do that, so I suggest holding off on the "nuclear" option for now.
Hello Stephen,
thank you for the answer. The disks seem ok, the system indicator is healthy but I cannot find any explicit indicator for the RAID Volume. Disk logs and mdstat.log indicate everything as ok (raid level 1, state clean for the two system set and the data set)
So, I guess I do not have to worry about loosing data shortly (and I got backups too).
But why is the device stopping to work when one disk is pulled and even not willing to boot? Doing a factory reset is honestly not a good solution. To protect from such disk failure szenarios I got the RAID1 setup. But this makes no sense when I have to do a factory reset every time a disk fails........
Greetings
Chris
chris2531 wrote:
But this makes no sense when I have to do a factory reset every time a disk fails........
Of course you shouldn't have to do this, so something is definitely wrong.
An alternative to the factory reset is to remove the mis-behaving disk, and then unformat it in a PC. With windows you can do that by going into the windows disk manager and deleting each of the "Volumes" it shows on the disk (they are really partitions). You can also zero the disk with vendor diags (The quick zero test in WDC's lifeguard program will do the job nicely). Then hot-insert the drive, and the NAS should rebuild it. That includes rebuilding the OS partition on it, and any boot sector information.
Though I still recommend waiting for a bit to see if Netgear wants to take a look. Your problem is a bit unusual I think.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
