Some friend called me to help with a celerra NS20. About 1TB data is not accesible and is very worried.
This morning 1 drive faulted in group raid with two other disks already faulted, so now, there are 3 disk faulted of 5. But, i'm not sure that this is related to the problem, because i understand that with two disk faulted this raid group was innacessible since 2014 when other disk of the group become faulted.
Over this raid there are two LUNs and about 1TB data space. There is only one storage group forme by all LUNs of the system.
I connected to navisphere and in alert panel i read the log for the disk faulted but no other events shown. Since 2013 there is no events related to luns become offline.
Some folder are accesible and some others not. But i dont know how can i see the mapping between luns and data or how to try to mount the folder. Also, dont know how can i access data from navicli or control station.
if file system is corrupted, i am amazed datamover has not paniced. Have you connected to the control station and start investigating while file systems are there, which are still mounted.
The first command shows me a list of directories, i only recognized two of them, i think they are the two main storage filesystems. One can be accesed, the other not.
Second command gives me an error 4000.
Is there a way of extract raw data of LUNs online? There are some programs that analyze data trying to find files like excel or doc file types.
I think the main issue is to try to repair the faulted disks. If the raid group was a Raid 6 which can withstand a two disk failure and still function and a third disk fails, then the raid group will be faulted and access disabled to the LUNs from the raid group.
The way to fix this is to replace the last disk that failed (the 3rd one) first and let the raid group rebuild, then replace the second disk that failed, let raid group rebuild, the replace the first disk that failed. This is only safe way to fix this issue, removing or moving around any of the faulted disks could result in the correct disk getting into the wrong slot. It's very important that if you do remove a disk, always return the disk back to the same slot before during any other work.
The raid group should be OK and the LUNs should come back on-line once you replace the disks.
dynamox
11 Legend
•
20419 Posts
•
87439 Points
711
0
Posted December 22nd, 2015 19:00
if file system is corrupted, i am amazed datamover has not paniced. Have you connected to the control station and start investigating while file systems are there, which are still mounted.