inactive volumes on snap4500

[mortenm]

hi cool guys

i have a snap4500 which just went down because of a short powerloss and no UPS

now i cant access any of the shares and when i look in the webinterface it says that my 3 volumes are inactive.

how can i "activate" them again? i really doesnt hope that all of my data is lost :s

thank you very much in advance

//mortenm

[blue68f100]

I haven't heard that one. Normally it see everything run's check disk then mounts.

[mortenm]

hi blue, thanks for reply...

i found this thread: http://forums.procooling.com/vbb/showthread.php?t=13487 - butit doesnt seems that the problem got solved there neither...?

[blue68f100]

If this is a supper block error, SpinRite may be able to fix it.

That thread was over 4yrs old, no wonder I didn't remember it. I did not have my GOS unit then.

[mortenm]

yeah i know its an old thread

spinrite? but it will delete data or am i wrong?

if not, how am i going to do?

thanks, i really appreciate your help

[mortenm]

i dont know if this can help anyone

[blue68f100]

That can not be good. If the data is a must have a recovery service may be you only option. If you know what the volume names are you may be ok.

I would try Spinrite, at this point you do not have a lot of options. SpinRite will not delete data, or never has for me. Once it sees things are not right it drops into a recovery mode. In recovery mode it queries multiple reads on the same bit. If says by reading before and after it can determine what it should be. In theory this is what happens when you start loosing power, speed and timing is off causing it to miss it's target.

There was one uses that found some recovery software he was able to use on his own. But I think he new the file system and the info needed to plug in the correct variables.

I'm glad I run UPS's on all of my equipment. I also use snapshots and do make the diaster recovery file which holds the info your looking for, and store it on 2 different servers.

[mortenm]

i have snapshots, but those i cannot access either.

i think i will try delete the first volume, the it says only can be deleted, but i cant delete it via the admin-page, because it has no name. how do i delete it via ssh?

[mortenm]

well, ive found out that it can't initialize XFS library - i will do some more research to find out what to do...

[bitor]

mortenm, let us know your progress on this problem. Thanks

Originally Posted by mortenm

well, ive found out that it can't initialize XFS library - i will do some more research to find out what to do...

[mortenm]

i have to give up on this - i had a linux-geek to look at it, but we couldnt fix it.

now i want to start all over but i cant delete my raid set, neither the volumes - can i take the hdds one by one and clear it in a pc?

what about the os?

thanks

[mortenm]

hmm all of my drives says disk failure and stands as unassigned, but it boots fine...

when i go under global spares, it says that theres no unassigned disks :S

[mortenm]

i have downloaded freenas livecd, and tried to boot from a cdrom drive connected to the one ide-channel instead of a harddrive.

i tried to do as explained in post 2 here: http://forums.procooling.com/vbb/sho...highlight=4500

but what am i supposed to do when i have set those 2 lines?
how to reboot the kernel

sorry, but i am not a linux-guy...

[blue68f100]

To get rid of the old shares that are lost requires you to use a terminal program like putty to ssh in the shell. You have to up your access rights to admin/root once logged in. Then you need to backup to the root directory. Under the root directory there will be a folder called "hd" this is where you will find vol-md0 and all sub shares. Then use the rm /"share or folder" -r . This will remove shares and all files from the snap. Use "" if you are deleting file/folders that are 2 words. the -r flag forces it to delete folder that are not empty. Remember in the Linux/Unix world it's case sensitive and is not forgiving. So make sure your right before hitting the enter key.

[mortenm]

hi blue

thanks, i will try this on monday, when i get back to work...

[mortenm]

how do i get into root directory?

when i login via putty and up to root - and write ls, nothing happens...

i have /dev/volgr0 - it is not that youre thinking of?

[mortenm]

this is what ive got:

[mortenm]

ive been in the hd-directory and ther was:

mnt1
mnt2
mnt3
mnt129

129 was the one that was weird in the webinterface så i deleted it, but id didnt make any difference...

other ideas?

[blue68f100]

Did you reboot the server for it start on a clean state?

Your mount points are different than what the GOS use for default, but that is not an issue. It's very easy to change mount points.

I will need to look to see if there is any other place that this info is stored at..

[mortenm]

hi blue

yeah, before i made that screenshot i did reboot...

[blue68f100]

Did you go back in and see if the share was still there. The reason I ask is markers are scattered in several different places. I have deleted some stuff to find it back after a reboot. If this is the case I did not provide the correct location for the shares, which worked for me. I do go through times that I would delete a file and it would be back after reboot. If this is the case you need to browse other folder for that particular share.