sunbound

I might look into this, and try the demo... if it sees the drive when i hook it up to my USB adapter, I might give it a go. I tried a few things the other day, and nothing would see the drive. and thanks for the "shrink array" link. I'll probably try that in the morning... after some sleep.

ok... I took all the disks that where part of the array, and mounted and shared them using "unassigned devices" 1) yes 2) at one point, all the drives (with the exception on the drive i was pulling data from) where not physically attached to the server 3) i mounted the 5 data disks (and some disks that still had data on them that where replaced with bigger dives at one point) 4) I moved the files, over the network, to USB drives on my desktop. When you say "New Config" is the way forward, I'm guessing that means assuming the worst, wiping the drives, and starting over. Is there any way to get these drives to the point that I can test and try to repair them to recover data from the individual drives? that disk 6, i placed in the disk 6 position by mistake, and haven't remembered how to remove it. tower-diagnostics-20190816-0341.zip

I took each of the disks that where in the array, and mounted them as unassigned devices. I then moved as many files off of those drives as i could, I figured i could assess what i was missing after moving everything off. Now that i have done that, I am unable to run those previous tests/fixes against the drives that where giving me issues. I wanted to try and "hedge my bets" against further losses. At this point, i would like to try and run those fixes on the problematic disks, and see what I can recover, if anything. It appears by doing this, the server things all the disks are new. Once I'm done trying to recover, I'm going to try and start everything from scratch.

All the physical disks are assigned to the same slots, and are attached to the same ports (both logical, and physical) they where previously. Everything is in the same place (I believe) that it was in a few days ago, before I started pulling files. I will post the screenshot and diagnostics when I get a chance. (Not at home right now) Thanks, as always

i actually have all the drives in the same positions they where in last time i was able to run the tests. after the last reboot, it says "stopped, invalid expression" and "You may not add new disk(s) and also remove existing disk(s)." I did throw an old disk in to see if i could recover data from it, and its showing as disk 6, missing. wondering if that has something to do with it. Also, I just noticed that the array thinks all these disks are new disks; showing blue squares next to them. when i replace a disk, i tend to just put the old disk aside, just in case i could get needed information off it.

I'm starting to think that ive hosed my system. in trying to preserve as much info as i could, i moved as much data as i could off of the drives while i mounted them as unassigned devices. Now im trying to go back and "repair" the drives, but it wont let me start the array in maintenance mode, to even run the tests on those drives.

At one point I did go though it... and systematically troubleshot the connection issues. I had narrowed it down to a power adapter from the PS that converts from MOLEX to SATA. I'm using 2 separate power adapters because the server case I DO have doesn't have enough ventilation for the drives to keep them cool. After figuring that out, I was able to get the server to see all the drives, but by then, I had already started messing with stuff, and now i'm just in damage control mode. This was before I decided to ask you guys for help... in hind-sight, I should have come here first. I am 3D printing different stack-able drive caddies for the drives, so that I can both cool them effectively, and hot-swap as needed without having to pull the server out of the closet I have it in, remove the cover, and pull the whole array just to swap out 1 drive as needed, for expansion/replacement purposes.

OK, thanks... so, because im dealing with a few different disks here... going back to disk 3... with all the "disconnected inode 6442453243, would move to lost+found" errors... i have yet to do the -Lv test. any harm in doing so? and can i run these tests, on different disks, at the same time, or is this un-advisable? Thanks

No, i didnt. I'll let it keep going... nothing to lose at this point, right? I'm assuming it will eventually give up, correct?

So, is there any way to recover the information from the 2 drives that where unable to find the "superblock". If not, I'm thinking the best course of action, for me is to migrate all the salvageable information to external drives, and start all over from scratch. Am i right? Thanks

Thanks for the input... I guess its s good thing i'm just a "hobbyist" with a "digital pack rat" mentality. I'm glad it wasn't corporate stuff this server was working with. I was looking at online backup stuff... will backblaze (not B2) work with this? like i said... its just family and private stuff... movies, pictures, documents, software, etc...

yes, that is correct. I tried mounting it via USB, but the server was unable to see it, so i pulled disk one from its position, and put the original disk 4 in its place... for now

I'm wondering what the safest way to move these files from this drive is. Since i can mount and share the drive, should i just move them off the drives onto USB drives, and just migrate it back once i have it all on externals? My dumb behind procrastinated on a backup plan, and now i guess i'm paying the price. Either way, i'll be looking for an affordable online backup soon...

I just tried the same test on disk 5, and im running into the same error as i did on 4... Phase 1 - find and verify superblock... couldn't verify primary superblock - not enough secondary superblocks with matching geometry !!! attempting to find secondary superblock... ................................................................................................................ "not looking good, is it, Doc?" i had to remove my "disk one", to make room for the other disk, and i can view the file structure. SMART report attached. tower-smart-20190811-0710.zip

OK, so i started the -nv check on disk 4 (the first problematic disk, and one that was rebuilt from the parity drive, using a new disk), and its not looking good... Phase 1 - find and verify superblock... couldn't verify primary superblock - not enough secondary superblocks with matching geometry !!! attempting to find secondary superblock...................... its been stuck on this for the past few minutes... should i place the original (disabled) drive back into the array, and try this test again?