Jump to content
Nebula

HELP PLS - Inode allocation btrees are too corrupted

13 posts in this topic Last Reply

Recommended Posts

Hello everyone,

 

I run unraid for 6-8 month and now I moved to a new house and finally got my server up and running and this happened.

 

Now Im begging for help because something bad happened with the array. OFC I got important data on the disk and thought it was copied to the parity disk as well, but it dossent seem to. Got some of the data on another HDD outside the array, but ofc it's not updatet with all data.

 

 

Attribute = 5Reallocated sectors count

Attribute = 187Reported uncorrectable errors

Attribute = 188Command time-out

Attribute = 197Current pending sector count

Attribute = 198Uncorrectable sector count

Attribute = 199UDMA CRC error rate

 

Lot of text I dont understand in the "Check Filesystem Status" The end message was this, not sounding good:


No modify flag set, skipping phase 5 Inode allocation btrees are too corrupted, skipping phases 6 and 7 Maximum metadata LSN (1:410451) is ahead of log (1:409427). Would format log to cycle 4. No modify flag set, skipping filesystem flush and exiting.

 

Please help me out if its possible, unraid server just want to format the harddrive

luftwaffe-smart-20190516-2217.zip

Share this post


Link to post
3 minutes ago, Nebula said:

I got important data on the disk

You must have another copy of any important data, preferably on another system. Parity is not a substitute for backups.

1 minute ago, Nebula said:

thought it was copied to the parity disk as well

Parity disk doesn't contain any data. Parity plus all other disks allow the data for a missing disk to be calculated.

 

That SMART report doesn't contain any useful information.

 

Go to Tools - Diagnostics and attach the complete diagnostics zip file to your next post.

Share this post


Link to post

But why does my diagnostics data not contain any SMART Attributes Data compared to others? Everything is default under disk settings only changed spindown delay.

 

I attach the "check filesystem status" on disk 1

 

Check Filesystem Status

Share this post


Link to post

Please don't try to do anything else without further instructions.

 

Disk 1 has filesystem corruption, it is disabled, and it isn't giving us any SMART information. Even worse, your parity disk needs to be replaced.

 

I notice warnings from Fix Common Problems, in particular, one saying you haven't got email notifications setup. So you probably didn't know you had a problem until it became multiple problems. And you are running a year-old version of Unraid.

 

Go to the command line and see if you can get SMART enabled on disk 1 by entering this:

smartctl -s on /dev/sdd

Then post a new diagnostic.

Share this post


Link to post

Disk1 looks OK. Unfortunately, your parity disk doesn't, and you will have to rebuild disk1, either before or after fixing the filesystem.

 

So, you have 3 serious problems. 1) Disk1 is corrupt, 2) disk1 is disabled and needs to be rebuilt, and 3) parity needs to be replaced.

 

Obviously parity can't be replaced until rebuilding disk1. I guess you could use the parity swap procedure which would allow you to replace parity first, then instead of reusing it to rebuild disk1, rebuild to a different disk. I'm not sure there is anything to be gained by that though.

 

I am going to tag @johnnie.black and see if he has any suggestions for the order in which to address these problems or any other ideas.

 

Share this post


Link to post
Posted (edited)

 

Thx man, I really mean it THANK YOU! 👍

 

Some good newz to me!

I meantime waiting for my sentences from you, I found allmost the rest of the data thx to google on multiple accounts, but still missing some.... Really have been so damn secure about this data and protected this in many years (allways with one extra hdd data outside arrraws) using synology NAS, Win10 array and and now I wanted to test UNRAID and this happens. Maybe I'm a bit of noob and not read carefully how to set things up. I must admit the learning curve is a bit harder than Synolgy/Win10 server but it serve much much much better my needs as mediaserver.

 

Its past midnight now, unfortunately I'm away the hole weekend and back again on sunday. Unfortunately I cant do anything before that, need to take care of other stuff before I go to bed. This array meltdown sets everything unhold!

 

I will return on sunday and buy some couple of new hdd as well for this rebuild. Any hdd I should stay away from?

 

Again thx for help

 

 

 

 

Edited by Nebula

Share this post


Link to post
1 hour ago, trurl said:

I guess you could use the parity swap procedure which would allow you to replace parity first, then instead of reusing it to rebuild disk1, rebuild to a different disk. I'm not sure there is anything to be gained by that though.

I'm not entirely sure if you could even get Unraid to allow this. I've never done the parity swap procedure myself. I don't know if there is a point in the procedure where you could substitute a new data disk after the parity copy or not.

 

36 minutes ago, Nebula said:

Its past midnight now, unfortunately I'm away the hole weekend and back again on sunday. Unfortunately I cant do anything before that, need to take care of other stuff before I go to bed. This array meltdown sets everything unhold!

You should probably shut down until you get back and we are ready to proceed.

 

36 minutes ago, Nebula said:

I will return on sunday and buy some couple of new hdd as well for this rebuild. Any hdd I should stay away from?

The real question it seems to me is what size disks to get. No point in going for less than 4TB for each of the 2 disks, since you are going to replace a 4TB parity and a new data disk can be as large as parity. But it might depend on what we come up with whether you can go larger with both or not. Maybe you don't want to go larger anyway. And the original disk1 can probably be reused and assigned to a new slot after everything.

Share this post


Link to post

I'll add some more thoughts as they occur to me while we wait.

 

Perhaps the filesystem on the physical disk is not actually corrupt, but the issues with the parity disk make the emulated disk unmountable.

 

If we only try for a 3TB rebuild, then that could slightly reduce the chances of hitting the pending sectors on parity.

Share this post


Link to post
Posted (edited)

Hello,

 

Server is stopped and shutdown for now. Going for 4TB disks and will buy two hdd for sunday. My question was more about the type of hdd but maybe it's not that important and they work in the same way. Some just keep telling me to use the NAS drive like WD Red but why, my hdd's don't run 24/7 so why pay more. They don't secure my data anyway if writes get corrupted.

 

I will keep an eye on this thread during the weekend.

 

 

Thanks for all input. You have been very helpful for now.

Edited by Nebula

Share this post


Link to post
Posted (edited)
8 hours ago, trurl said:

The real question it seems to me is what size disks to get. No point in going for less than 4TB for each of the 2 disks, since you are going to replace a 4TB parity and a new data disk can be as large as parity. But it might depend on what we come up with whether you can go larger with both or not. Maybe you don't want to go larger anyway. And the original disk1 can probably be reused and assigned to a new slot after everything.

 

Doesn't matter minimum 4TB but if 6 TB is better options for rebuilding data I take this instead.

Edited by Nebula

Share this post


Link to post
Posted (edited)
10 hours ago, trurl said:

I am going to tag @johnnie.black and see if he has any suggestions for the order in which to address these problems or any other ideas.

I would first try to repair filesystem on the emulated disk1, and for now make sure to keep old disk1 intact.

 

IMHO parity swap isn't really an option since if parity is really failing it will fail during the parity copy phase and abort, so if it comes to that might as well just try to replace disk1, maybe parity is not so bad despite the SMART report, and if it is the rebuild will at least continue skipping the read errors (but would obviously result in some corrupt data).

Edited by johnnie.black

Share this post


Link to post

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.