On Mon, Aug 22, 2022, at 7:41 PM, George R Goffe via test wrote:
> Howdy,
>
> Thanks to all who responded...
>
> Chris,
>
> I tried the superblock mount but that still fails. Then I tried the 
> rescue=all mount. That succeeded... I have copied all the files I need. 
> From what I read, corrupted files get copied too... and there were some 
> which I removed from the copies.

It does disable data checksum verification, so it's possible. It's a valid 
option to use the individual option method, leaving data checksumming in place, 
e.g.

mount -o ro,rescue=usebackuproot,nologreplay,ibadroots

in this case it will complain about corrupt file blocks and won't let them be 
copied out. How this gets handled depends on the application - some 
applications stop at the first bad file. Others continue reading the rest of 
the same file, then stop. Still others continue reading all file blocks.



> From 
> "https://lists.fedoraproject.org/archives/search?mlist=users%40lists.fedoraproject.org&q=parent+transid+verify+failed";
>  
> I found some other commands that you wrote. I presume after unmounting 
> this problematic filesystem. Is it time to run them?

Try "mount -o rescue=usebackuproot" this is also used by rescue=all, but 
permits rw mount. If it works, the current damaged root tree will be replaced 
by a good one.


> btrfs rescue zero-log 
> btrfs check --repair --init-extent-tree 
> btrfs check --repair

I don't think any of these are indicated for this problem. 

Zero log is harmless, but the messages don't indicate a problem related to the 
tree log.

Init extent tree is taking a big chance It might work, it might make things 
worse. And it'll take a while. I'd just take advantage of the ro rescue mount 
to get your data out such as it is. And then mkfs time. Wipe it and clean 
install, restore from backups. It's the more reliable way of getting back to 
good.

But it's useful to look at logs over the last couple days to see if there's 
prior evidence of problems and what they might be. Because it sounds to me like 
the hardware has become unreliable in some way.


-- 
Chris Murphy
_______________________________________________
test mailing list -- test@lists.fedoraproject.org
To unsubscribe send an email to test-le...@lists.fedoraproject.org
Fedora Code of Conduct: 
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: 
https://lists.fedoraproject.org/archives/list/test@lists.fedoraproject.org
Do not reply to spam, report it: 
https://pagure.io/fedora-infrastructure/new_issue

Reply via email to