On Wednesday 01 March 2017 17:32:35 Qu Wenruo wrote:
> At 03/01/2017 04:23 PM, Marc Joliet wrote:
> > On Tuesday 28 February 2017 23:14:54 Marc Joliet wrote:
> >> I think I'm at that point now myself, unless anybody has any other ideas.
> > 
> > For example, could the --init-extent-tree option to btrfs-check help,
> > given
> > that I needed to pass -w to btrfs-image?

(First off, thanks for taking the time to respond!)

> --init-extent-tree should be avoided normally, unless you're 100% sure
> about only extent tree is corrupted, and you have enough faith that
> --init-extent-tree can finish.
> 
> Or it will mostly make things worse.

OK, that's why I asked :) .

> Before trying any RW btrfs check operations, did you run btrfs check on
> the image?

If by image you mean the device in question, not since the last time (see my 
previous messages).  Or can I really run btrfs check on the image?  In any 
case, I started btrfs-check on the device itself.

> And btrfs check --mode=lowmem is also recommended as in some rare case,
> low mem mode can detect bug which original mode doesn't.

I did see differences in output the last time around (again, see my previous 
messages in this thread), so I'll run with lowmem.  It won't be done until 
tomorrow, though.

> (And it also helps us to enhance lowmem mode)

OK

> > Also, the image is complete, so I only need to find somewhere where I can
> > upload a 9.4 GB file.
> 
> Is it a compressed dump? Dumped with btrfs-image -c9?

It was created with:

btrfs-image -s -w /dev/sdb2 - | xz -9 --stdout > ./btrfs_backup_drive_2.img.xz

(Mainly because I felt more comfortable using a separate compression utility, 
not for any rational reason.  Although if you really meant "image" above, I 
have the feeling I'll regret this decision.)

> If so, such large one won't help much unless there is some developer
> really interested in inspecting the image.

When I started without compression, it was something like 46 GB, after only 
1-2 hours, so I expect the uncompressed size to be... very large :-/ .

> Thanks,
> Qu
> 
> > Greetings
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Greetings
-- 
Marc Joliet
--
"People who think they know everything really annoy those of us who know we
don't" - Bjarne Stroustrup

Attachment: signature.asc
Description: This is a digitally signed message part.

Reply via email to