How closely is this related to my issue?

 

http://lime-technology.com/forum/index.php?topic=27450.15

One and the same.

Ok try this.  Make a list of your data disks that contain the top-level directory "unsorted".  For each one list the xattrs, for example, let's say 'unsorted' is on disk1, disk3 and disk7, you would type this:

 

getfattr -d /mnt/disk1/unsorted
getfattr -d /mnt/disk3/unsorted
getfattr -d /mnt/disk7/unsorted

 

here is the output (nothing for disk2):

root@UNRAID:~# getfattr -d /mnt/disk1/unsorted/
/mnt/disk1/unsorted/: user.org.netatalk.supports-eas.4VNO7Q: Input/output error
root@UNRAID:~# getfattr -d /mnt/disk2/unsorted/
root@UNRAID:~#

 

please let me know if there is anything else i can do to help

Ok try this.  Make a list of your data disks that contain the top-level directory "unsorted".  For each one list the xattrs, for example, let's say 'unsorted' is on disk1, disk3 and disk7, you would type this:

 

getfattr -d /mnt/disk1/unsorted
getfattr -d /mnt/disk3/unsorted
getfattr -d /mnt/disk7/unsorted

 

here is the output (nothing for disk2):

root@UNRAID:~# getfattr -d /mnt/disk1/unsorted/
/mnt/disk1/unsorted/: user.org.netatalk.supports-eas.4VNO7Q: Input/output error
root@UNRAID:~# getfattr -d /mnt/disk2/unsorted/
root@UNRAID:~#

 

please let me know if there is anything else i can do to help

It looks like there is some kind of corruption with that extended attribute on that particular directory.  I think you should be able to fix it like this:

 

First, goto Settings/AFP and disable AFP

 

Next type this sequence of commands:

 

user nobody
mkdir /mnt/disk1/unsorted-
mv /mnt/disk1/unsorted/* /mnt/disk1/unsorted-
mv /mnt/disk1/unsorted/.* /mnt/disk1/unsorted-         [this to move files starting with '.' - it will produce error msg - thats ok]
rmdir /mnt/disk1/unsorted
mv /mnt/disk1/unsorted- /mnt/disk1/unsorted

 

Now you can re-enable AFP, and everything should be ok.

 

You see what we're doing here: we're creating a temp directory, moving all the files from bad directory to temp directory, deleting the bad directory, then renaming the temp directory to what was the bad directory's  name.

 

The reason the 'mover' failed was because 'rsync' was not able to make the extended attributes of the destination match those of the source, so it errored-out, leaving the original source on the cache drive.

To everyone else, let Tom work this out with joyless first, then he will let the rest know.

First, goto Settings/AFP and disable AFP

 

Next type this sequence of commands:

 

user nobody
mkdir /mnt/disk1/unsorted-
mv /mnt/disk1/unsorted/* /mnt/disk1/unsorted-
mv /mnt/disk1/unsorted/.* /mnt/disk1/unsorted-         [this to move files starting with '.' - it will produce error msg - thats ok]
rmdir /mnt/disk1/unsorted
mv /mnt/disk1/unsorted- /mnt/disk1/unsorted

 

Now you can re-enable AFP, and everything should be ok.

 

You see what we're doing here: we're creating a temp directory, moving all the files from bad directory to temp directory, deleting the bad directory, then renaming the temp directory to what was the bad directory's  name.

 

The reason the 'mover' failed was because 'rsync' was not able to make the extended attributes of the destination match those of the source, so it errored-out, leaving the original source on the cache drive.

yes, this indeed fix the issue. thanks!

yes, this indeed fix the issue. thanks!

I'm going to mark this "solved" but I can't explain how the corruption occurred in the first place.  There is a newer version of netatalk that I will be updating after 5.0 'final'.

To fix my issue, would I follow the same steps? Or is this a unique solution for joyless?

 

Thanks!

To fix my issue, would I follow the same steps? Or is this a unique solution for joyless?

 

Thanks!

The method is generic, the paths in the command are specific. Substitute your paths. If you don't understand what I just said, then post a syslog with your mover errors and someone can tell you exactly what to type.

That makes sense. However, I am unclear on which disks I should run this on.

 

I have five different disks with shares spread across multiple disks. Will I need to run this on each share on each drive?

I have five different disks with shares spread across multiple disks. Will I need to run this on each share on each drive?

I don't have the issue, and am only following it peripherally, but it's my understanding so far that only certain shares were corrupt for most people, so it's only needed on the root folders that correspond to those shares on each disk. Do you have a list of errors in your syslog that tell you which shares are having problems?

That makes sense. However, I am unclear on which disks I should run this on.

 

I have five different disks with shares spread across multiple disks. Will I need to run this on each share on each drive?

Please move discussion here:

http://lime-technology.com/forum/index.php?topic=27450.30