Originally posted under the wrong heading (v6) due to bad eyesight.

 

Can anyone please advise with this problem as I can't find a thread or guidance on this.

unRAID 5.0.5 Pro, WAS working fine. Two main shares Movies and Music across 2 disks using Most Free allocation.

In between rips, with no other activity going on from elsewhere, decided on a whim to switch the 'Movies' share back to 'FillUp' from 'Most Free' so as to force unRAID to use up a large chunk of unused space on disk1.

Since then, Movies share refuses to browse from the GUI, whereas Music (unchanged) still works fine. Have tried starting and stopping the array, but to no avail.

My IE browser says 'The page can't be displayed'

Have tried rebooting, no difference. Have tried changing Allocation back to Most Free.

I can inspect the data from disk 1 to folder 'Movies' and it's contents are fine. Similarly with Disk2/Movies, all fine.

Note that if I browse to /mnt/user/ and try and open /Movies I also get 'The page can't be displayed' behaviour

With Firefox I get 'The connection to the server was reset while the page was loading'

Seems like the folders on the disks are fine but the share functionality across multiple disks is 'broken'

Note that the 'broken' share is still accessible from Windows via samba BUT if I try and open the share from WebGUI (to list what's on each disk) while it is being accessed (ie during a rip) it resets the share and the rip fails.

As I said, the share I didn't change (allocation method of) is completely unaffected.

Can anyone please advise how to fix this.

Thanks

Not sure what your problem is, but the allocation method will be a red-herring as it never affects access, just where new files get created.

 

I would suggest your provide a syslog that covers the period where you are seeing the problem.  That should show if there are any underlying errors happening.

Update - turns out .. this has nothing to do with having changed the Allocation Method, that was purely co-incidental

In the process if copying syslog I noticed lots of filename errors on the console BUT the filename did not match the name of the file on disk, ie IT WAS MISSING A CHARACTER.

This led me to suspect an invalid character, and I was right.

This problem was again due to invalid characters in a filename.

The same effect was/is caused by an apostrophe.

I though it was something I'd done but infact I just hadn't browsed the files on disks via WebGUI for a while, so the 'share' which appeared to be 'broken' was in this context a problem with the WebGUI navigation function of unRAID, which was 'broken' by an invalid character in the name of one folder.

The offending char in this case was like a full stop at half height, ie like the top half of a 'colon', which was present in pace of a dash (-).

How, why, who knows, but at least the cause is indentified and the problem now resolved.

To fix it I needed to map and browse the disks individually and look for the folder which the console was creating errors about; find the offending folder and rename it.

PROBLEM SOLVED.

All stable and working again. 

Hopefully this will get fixed in v6