dlandon

My advice is to put the ownCloud /config on an SSD cache share. Just putting the /data on disk 1 probably won't accomplish what you want depending on the activity on ownCloud. Any /data file activity would spin up the disk. i.e. uploading, downloading, or viewing files. Put the ownCloud /config on a SSD cache, and back it up regularly to the array. ownCloud has a plugin in the marketplace that does a database backup regularly, and you can safely copy the /data folder to the array with a User Script, even with ownCloud running. You will also find that ownCloud will be more responsive on a SSD share.

Are you having any other issues with other shares or files not purging out the aged files? Run this command and see if there are any errors. /usr/local/emhttp/plugins/recycle.bin/scripts/rc.recycle.bin purge I wouldn't expect to see any errors. The other thing you can check are the dates of the files. Some time ago the aging determination was changed and these files might have an issue with the dates. 2017.05.07 Add: Changed the recycle bin aging to the last access date and don't change the modify date when the file is put in the recycle bin.

From what I can see, your shares are all on disk1. You're mapping directly to the disk and not the share. Don't do that, even if you have shared the disk1. Do it this way. Create a share called 'owncloud'. Set it to use disk1 only. Then map the config folder of ownCloud to '/mnt/user/owncloud'. The /data folder will then be in this share and all of the /config will be located in the 'owncloud' share on disk1. Always refer to shares by the mount point '/mnt/user/share'. Never map to disks directly. You will create problems!

Post your diagnostics. I think your issue might be the way you've set up your shares.

As far as I know these have been addressed. There was a change in the default timeout on a CIFS mount that was shortened from the original default of 120 seconds. I adjusted the mount timeout to 120 seconds and that should have taken care of the timeout messages in the log. If anyone is having issues, please post a diagnostics so I can have a look.

Looks to me like you have a network or server issues. They are going off-line. Apr 1 15:01:00 Moria kernel: CIFS VFS: Server NAS02 has not responded in 120 seconds. Reconnecting... Apr 1 15:01:00 Moria kernel: CIFS VFS: Server NAS03 has not responded in 120 seconds. Reconnecting... Apr 1 15:01:02 Moria kernel: CIFS VFS: Server NAS01 has not responded in 120 seconds. Reconnecting... This is at the CIFS level. My guess is a network issue.

You'll have to post diagnostics so someone can help you. Whenever you post an issue, add your diagnostics so I don't have to waste time asking.

This is where the username and password are entered. Obviously you aren't entering a username. Wait for a bit and I'll issue an update to the plugin.

What username are you entering?

It appears a device was removed or dropped off-line without being unmounted.

Ok, read the log message: Error message: username specified with no parameter Remove the remote SMB share and re-enter it. There is an error with the credentials.

Try the update. Should fix it. I was using .pl, not .pm in the copy.

It looks like the addition of the domain to the user with user@domain is causing too many problems. If a share contains a '@', UD thinks that there is a domain. I've added a domain line to the add SMB share dialog to define the share and domain separately. This won't affect any current remote shares already defined, it will just be different when an SMB share is added new.

Looks to be a problem with the copy. What is the extension of the control file? .pm?

As requested, post diagnostics. Reporting that something is broken without diagnostics only makes me have to ask and waste time.

You can ABSOLUTELY cut the DRAMA and give me some information. Just posting that it's broken has NO value and does NOTHING to get you an answer. Whenever you post a problem always attach your diagnostics. Please provide the following: - Post diagnostics. - Does your share name have a '@' character in it? - Is it a Windows share?

I don't think the Rx and Tx ring buffers will be a limit on your throughput. It kind of depends on the NIC though. I've read where to large a setting on the buffers can actually be counter productive, but I am not an expert on NICs.

I'll add it as a choice.

You have a database issue. Are you using the built in mysql database or an external database? Making a new container won't fix the issue.

The differences in the mounts is the correct way to mount SMB v1 and SMB v2/v3. If the one server doesn't have SMB v1 enabled, it won't mount with SMB v1 being forced. Windows 10 has SMB v1 turned off by default and you have to specifically enable it. Your server may have the same issue. SMB v1 is on its way out and it doesn't make sense to invest a lot of time in it. I expect that SMB v1 will eventually be removed from samba because of the security issues. You need to upgrade the old server, or replace it.

Good for you. I’m sur I’ll hear about it if there is a problem.

Did the update fix the problem?

In the future, you can get more help by posting your diagnostics.

Windows works that way. You have to enter the user login credentials just to get a listing of the shares in UD.

Screen shot please. I think there may be an issue with parsing the user and domain. Also, copy the contents of the /flash/config/plugins/unassigned.devices/samba_mount.cfg so I can see how the user and domain were parsed.