denishay

One could argue that syncing is not backup and not meant to protect against deletions or corruption.

Open a console (either from main unraid dashboard or SSH session) and type: docker exec -it nextcloud sudo -u abc php /config/www/nextcloud/occ files:scan it will run the OCC command in the nextcloud docker with the abc username and scan for any missing files. Not sure why, but someone found funny to use a completely non-standard name as Nextcloud data owner... Hoping it saves you countless hours of research.

This was discussed already in this thread if you search. You can edit the config file to disable de-duplicaiton and presto, Crashplan will use your max upload... With a nice upload like you have, it's way faster than having Crashplan trying to guess what is necessary or not to upload, hence reducing the upload strongly. Edit: you can also see that here: https://support.code42.com/CrashPlan/4/Configuring/Unsupported_changes_to_CrashPlan_de-duplication_settings

Nope, an single SSD. 240GB, nothing extraordinary. Filesystem 1B-blocks Used Available Use% Mounted on /dev/sdc1 239947935744 97084276736 142863659008 41% /mnt/cache /dev/md1 1999422144512 1214729420800 784692723712 61% /mnt/disk1 /dev/md2 999716474880 247712878592 752003596288 25% /mnt/disk2 /dev/md3 2999127797760 1427599224832 1571528572928 48% /mnt/disk3 /dev/md4 2999127797760 2593694089216 405433708544 87% /mnt/disk4 /dev/md5 3998833471488 3454259048448 544574423040 87% /mnt/disk5 /dev/md6 3998833471488 2016894824448 1981938647040 51% /mnt/disk6 rootfs 4140683264 691458048 3449225216 17% /mnt shfs 17235009093632 11051973763072 6183035330560 65% /mnt/user shfs 16995061157888 10954889486336 6040171671552 65% /mnt/user0 Unfortunately, in between, I had completely uninstalled and reinstalled it in hope of "fixing" what I thought was an issue. So the log is now empty.

Thanks a million! That was indeed the case. It was my nextcloud share, and somehow, between the trashbin and a few redundant folders, I had more than 5.1 TB on this share! It's weird Unbalance offered the cache drive as destination then, or didn't warn me there wasn't enough space. But thanks! Mystery solved now!

Hi all,  I'm puzzled. I just added a new drive to my array (formatted it, etc.) and I can move files to/from it just fine in a shell using mc, but somehow, Unbalance is completely unable to offer me this brand new emtpy 4TB drive as a destination for a "gather" operation. I would like to move all files pertaining to a share to this new drive, but alas, nothing but my cache drive is offered as a destination (the other drives are quite full, up to the 512MB set as a limit).  Of course, I tried stopping/restarting unbalance, the array… I even rebooted the whole server, but no luck in getting this new drive as a destination. How can I "force" that as a destination in a gather operation?

Ahh... that makes a lot more sense now ! thanks! Indeed. Probably I got confused, but probably the message wasn't extremely clear either Who knows…

Today, I got very surprised to see The Fix Common Problems plugin giving me the warning below: The statistics plugin is installed, but the preclear disk plugin is not installed. There is no reason to have statistics installed but not Preclear. It is recommended to uninstall statistics here [followed by a button to access the plugins page] Really? My initial reaction was of the WTF?? kind. Those are two very nice and useful plugins, but they have nothing to do with one another. Can we honestly state that there is "no reason to have Statistics installed but not Preclear"? Really? I don't plan to add drives anytime soon. And if I do, I'll simply add them to the array. I do not necessarily *need* to format them outside. Plus, the Stats have nothing to do with Preclear. Again, both are nice, but I feel that this kind of "forced" recommendation (what the heck, recommend uninstalling something I use??) is counterproductive and does not help people in doubt. It just throws in more confusion and disinformation. This has nothing to do as a warning.

Instead of "simply" I should have written "in a simple manner". The new UI is more than unintuitive. Instead of having a nice list of all top-level folders you have selected, you are forced to navigate deep into the directory tree (instead of list of shares) under root/storage instead of seeing as a root only your list of shares. So can you do it? Yes? Was it way simpler before? YES! As a side note: thanks a bunch for your work on this Crashplan Pro container and many contributions here. Without all that extremely great work you do, I would have probably chosen some other mainstream solution for my NAS like a Synology, but now, with every day passing, I am happier and happier to see everything my unraid server can do! So Djoss, you have all my admiration and sympathy!

Same here, switched to Djoss' Crashplan pro container and it solved the issues. All was very well explained in the link on how to retain your existing backup (it took me approx. 4 hours to check my 4.1 TB currently backed up). … and I also hate the "new" UI, which doesn't allow you to simply select what you want to back up. You're better off choosing that from Crashplan's website.

It could have been the Pro version overwriting your my.services.xml file during an update (see my post above).

Hi guys, It took me a while to figure out why Crashplan wasn't maxing (or at least near-maxing) my upload capacity. The thing is that Crashplan seem to have been developed with much lower connections in mind than the ones we have today. • Stop the Crashplan docker • Go to your appdata folder, navigate to your Crahsplan folder • Edit the "my.service.xml" file in the "conf" folder And change the following value to 1: <dataDeDupAutoMaxFileSizeForWan>1</dataDeDupAutoMaxFileSizeForWan> • This will effectively disable the data de-duplication for any file larger than 1 byte for any backup done over the internet • Start your Crashplan docker again This always had me max my connection again. Otherwise, I rarely went beyond 2 MB/s. (currently happily going over 25 MB/s with that) Kind regards, Denis

Indeed, you need to enable guest access. But, if you cannot or do not want, with Windows 10, you need to have the same username and password created on unraid. That shouldn't be a problem if you are using a local account, but given how Microsoft has been forcefully pushing their Microsoft accounts, if you made the mistake of going with a Microsoft account for login in to Windows, you can also simply create on Unraid a user with your MS email and password. That works too. for those who have enabled guest access, the very first time you will try to access Unraid shares after that, enter ".\Guest" as username and leave the password field blank. Check the "Remember credentials" option if wanted.

Have a look in the network settings of unRAID. Make sure enable bonding is set to no. (not my idea, taken from https://www.youtube.com/watch?v=I58LTMKyeYw&t=272s)

8GB DDR3. Running typically 2 or 3 dockers and 1 Windows 10 VM. never had any memory issues.

Same here. Would gladly pay to have that back. Sent from my ONEPLUS A3003 using Tapatalk