I hope not too often. The problem here was that many systems have been using an old configuration for ages. In the course of time, however, changes in the kernel and UDEV have resulted in the device names now being completely different than they were back then. These were changes over which the OMV project has no influence.
No idea what you mean with old configuration. I bought my N100 9 months ago. OMV fucked up that's about it.
Thanks, be interesting to know if it's a networking issue or something else
Network issues. Resetting the network interface fixed it.
Seems at lot as though all the units experiencing this issue are rpis
Is there anyone who has had this problem running other hardware?
Running OMV on a N100. Server is down since upgrading. I am going to troubleshoot in a few hours.
Let me start by saying I really appreciate all the work you have put in. My comment was not meant to criticize your work. I think you made a great plugin to get people started with Borgbackup. It could be made even better if it were made more fool proof for people like me.
I use this plugin myself and haven't had any issues after the upgrade. Since I can't reproduce it, I would need people to help.
I understand that. Unfortunately I cannot help you with finding the bug anymore. I use my NAS also for my business files, so I don't want to try it again. I hope you understand.
QuoteHave you reported your issues previously so I could fix them?
No I have not. I should have done that so I could have helped you find the bug.
QuoteOut of curiosity, what did you think it was doing? What should the plugin say? The plugin uses --encryption=keyfile and the borg manual (linked to in the plugin) is quite clear where it stores the keyfile:
The other encryption methods are interactive and would not work with the plugin. I can look at adding an export button to export the keyfile.
I think that during the creation of the repository there should be an option to download the keyfile. It should also be made extra clear that without the keyfile the backup is useless. Maybe it should also display the location where one can find the keyfile again. And maybe a button to import/export a keyfile in the repository screen.
The borg manual was to hard to understand for me in the beginning. After many hours of studying how borgbackup works, I finally understand begin to understand it.
I never realized what a keyfile was until I reinstalled my server. That was my mistake but borgbackup is kinda complicated to learn.
QuoteIn the seven plus years the plugin has been around, it has always deleted the repo when you click the Delete repo button and no one has ever mentioned deleting the repo was unexpected. I'm not against adding a button to delete just the database entries though. It just seems weird because the plugin requires deleting all archives before deleting the repo. I don't have a good way to import the archives when importing an existing repo.
I didn't expect it to delete all the files. I thought it would only delete the reference in the borg plugin in OMW. Maybe a warning that this action will also delete all the files.
QuoteThe point of OMV and the plugin are to be able to manage everything from the web interface. That flexibility does allow dangerous actions but they will not happen without the user triggering them. If the plugin didn't delete files, a user could potentially fill their drive and not know why. I can add additional warnings to the delete buttons
Yes I agree. I think the plugin could be made even better by making it more fool proof for people like me. Borgbackup is kinda hard to learn. Repositorys, keyfiles, archives, what option to choose and what not, it's a lot to digest. The manual has so many details that the manual was not really useful in the beginning.
Thank you for the answers. I enabled it on all the drives.
I had the same issue while upgrading to OMV7. The same issue even persisted while transferring the backup files to a new server. I could never fix this issue. Always End of Line. The borg commands in the terminal were working fine and giving the correct output. I never was able to fix this error. I ended up deleting the addon and manually setting up my borgbackup. It's not the first issue I was having with this addon.
Other issues I was having:
- Nowhere during the setup it is stated that the Borgaddon uses a repokey if you select encryption. So unless you manually backup your repokey, the repository is useless! I can't understand why there is no mentioning of this. If your server crashes, say bye bye to your backups. You should manually run (maybe with sudo) and save the file to a backup location!:
borg key export /path/to/repo > encrypted-key-backup- If you move to a new server you need to manually import the repokey.
borg key import /path/to/repo > encrypted-key-backup- Removing the repository in the addon really removes the repository. It deletes all the files in the repository. This should never be the default behavior. After the End of File error I tried to remove and re-add the repository in hoping to fix the issue. It managed to fix the error by deleting the entire repository. Not exactly the solution I had in mind.
I decided for myself that using this addon is to dangerous. I have to be able to rely on my backups to be safe. So now I manually learned the borg commands. No one to blame but myself if it goes wrong.
Hi,
Setup:
1 SSD
2 HDD.
I run my Debian and OMW from my SSD.
Should I enable write-cache on the SDD and/or on both HDD's? What would be the benefit of doing so?
Thanks!
What do you mean with that?
Hybrid raid was brought in by Synology in DSM 3.1, basically it creates partitions on the drive to maximise capacity and presents itself to the user as an LVM. The other advantage is you can mix and match drive sizes, could you do it using mdadm ..... possibly, but with two drives there's no redundancy.
Yes that was what I meant. But i'll try Rsync or Rsnapshot first. Looks likes a better idea.
I understand what you mean. That solution also popped up in my mind. One thing I don't quite understand yet is what the advantaged would be over partitioning or running some form of hybrid raid.
Hi,
I am running OMV on an Odroid HC4 with Armbian.
I have 2 harddrives.
1x 1 TB
1x 2 TB
I would like to have some redundancy for the 1TB to be mirrored. So that if one drives fails I always have the other. But how can I use the remaining diskspace? What raid/filesystem would I need for that that's not hard to setup? I am a little bit lost at this point.
Thank you for your time and help.
