zephyr.1.2020

Hope it gets worked out. I assumed my config was breaking due to my customizations over the years. So starting over from a fresh install seams to have been it for me.

Sorry for the delay in responding @andreidelait I ended up starting over from scratch pretty much. I changed the port numbers for the original docker containers(mariadb, nextcloud, swag) and started over with the most recent nextcloud video from Space Invader One. This time, instead of using Duck DNS, I decided to buy a domain name off of name cheap($10/year ish) and try again. This time it seams to be working. Once past the initial install, I copied over the original data from the Nextcloud1 share into the Nextcloud2 share. Once the files were transferred, I had nextcloud rescan them by running "occ files:scan --all" That's gotten me going again with a basic install. Still have a few things to tweak, mainly getting it able to upload and download larger files. But if memory serves, I had these same issues with the last time I set it up and was able to find a solution so I'm sure it will get figured out eventually. I would like to be able to get document editing figured out but still no luck with that.

How is that going for you? I've not tried it yet my self but am considering it or a VM route. There are so many ways to do it, it's a bit overwhelming haha. My initial attempts at a VM failed, but may be worth giving it another go this afternoon.

Still no luck. Added .conf to the nextcloud file name. Rebooting both swag and nextcloud got me back into the nextcloud interface for a wile. Poked around some random files I've not accessed in a while, could use them as normal. Phone app said unrecognized server name.....So I decided to reset it's config from scratch. Error came back with that but I then remembered having an issue with the phone communicating over the LAN to the server, so dropped to LTE and the error went away and I could access random files. But about 5 minutes into it, the original issue came back. CPU usage racked up to about 70%-80% and lost all control of nextcloud and unraid's ui. Any help is greatly appreciated!

@andreidelait are you talking about the config file located at /user/appdata/swag/nginx/site-configs? I have 3 files in this directory. default.conf, nextcloud, and default.conf.sample. I changed nextcloud to nextcloud.conf and no luck. Were you talking about a different file?

Ah. didn't think to try to send it twice. that is good to know.

Over this last few weeks, something is causing Unraid's webui to hang and the only way I've been able to get it back is force a reboot with the power switch. I think it could be caused by a broken config in nextcloud because I can recreate the issue when I navigate to nextcloud's webui, and search for an update. Shortly after launching everything just goes in limbo. I tried stopping the following dockers when this happens and it fails. Nextcloud, Swag, Plex. But I could stop sonarr......I just forced a reboot and captured the diagnostics. If I need to capture while the error is happening, please instruct me on how to do that. I do have SSH ability while it's hung but trying to force a reboot over ssh didn't work(or I was too impatient) so I don't know if that's working correctly when this happens either. Thanks in advance for the help! zephyr-1-diagnostics-20220930-0635.zip

I've been using Rsync running on my backup Unraid server to pull files from my home Unraid server for an offsite backup. I mount the root SMB share on home server over a tailscale vpn with a script, a second script triggers a list of rsync operations to run, backing up the files, then unmounts the share. For the longest I thought this was a rock solid backup strategy then I realized some media files weren't backing up. After some trial and error, I was able to narrow it down to only media files with 'special' characters in the name that wouldn't be transferred. For example: it doesn't like the é in Pokémon Detective Pikachu but Pokemon Detective Pikachu is fine, it doesn't like the dots and ? in What if...? but if the folder/file is named What if it all works. I was hoping I could use Sonarr/Radarr to name these different to avoid it at the source but no luck. Then I was hoping something could be added to the rsync syntax since I found a few forums mention something about iconv but I've not been able to get that working either. Any help will be tremendously appreciated. If you know of another easy way to schedule routine backups, I'm all ears! Thanks in advance.

That led me to the solution. Instead of being on the landing page after logging in, the setting I needed was on the individual tv show itself. I dove into using plex's live TV/DVR option a month or so ago and I thought the option to keep only x number of unwatched shows deleted what was recorded, apparently it deleted the previous files I imported and had already watched as well. Good to know. Thanks for all the help Jonathan!

Update: I've been able to confirm that Plex is somehow the culprit. When I add a season folder, then force it to run it's scheduled tasks, it deletes them. I see under the alerts section of plex that it deletes the media. I disabled all scheduled tasks that could cause deletion. Unchecked the options to empty trash after each scan, and the one that allows media deletion. But when I force it to run scheduled tasks, it still deletes the media.... Any ideas? Would making plex read only solve the issue? if so, how would I go about doing that?

I will try that and report. Thanks!

Well I think I can rule out issues with the disk. I used the unbalanced plugin to move all of the show to a newer disk, confirmed that the files were written to the new disk and were playable. Checked this morning and all but the current season was deleted again. Any ideas?

Correct. I first noticed something odd when I was running a rsync to my backup server and it flagged a warning that some files vanished. I'm fairly sure I had the TV share set to fill up each disk. So it stands to reason each time I would write the show back, it would go to the same disk. Does this seam logical that it could be a hardware issue?

Is this what you're referring too?

Here's my syslogzephyr-1-diagnostics-20211207-0900.zip