space issues

[falcomm6]

Hi guys,

I love the new OmniStream, im just having some continued issues with space for some reason. I thought it was due to misconfigured sabnzbd, but now ive tried migrating from gooby with nzbget and still having it.

Ive tracked down a lot of data being saved in var. Im not good at linux but i was able to track a good amount of data in var at the below directory. Ive had 500gb full disk twice in 2 weeks, where i never had this issue with Gooby. Any tips?

it doesnt let me cd to docker from this location to dig deeper for some reason.

[kelinger]

This may be an easy one: we're not storing anything specifically for OmniStream's code in /var. But, I'm sure Docker is. If you don't regularly run omni update then Docker will be keeping older copies of containers around (eg, you're running Sonarr and a newer container comes out, that will be downloaded and installed but the previous container remains locally cached by design). Some commands may help. Note that there is no need to run more than one of these:

omni update - updates the code base, downloads any newer containers that weren't previously cached, swaps out the old for the new, and then purges all non-used containers. If there are a lot, this may seem to hang for a minute at the end of the update but trust me... you'll see a bunch of data scroll by and a summary of how much space was reclaimed when it's finished.

omni rebuild - same as above without the update portion. This was written with the intention of being used when you add or remove components from a running system. For example, if you enable SABnzbd and disable NZBget, the changes to the config are made immediately but no containers are stopped/started in the process. The rebuild causes OmniStream to reparse the configurations, see which containers need to be taken down or started up and then it cleans out anything leftover, including the cached containers. Without making any changes, the configs would be the same so this, in effect, simply runs the purge process to pick up the trash.

omni clean - very brute force. This takes down everything and then purges all containers--old and new. It does not delete any of your configs, data, personal shares, etc. but basically leaves you with a clean system and OmniStream still installed but not running. Like the above, it will display the amount of space reclaimed. Unlike the others, you will need to start OmniStream when finished (if you wish) and this will redownload all containers that are enabled. I want to reiterate that this will not destroy any personalization or customization but it should be followed by omni up when you're ready for things to start running again.

[kelinger]

Additional notes:

Depending on what you're running, some container publishers have daily updates or even multiple updates per day. The "watchtower" container keeps an eye on these and predownloads them so the next omni update can just swap out the old and replace it with the new with minimal downtime (in most cases, the swap is a "blimp" and only noticeable if you're actively using the specific container--or a dependency--when it is switched over).

I personally give my system an omni update everytime I log in just to a) make sure I'm on the latest and greatest and b) clear out the previously used containers. It usually only takes a minute and I can control when I do it (ie, not while a spouse is watching a movie on Plex).

[falcomm6]

Thanks for this. Looks to still be 100GB, not entirely sure whats going on. Maybe im saving something somewhere I shouldnt.

Do you know how I can navigate to the docker folder? right now it says I cannot cd there, perhaps its mounted and lives elsewhere. Curious to see exactly whats happening. it says command cd not found - even when I sudo. Never had this before.

Thanks for the tip though. Nice to reclaim 5Gb each login

[kelinger]

You probably need to use sudo.

sudo -s will start a shell essentially as root. Then you can cd anywhere you want and you don't need to use sudo again for things like du when you try to find the culprit. When done, type exit to return to your peon-level non-root login 😄

[falcomm6]

LOL. I definitely need to stay peon level as much as possible. Here is what im getting. I tried -s and it didnt help me either.

[falcomm6]

NVM I got it now. Thank you so much for helping this peon! lol

[falcomm6]

Is this normal? All the data is in overlay

Last question I promise! thank you so much, again.

[falcomm6]

I cleaned a LOT using something I foung on Google. Below deleted the volumed which were not running. Extra 100gb. Time will tell if I broke something lol.

docker volume rm $(docker volume ls -qf dangling=true)

Thanks!