crafty35a

I just yesterday noticed that there was a duplicate entry for this plugin in CA and my previous install was not updating (was stuck on 2023.12.19). I got the new version, and was very happy to see some of the new options (especially free down/prime up)! However, I've noticed some issues: It seems like the ignore list option is not working (files listed inside the ignore file are being moved even with this option enabled). It did work for me until I migrated to the new version of the plugin. With the ignore list enabled, more data is moved than when this option is disabled. The difference seems to be roughly equal to the total size of the files in the ignore list. The percent used calculation seems to differ from what is shown in the Unraid GUI. The GUI shows 88.7% full (1.71 TB/1.9TB), while the plugin shows 85% used (the logs say 1.6 TiB / 1.8 TiB). This is a mirrored ZFS cache pool of 2 devices. Not a huge deal if the difference is consistent, I just don't want to risk running out of space if the difference between the two free space calculations were to grow. You can see my options in the screenshot below. Pretty simple, I just want to trigger the mover at 85% and move files until the Cache is down to 80% full, while ignoring certain files.

Great idea with this, I was looking for exactly this behavior in the Mover Tuning Plugin! Have you considered allowing us to specify a desired amount of free space to maintain, rather than specifying the size we want to TARGET_DIR to be? Unless I am missing something, it would be pretty easy to calculate the MAX_SIZE from the user specified free space requirement: (total size of cache pool - desired free space) - ((total current size of data in /mnt/cache) - (total current size of data in TARGET_DIR ))

Read the thread, there are issues right now with this plugin on 6.12

Thank you, but I don't see that warning message anywhere (and also I have 6 CPU cores)

I feel like I must be missing something obvious, but I can't seem to select my SSD cache drives for testing. They are listed, but grayed out/unselectable. I did add the path value specified in the FAQ.

This is suddenly happening to me as well. If I toggle the tunnel a few times, it will eventually handshake. But even then, the tunnel does not show up as a network for docker containers (I have "VPN tunneled access for docker" selected), so it doesn't appear to actually be working. Edit- I imported a second tunnel, and suddenly both appear as available networks for containers. But they still don't seem to work at all, I can't get any traffic over the tunnels. Edit2- I was able to fix this after downgrading from 6.12 RC2 to stable. Oddly, it took multiple reboots to start working. I was assuming this was the same exact issue as OP, but maybe not since I was on 6.12.

I can confirm, there are a couple issues regarding sort order on 6.12 RC2 with Docker Folders. Containers inside a folder cannot be re-ordered. The orange up/down icon appears next to the container, but the containers are not draggable. Containers outside of a folder can still be dragged. Re-ordering folders themselves does appear to work, but the order does not stick. After a reload, the previous order returns.

The instructions I had followed were to map the /netdata/dbengine subfolder, not the entire netdata folder -- I will try it the way you show here and let you know if I lose history again tonight. Is this the only change you made from the template defaults? Also, did you make the "delete obsolete charts files" change or was that not necessary? Edit - I'm not quite sure how to set this up, the container won't start with this setting ('netdata ERROR : MAIN : Failed to initialize database at /var/cache/netdata/netdata-meta.db, due to "unable to open database file'). Guessing I need to copy the db over manually before starting, but I'm not clear on how to do so

I tried that back in April when I first started using Netdata and ran into this issue, and it definitely had no effect. I just changed that setting again and switched back from the stable tag to netdata/netdata to force an update, and lost my history. I can check to see if history is cleared again for the next pushed update as well, but I'm pretty sure it will be.

I'm aware of that, but it sounded to me like that was referring the visual aberrations (e.g. the misaligned separator). Either way, I hadn't seen anyone else report this so it seems worth mentioning.

Minor issue - on the dashboard, the "All apps/Started only" toggle seems to be reversed for containers within a folder. It works properly for containers that are not in a folder, and for folders themselves.

Understandable - thanks for taking the time to get this plugin up to speed, it really is crucial to a lot of us.

Since updating to the fork, folders do not show up on my dashboard anymore - just the standard big list of containers. Anyone else? I initially updated directly. But then I removed the plugin, rebooted, reinstalled the plugin and created a new folder - same result, folders do not show up on the dashboard.

This is awesome! The only thing is we now lost the ability to copy the current path to clipboard (very useful when simultaneously using a terminal). Would you consider using a bold / character, instead of the stylized chevron that you now use?

Mine all look correct. But I notice that your containers are showing different numbers for total RAM available. The 1.78Gib total you are seeing is the sum of the two containers that show 8GiB total RAM. Are you somehow limiting the RAM available to some containers? if so, that's probably what's causing the numbers to not be included in the total.

Just a quick suggestion - it would be really cool if you could make the full path at the top function as a clickable breadcrumb style navigation (like Windows Explorer, Krusader, etc). Clicking "Parent Directory" repeatedly gets pretty tedious!

How do I get glances to show the full command, like htop? e.g. by default it is just showing "python3" instead of "python3 -m glances /glances/conf/glances.conf -w" It looks like glances by default should be showing the full command line, but it doesn't when I use this container.

What is the proper way to make my chart data persistent? I already bound the dbengine folder into my appdata. Data persists through a restart of the container, but not an update (and updates are very frequent!). Also, just a heads up- there is a typo in the OP instructions on mounting netdata.conf (it's misspelled as netAdata.conf on both sides)

If I have a stack using depends_on to control container start order, how does that interact with the the ordering on UnRAID's Docker page which typically controls startup order?

Great work on this! I'm completely new to Docker, but did eventually get this working on Windows in Hyper-V mode (Mullvad/Wireguard). My question - is it possible to get this working with Docker set to WSL 2 mode? So far I haven't been able to get it working. The main error seems to be: 2021-02-04 15:44:05,318 DEBG 'start-script' stderr output: iptables-restore v1.8.7 (legacy): unknown option "--save-mark" Error occurred at line: 5 Try `iptables-restore -h' or 'iptables-restore --help' for more information. Which I believe means the WSL2 kernel does not include the CONNMARK module. But I swapped out the WSL2 kernel for one built with wireguard/connmark support, and I'm still getting errors. --EDIT-- Got it! I had to manually compile the WSL2 kernel with CONFIG_NETFILTER_XT_MATCH_CONNMARK=y Everything works great when using that modified kernel. Instructions here (not mine) in case anyone else runs into this in the future.