Kaastosti

Members
  • Posts

    14
  • Joined

  • Last visited

Everything posted by Kaastosti

  1. Updated from 6.11.5 to 6.12.4 yesterday. Since then, symbolic links can no longer be used as a share. They still exist on the disk, but are not shown in the /mnt/user/ folder or in the shares list. For example I have /mnt/user/test pointing to /mnt/user/data/specific_folder/, which under 6.11.5 shows up in the shares list (see attached image). As a result, I can use it as a network share in Windows machines. Since the upgrade, it's gone. Just performed a rollback (which went smooth), and *poof* there it is again, working. Was this deliberate and did I not read release notes too well, or is it a bug?
  2. Right now I was focusing on the appdata. I put it back on cache only to see whether the message was still there. Which it was. Now put it back on 'prefer', disabled docker, disabled VM, invoked mover. Done instantly. When I check /mnt/user/appdata, there are even folders there for dockers that no longer exist. When I check /mnt/disk1/appdata, I only see folders for binhex-krusader and pihole there. Did some comparison, both were safe to remove from the array. Done so... set shares back to cache only, re-run FP, no issues left Thanks for the help!
  3. I'm probably missing something obvious... but not sure how to fix. I'm getting the "Share x set to cache-only, but files / folders exist on the array" message. And indeed, there are files for that share on the array. So, following instructions for once, I set share x to 'prefer' cache, disabled docker and vm... and invoked the mover. That starts to do its magic, after which there is more data on the cache drive. Hooray. However, I can still see the data on the array as well. Was expecting that to have been moved/merged to cache. And sure enough, when I set the share back to cache only, the message pops up again. Not feeling comfortable just moving this manually and deleting from array, since that might overwrite new(er) versions. What's the safe move here? Think it's a bit overkill, but including the diagnostics anyway thanks in advance for the help! kaasnas-diagnostics-20230408-1953.zip
  4. Update: Turns out the mover was actually doing something. One hour wasn't enough to make a difference, but the entire night was. Got back this morning and there was just 300MB left on the cache drive. That I can manage. In with the new, out with the old... so far it seems to work like it should. Sorry to bother you 😬 - - - - - - - - - - A couple of days ago my cache SSD started throwing CRC errors. Luckily I just freed a larger SSD drive from my desktop system, which I was going to install this weekend. So of course today was the day the system properly f'ed up. VM nowhere to be found, dockers not starting, the stuff of nightmares. Breathe in, breathe out, let's replace the cache drive. One of the things I would like to do first is move everything from the current cache drive to the array (which is still working perfectly). So stopped dockers, stopped VM and set the shares to all use cache and invoked the mover. That's been running for over an hour without much of a change on cache usage. I'm aware that I'm using the cache disk for more than just cache... but I was expecting that either the mover would move stuff, or simply stop because it's 'done'. The rest would be my own problem. System log doesn't show any mover action... but I can't stop the array since the process is still running. Don't want to do anything that might cause more harm, so going to let it run for the night. Honestly not expecting much progress in the morning. What would be a good (and somewhat safe) strategy to replace the disk? Thanks in advance for the help!
  5. Thanks! Everything back up again Just out of interest... where can you find all available versions/tags? How'd you know to add 24.0.5 specifically for a still working PHP version?
  6. Same issue here. Find Nextcloud in the list of dockers and click the icon in front of it. Select 'edit'. Then, where it says "Repository: linuxserver/nextcloud", add the :140 or any version to the latter. Click 'Apply' and the selected version is downloaded. While this does downgrade Nextcloud, at my instance the problem persists, just with a lower version 😛
  7. Thanks for the reply, that option wasn't there for me. Changing network settings require both Docker and the array to be down... and my array was in the process of building the parity. I found earlier that when shutting down the array with 2 hours (of 16) of parity building to go, it will simply restart the entire process. Wasn't going to make that mistake again When it was finished, turned it off, turned bridged mode off, turned it back on and... voila, there was the br0 option. Selected it, worked like a charm the first time. No clue why it wasn't there to begin with, I haven't been messing around with the network settings.
  8. It's almost three years old, but glad you posted it, thanks! Couldn't figure out why my instance wouldn't connect...
  9. No doubt I'm missing some obvious answer, but my pihole installation keeps insisting on having the IP of the server itself. In the instructions it is advised to use bridge mode and give pihole its own IP. But even then it stays the same. When selecting the WebUI it can't be opened. I've re-installed, removed, installed and changed settings all evening. Won't budge. Config: Installing: Mapping: Update: Turns out the drop-down on top does not show the br0 interface for me. So somewhere in my config there seems to be an error... will have to investigate how to get it.
  10. Hahaha I know the feeling, "if only we would've known that when we started"... or "holy crap, did I really write that :|" Only one way to find out, will give it a go!
  11. Thanks for the quick reply! We can't mess around in the coding ourselves, can we? No idea what language it's in, but would love to give it a try. If not to bend it to my will, then at least to know how plugins can be built
  12. New to the whole unRAID party, just finished moving all data the the shiny new array Cache disk ready to go, parity is being calculated, and I'm getting into installing all kinds of apps and services. So far I'm loving it! Not sure whether this is a bug, but when in the CA tab, it only seems to remember my last tab if it was in the first five of the 'Categories' submenu. I'd like APPS to always start in the 'Pinned Apps' menu... so far, no luck. If I select it, switch to TOOLS and come back, we're back at one of the 'Categories'. Am I missing an option, or is this intentional? Thanks in advance for the help!
  13. Have you tried to send a message to your bot using the HTTP call? Simply from a browser. As long as your bot token is correct, the chat id is correct and the bot is in the chat... should work. Then at least you know technically it could work and the error is on unRAID side.
  14. Ah yes the Telegram API requires both chat id and message text to function: https://core.telegram.org/bots/api#sendmessage The chat_id parameter is part of the test script in Telegram.sh, but in my case (and I guess I'm not alone) the 'chatid' file that is referenced does not exist. If you create it like DavKBe mentions, it is read and the test script works immediately. What I did was create my own Telegram channel, so I can have multiple bots in there. Applications report to their own bot, and by setting the 'Sign messages' option, the bot name is shown. That way I know where the message is coming from. When the channel is there, add @get_id_bot to it. This bot will respond to each message with the chat ID. Once you have it, remove the bot. That is the chat ID you need to add to the 'chatid' file (including the - prefix). Creating you own bot is easy using @BotFather. Just send that bot a message and start the conversation with /newbot. After naming your shiny new companion, the process will end with the BotFather confirming your bot creation and the access api you need. You can test your bot manually by simply using a HTTP call (replace <bot_access_api> and <chat_id> with your values): https://api.telegram.org/bot<bot_access_api>/sendMessage?chat_id=<chat_id>&text=test