limetech Posted September 19, 2016 Author Share Posted September 19, 2016 - Unlike 6.1.9, the Docker system in 6.2 no longer supports the docker.img file to be located on a disk mounted with the Unassigned Devices plugin. You must locate it either on the Cache drive or on the array. Even if this was never officially supported, this regression is REALLY sad. No that restriction is not true. We'll update the OP. Many many people have trouble with this. Some do not. Probably a race condition but I still felt my contribution was justified. Sent from my LG-D852 using Tapatalk I haven't looked at the UA plugin for a while, as long is it mounts devices in response to 'disks_mounted' event, which takes place before services are restarted, it should work ok. Quote Link to comment
Squid Posted September 19, 2016 Share Posted September 19, 2016 - Unlike 6.1.9, the Docker system in 6.2 no longer supports the docker.img file to be located on a disk mounted with the Unassigned Devices plugin. You must locate it either on the Cache drive or on the array. Even if this was never officially supported, this regression is REALLY sad. No that restriction is not true. We'll update the OP. Many many people have trouble with this. Some do not. Probably a race condition but I still felt my contribution was justified. Sent from my LG-D852 using Tapatalk I haven't looked at the UA plugin for a while, as long is it mounts devices in response to 'disks_mounted' event, which takes place before services are restarted, it should work ok. Like I said works for some fails for most coming from 6.1. Easiest solution is to just switch to using a cache drive and since I don't use UD for that purpose I haven't bothered figuring it out beyond a solution I know works. Sent from my LG-D852 using Tapatalk Quote Link to comment
dlandon Posted September 19, 2016 Share Posted September 19, 2016 - Unlike 6.1.9, the Docker system in 6.2 no longer supports the docker.img file to be located on a disk mounted with the Unassigned Devices plugin. You must locate it either on the Cache drive or on the array. Even if this was never officially supported, this regression is REALLY sad. No that restriction is not true. We'll update the OP. Many many people have trouble with this. Some do not. Probably a race condition but I still felt my contribution was justified. Sent from my LG-D852 using Tapatalk I haven't looked at the UA plugin for a while, as long is it mounts devices in response to 'disks_mounted' event, which takes place before services are restarted, it should work ok. UD mounts its devices on the 'disks_mounted' event. Quote Link to comment
Bjonness406 Posted September 19, 2016 Share Posted September 19, 2016 Time to add some new feature request now? What about a working powerdown now that the powerdown plugin is deprecated? (Not blaming dlandon here, he have done an awsome job with it. Now it is LT time to fix this issue for real!) Quote Link to comment
BRiT Posted September 19, 2016 Share Posted September 19, 2016 By working, I would say one that does not depend on the web UI being responsive. Quote Link to comment
bonienl Posted September 19, 2016 Share Posted September 19, 2016 I have another wish, something discussed in the past as well, but it requires core changes (read: long development). Quote Link to comment
Bjonness406 Posted September 19, 2016 Share Posted September 19, 2016 I have another wish, something discussed in the past as well, but it requires core changes (read: long development). And that is? Changing emhttp to something else? Or something entirely different? Quote Link to comment
JonathanM Posted September 19, 2016 Share Posted September 19, 2016 I have another wish, something discussed in the past as well, but it requires core changes (read: long development). Let me guess. Multi-threaded webserver for the main gui? Preferably using apache or something mainstream? Quote Link to comment
limetech Posted September 19, 2016 Author Share Posted September 19, 2016 Time to add some new feature request now? What about a working powerdown now that the powerdown plugin is deprecated? (Not blaming dlandon here, he have done an awsome job with it. Now it is LT time to fix this issue for real!) Please refer to the existing defect report: http://lime-technology.com/forum/index.php?topic=51983.0 Quote Link to comment
limetech Posted September 19, 2016 Author Share Posted September 19, 2016 I have another wish, something discussed in the past as well, but it requires core changes (read: long development). Let me guess. Multi-threaded webserver for the main gui? Preferably using apache or something mainstream? It's going to be nginx. Quote Link to comment
binhex Posted September 19, 2016 Share Posted September 19, 2016 I have another wish, something discussed in the past as well, but it requires core changes (read: long development). Let me guess. Multi-threaded webserver for the main gui? Preferably using apache or something mainstream? It's going to be nginx. Awesome, I am going to be looking forward to that, should seriously improve performance of the webui, not to mention reduce locking conditions. Sent from my SM-G900F using Tapatalk Quote Link to comment
Ice_Black Posted September 19, 2016 Share Posted September 19, 2016 I have another wish, something discussed in the past as well, but it requires core changes (read: long development). Let me guess. Multi-threaded webserver for the main gui? Preferably using apache or something mainstream? It's going to be nginx. Websocket would be good as well for realtime. Quote Link to comment
ken-ji Posted September 19, 2016 Share Posted September 19, 2016 Since we're asking for improvements: How about the early patching system I'm requesting? http://lime-technology.com/forum/index.php?topic=51562.msg494796#msg494796 Its obvious that LT is relying on Slackware as the core, which should make patching a breeze if done early enough - before you start the network or any other service. I wrote a braindead upgrader once for a fleet of desktop machines, of course since the patch is late and to local disk rather than RAM, I would need to reboot the machine after the patches are applied. Quote Link to comment
bonienl Posted September 19, 2016 Share Posted September 19, 2016 Perhaps a new topic "What do you like to see in unRAID 6.3" needs to be started? Quote Link to comment
perhansen Posted September 19, 2016 Share Posted September 19, 2016 Just upgraded from 6.1.9. VMs and dockers running. Installing second parity as we speak. Everything is running smooth. Thanks LT and other contributers for making this product so awesome. Sent from my iPhone using Tapatalk Quote Link to comment
trurl Posted September 19, 2016 Share Posted September 19, 2016 Just wanted to take a moment to give a special thanks to our moderators who have been doing an awesome job in helping support the release of 6.2. You guys are fantastic and thank you for doing what you all do. I know I'm late to the party but just wanted to +1 this. I have been (and still am) on vacation with limited connectivity. I haven't even done the upgrade myself (just the same as rc5, right?). Don't want to have to reboot since I suspect the housesitter watches plex most of the time. Quote Link to comment
wirenut Posted September 19, 2016 Share Posted September 19, 2016 6.2 Single vs Dual Parity testing on the D525 X7SPA-HF-D525 @Garycase File Copy during parity Check Note that I do NOT copy files (to the Cache) during parity checks normally, however mover is suspended during parity checks so in theory this should be ok I was unaware of this.... Is there a setting for this or is it automatic? Quote Link to comment
landS Posted September 19, 2016 Share Posted September 19, 2016 Oh shoot.... I was confusing mover suspend during parity for a cache directory suspend during mover.... 1 - will correct pist 2 - will need to be cautious about no writes during parity. @Tom. Feature request: suspend mover during parity checks switch Quote Link to comment
Gizmotoy Posted September 20, 2016 Share Posted September 20, 2016 Kind of disappointed I need to either add a cache drive I don't want, or leave an array disk spinning 24/7 to use docker now. Not terribly excited with either option. Well, how else are you going to run Docker? You can't run it without having a drive, and the drive will be active as the apps within the Dockers will be accessing the drive. You also don't HAVE to use the cache drive to cache, you could put a small SSD in there and ONLY use it for Docker and appdata. A cheap 60GB or 120GB SSD isn't expensive, over here in the UK it's possible to get a 120GB SSD from a decent brand for less than £40. What you're trying to achieve is like having a car, wanting to drive it, but not being terribly excited by starting the engine. Well, I had been planning to run Docker the exact same way I have been since Docker support launched, as stated in the part of my post you cut you out of your quote: via an SSD mounted via Unassigned Devices. I have an engine. It works just fine on 6.1.9. The closest Engine-based car analogy would probably be needing to do an engine swap because you updated the ECU firmware. Others did mention that you can create a cache drive that doesn't cache anything, so it looks like I'll probably do that. Though Limetech said this should work, so I'm not sure what to do. Wait and see, I guess. - Unlike 6.1.9, the Docker system in 6.2 no longer supports the docker.img file to be located on a disk mounted with the Unassigned Devices plugin. You must locate it either on the Cache drive or on the array. Even if this was never officially supported, this regression is REALLY sad. No that restriction is not true. We'll update the OP. Many many people have trouble with this. Some do not. Probably a race condition but I still felt my contribution was justified. Sent from my LG-D852 using Tapatalk I haven't looked at the UA plugin for a while, as long is it mounts devices in response to 'disks_mounted' event, which takes place before services are restarted, it should work ok. UD mounts its devices on the 'disks_mounted' event. I'm going to preclear my new parity a couple times to make sure it's stable, then will probably try it this weekend. Sounds like it *should* work, but just doesn't. Interesting. Quote Link to comment
Shadowrunner Posted September 20, 2016 Share Posted September 20, 2016 Hi, Just a heads-up - after upgrading, all my shares were set to "Use Cache? - No". I only noticed this when the Mover failed to move anything. Is this expected behaviour? SR Quote Link to comment
binhex Posted September 20, 2016 Share Posted September 20, 2016 ok so ive got slightly screwed over by the dodgy dynamix update for 6.1.9 (CA auto updates dynamix by default), ive seen the fix for this but thought what the hell, this is a sign from the god-geek that i should just embrace 6.2.0 and plunge straight in :-). the issue i have is that i dont now have a working UI in which to prep for the update, so i cant get to the VM screen in order to disable auto boot on startup (part of the procedure for 6.2.0 is i noted, dont auto start vm before tweak performed). ive done a quick text search across the flash drive but cannot find any reference to auto boot settings. so the question in short is, does anybody know if its possible to temporarily disable auto boot of a named vm through CLI, im no libvirt expert so it could very well be buried in there somewhere maybe?. well it all went swimmingly in the end, strangely when i got home i tried the web interface and it all sprang into action (fixed dynamic release for 6.1.9 maybe?), so i was able to get to the vm tab and stop auto start on bootup, which is just as well as you cant configure a vm when the array is stopped, rebuilt loopback docker image in the new location, changed the docker template location for config using a bit of replace magic in notepad++ and finally switched the bridge for the vm, and voila, all is well, very nice release guys, now if LT could only fix that pesky ip table mangle issue :-P Quote Link to comment
itimpi Posted September 20, 2016 Share Posted September 20, 2016 Just a heads-up - after upgrading, all my shares were set to "Use Cache? - No". I only noticed this when the Mover failed to move anything. Is this expected behaviour? I would not expect existing settings about the cache usage by shares to be changed. However if you did not have cache enabled at the global level, or if you do not physically have a cache drive then that is the settings I would expect. Quote Link to comment
gundamguy Posted September 20, 2016 Share Posted September 20, 2016 Just a heads-up - after upgrading, all my shares were set to "Use Cache? - No". I only noticed this when the Mover failed to move anything. Is this expected behaviour? I would not expect existing settings about the cache usage by shares to be changed. However if you did not have cache enabled at the global level, or if you do not physically have a cache drive then that is the settings I would expect. I had a share change to "Use Cache - Only" the good news is that it's super simple to fix that. Quote Link to comment
itimpi Posted September 20, 2016 Share Posted September 20, 2016 Just a heads-up - after upgrading, all my shares were set to "Use Cache? - No". I only noticed this when the Mover failed to move anything. Is this expected behaviour? I would not expect existing settings about the cache usage by shares to be changed. However if you did not have cache enabled at the global level, or if you do not physically have a cache drive then that is the settings I would expect. I had a share change to "Use Cache - Only" the good news is that it's super simple to fix that. just for interest, which share was that. I was wondering if it was something like the appdata share that would normally have that setting. Quote Link to comment
landS Posted September 20, 2016 Share Posted September 20, 2016 Write Speed Test Write speed to the Cached Share, all other hardware and test file the same (4 GB ISO File) V5 near 100 MB/s V6.1.9 max of 40 MB/s V2.0 57 MB/s v6.2 Write speeds are still not as fast as v5, however it is nearly 40% quicker than v 6.1.9! The low Cache write speeds have been my biggest complaint with v6, however the benefits far outweigh this obstacle. I strongly recommend looking into adjusting your disk tunables, they can make a dramatic difference. It's possible you may be able to get close to v5 speeds again. Oh my goodness... Parity check post squids modified tunables script, set to Unthrottled unlimited/unteathered.... 2 threads around 80% 2 up to 40%. And I can now watch a DVD without fail or a 1080p with only an occasional stutter during the parity check. On track to shave a goodly chunk of time off of the check too! Awesome recommendation RobJ Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.