Can0n Posted November 14, 2018 Share Posted November 14, 2018 (edited) hello I have unraid 6.6.5 installed and on one of my unraid units it seems Mover is broken I i try to run mover manually i see writes to the cache and reads from one or more of my drives, I have made sure the cache settings on the shares im trying to move over are set to yes if i invoke mover the GUI shows mover is running but no traffic appears to be happening show its writing to the array, if i click the main button again mover is now showing as running anymore the log shows this when i try to run it Nov 14 09:31:14 Thor emhttpd: req (19): cmdStartMover=Move+now&csrf_token=**************** Nov 14 09:31:14 Thor emhttpd: shcmd (280): /usr/local/sbin/mover &> /dev/null & Attached is diagnostics and I have tried rebooting the server as well as uninstalling Mover Tuner plugin i thought was causing the issue any help would be muchly appreciated thor-diagnostics-20181114-0934.zip Edited November 14, 2018 by Can0nfan found reason Quote Link to comment
JorgeB Posted November 14, 2018 Share Posted November 14, 2018 Enable mover logging and try again, there should be more info logged. Quote Link to comment
Homerr Posted November 16, 2018 Share Posted November 16, 2018 I'm having the same exact issue. Mover logging enabled, here's what it says. Nov 16 05:26:47 unRAID emhttpd: req (18): cmdStartMover=Move+now&csrf_token=**************** Nov 16 05:26:47 unRAID emhttpd: shcmd (2433): /usr/local/sbin/mover |& logger & Nov 16 05:26:47 unRAID root: mover: started Nov 16 05:26:47 unRAID move: move: skip /mnt/disk1/system/docker/docker.img Nov 16 05:26:47 unRAID move: move: skip /mnt/disk1/system/libvirt/libvirt.img Nov 16 05:26:47 unRAID move: move: skip /mnt/disk3/appdata/CrashPlanPRO/conf/adb/LOCK Nov 16 05:26:47 unRAID root: mover: finished Quote Link to comment
JorgeB Posted November 16, 2018 Share Posted November 16, 2018 7 minutes ago, Homerr said: I'm having the same exact issue. Mover logging enabled, here's what it says. Files in use are skipped, you'd the need to shutdown VM and docker services first. Quote Link to comment
Homerr Posted November 16, 2018 Share Posted November 16, 2018 No VM's. Only Docker running was CrashplanPro and that shouldn't be accessing the Cache drive? Similar message after stopping CrashplanPro. I rebooted server and computer I used to copy files that ended up on the cache drive, again with similar message. All seemed to work okay prior to 6.6.5. Nov 16 05:51:35 unRAID emhttpd: req (3): cmdStartMover=Move+now&csrf_token=**************** Nov 16 05:51:35 unRAID emhttpd: shcmd (216): /usr/local/sbin/mover |& logger & Nov 16 05:51:35 unRAID root: mover: started Nov 16 05:51:35 unRAID move: move: skip /mnt/disk1/system/docker/docker.img Nov 16 05:51:35 unRAID move: move: skip /mnt/disk1/system/libvirt/libvirt.img Nov 16 05:51:35 unRAID root: mover: finished Quote Link to comment
JorgeB Posted November 16, 2018 Share Posted November 16, 2018 1 minute ago, Homerr said: No VM's. Not the VMs, the VM service: Settings -> VM Manager Quote Link to comment
Homerr Posted November 16, 2018 Share Posted November 16, 2018 Ok, set VM's to 'no'. Still same issue. Nov 16 05:56:25 unRAID emhttpd: req (7): cmdStartMover=Move+now&csrf_token=**************** Nov 16 05:56:25 unRAID emhttpd: shcmd (241): /usr/local/sbin/mover |& logger & Nov 16 05:56:25 unRAID root: mover: started Nov 16 05:56:25 unRAID move: move: skip /mnt/disk1/system/docker/docker.img Nov 16 05:56:25 unRAID move: move: skip /mnt/disk1/system/libvirt/libvirt.img Nov 16 05:56:25 unRAID root: mover: finished Quote Link to comment
JorgeB Posted November 16, 2018 Share Posted November 16, 2018 Then you likely have duplicate docker and libvirt images on cache and disk1, delete the ones not in use. Quote Link to comment
Homerr Posted November 16, 2018 Share Posted November 16, 2018 That went over my head. Can you point me to how to tell if this is the case and which ones to delete? Quote Link to comment
JorgeB Posted November 16, 2018 Share Posted November 16, 2018 Start by checking if they really are duplicated, post the outputs of: find /mnt -name libvirt.img and find /mnt -name docker.img Quote Link to comment
Homerr Posted November 17, 2018 Share Posted November 17, 2018 (edited) Thanks for helping, back from work. Here is what was returned, seems like single instance for each. Quote /mnt/user/system/libvirt/libvirt.img Quote /mnt/user/system/docker/docker.img Edited November 17, 2018 by Homerr Quote Link to comment
JonathanM Posted November 17, 2018 Share Posted November 17, 2018 Do you have menu items for Dockers and VM's on the unraid GUI? Quote Link to comment
Homerr Posted November 17, 2018 Share Posted November 17, 2018 On the Dashboard tab under Apps I have binhex-krusader, CrashPlanPro, and PlexMediaServer. All are stopped. Settings tab, System Settings I see Docker and VM Manager. Docker: Quote Docker version: 18.06.1-ce Docker vDisk location: /mnt/user/system/docker/docker.img Default appdata storage location: /mnt/user/appdata/ VM Manager: Quote Enable VMs: No I upgraded from 6.6.3 to 6.6.5 a couple of days ago, only change in setup I can think of. (Updated a couple of plugins at that time as well.) If it matters, plugins are: CA Auto Update Applications CA Cleanup Appdata Community Applications Fix Common Problems Preclear Disks Statistics unBALANCE User Scripts Quote Link to comment
Homerr Posted November 17, 2018 Share Posted November 17, 2018 I was also just browsing the contents of of the cache disk (clicked the folder on right side of cache line on Main tab). /mnt/cache/system/libvirt/libvirt.img is a 1.07gb file present there. Doing the same on disk1 shows: /mnt/disk1/system/docker/docker.img is a 21.5gb file present there. Quote Link to comment
Homerr Posted November 17, 2018 Share Posted November 17, 2018 Hmm, sometime in the last 3 hours the Mover started working and it is about 90% done. I haven't been around the computer during that time so it wasn't me... Quote Link to comment
JorgeB Posted November 17, 2018 Share Posted November 17, 2018 seems like single instance for each. That doesn't make sense, you should get at least two results from each even though they are the same file, e.g.: /mnt/user/system/docker/docker.img /mnt/cache/system/docker/docker.img or /mnt/user/system/docker/docker.img /mnt/disk1/system/docker/docker.img Quote Link to comment
Homerr Posted November 17, 2018 Share Posted November 17, 2018 (edited) Ok, has this OS gone full retard? It's doing it again. Johnnie, yeah there was only one. So... last night after I thought it was fixed and the Mover was done I moved around 3-400gb of files around which were already on the server, some may have been on the cache disk - I can't be sure - and it seemed ok. Then I copied around 200gb to the server and went to bed. The cache disk is a 640gb drive. Now it is doing this BS again somehow. 'Close all notifications'... yeah right. It just took me literally 35 minutes of clicking that to close all the dumb notifications so I could take the screenshot below. It seems Disk4 is near full too and unraid seems bent on jamming it with more. Is there some other dumb thing going on here that has been causing all this? It seems like everything, stuff already on the server even, is being run through the cache disk when moved around between shares...? I added the diagnostics file below. unraid-diagnostics-20181117-0640.zip Edited November 17, 2018 by Homerr Quote Link to comment
JonathanM Posted November 17, 2018 Share Posted November 17, 2018 In that screenshot it shows docker service is still enabled. When docker service is disabled, there will no longer be a docker item on the main GUI. Do you want the system share and appdata share on the array or the cache drive? Are they set to cache:prefer? Quote Link to comment
Homerr Posted November 17, 2018 Share Posted November 17, 2018 (edited) I need Docker for CrashplanPro and Plex. Have I missed something where I have to now choose between Docker and cache functionality? Based on your second comment I just changed system and appdata shares to 'No' to use the cache drive instead of 'Prefer'. I don't recall how I set this up, but I think I added the cache drive well after initial setup and didn't change system and appdata to this. I would have done the most straightforward implementation, i.e. just setting the disk to the cache slot and let it default to whatever the os wanted. After changing to No, and running Mover it shows this but didn't seem to do anything about actually moving data: Quote Nov 17 07:22:11 unRAID emhttpd: req (22): cmdStartMover=Move+now&csrf_token=**************** Nov 17 07:22:11 unRAID emhttpd: shcmd (3353): /usr/local/sbin/mover |& logger & Nov 17 07:22:11 unRAID root: mover: started Nov 17 07:22:11 unRAID root: mover: finished Edited November 17, 2018 by Homerr Quote Link to comment
Squid Posted November 17, 2018 Share Posted November 17, 2018 (edited) 36 minutes ago, Homerr said: I just changed system and appdata shares to 'No' to use the cache drive instead of 'Prefer'. Are you sure that's what you want. That setting means that the apps will basically run from the Array instead of the cache drive, with the resulting performance hit. If that is what you actually want, then in order to move all of the files from the cache drive to the array then you have to Settings - Docker - Disable the service. Settings - VM Settings - Disable the service Change the system and appdata shares to be use cache: YES Run mover After mover is done, re-enable the docker service. Since you don't use any VM, no point in re-enabling it) If this is NOT what you want, and want the performance gain from having the apps running from the cache drive, to move all of the files onto the cache drive Settings - Docker - Disable the service. Settings - VM Settings - Disable the service Change the system and appdata shares to be use cache: PREFER run mover After mover is done, re-enable the docker service (since you don't use any VM, no point in re-enabling it) (Press HELP when on the SHARES setting, and you will see the difference between what YES, NO, and PREFER do Edited November 17, 2018 by Squid Quote Link to comment
trurl Posted November 17, 2018 Share Posted November 17, 2018 1 hour ago, Homerr said: After changing to No, and running Mover it shows this but didn't seem to do anything about actually moving data: Mover doesn't move cache-only or cache-no. It only moves cache-yes from cache to array, and moves cache-prefer from array to cache, so this is apparently one thing you misunderstood that perhaps contributes to your complaints. 2 hours ago, Homerr said: It seems Disk4 is near full too and unraid seems bent on jamming it with more. Is there some other dumb thing going on here that has been causing all this? What disk Unraid chooses to write user share files to depends on 3 settings of the particular user share. Are you familiar with these? There is help in the webUI. Ask if you want further explanation. Hint: Split Level, Allocation Method, Minimum Free. Obviously Include and Exclude can affect which disks are used also. Quote Link to comment
JonathanM Posted November 17, 2018 Share Posted November 17, 2018 3 hours ago, Homerr said: Have I missed something where I have to now choose between Docker and cache functionality? No, you complained that the docker image file wasn't getting moved, and you have to stop the docker service to enable mover to operate on the docker image file. Mover will NOT move open files. You can reenable the docker service once you have the files where you want them. Quote Link to comment
Homerr Posted November 25, 2018 Share Posted November 25, 2018 On 11/17/2018 at 7:58 AM, Squid said: ... If this is NOT what you want, and want the performance gain from having the apps running from the cache drive, to move all of the files onto the cache drive Settings - Docker - Disable the service. Settings - VM Settings - Disable the service Change the system and appdata shares to be use cache: PREFER run mover After mover is done, re-enable the docker service (since you don't use any VM, no point in re-enabling it) (Press HELP when on the SHARES setting, and you will see the difference between what YES, NO, and PREFER do Been out of town this week, just getting back to this. I did what Squid offered here and all is good now. TY! 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.