mikedpitt420

Aha. Thanks very much for pointing me towards an answer instead of assuming everyone runs the same software / plugins / dockers setup! Thanks d00d

How do you mean? I don't understand. What more streamlined way of dealing with dockers are you referring to? I'm using unRAID's UI "docker" tab and am running 6.1.5. Is there something I have missed? If you're referring to something not included in the unRAID OS by default, why do you assume anyone has heard of what you're talking about?

So there is no way to add a line to the repositories and install it in this way with the "Add Container" button any longer? It needs manually installed somehow and not through the UI?

What do I add to the Docker Repositories to get the HTPC Manager plugin now?

I agree with your sentiment. However, why not take it a step further and say every single piece of software made by everyone has to have a little ? next to every option, explaining every function of every button in plain, non-technical speak? There would be nothing but "?" buttons and hovering explanations as far as the eye could see. Not trying to offend only playing devil's advocate. I think it's pretty well assumed that if you've decided to build and administer your own NAS, that you'll also spend some time learning the in's and out's of the system itself. Could some stuff be explained a bit better? Sure. Does every single button need ? with a non-tech explanation as well? I think not... They've gone as far as making a pretty darned conclusive manual for 6. That being said, it doesn't encompass plugins and add-ons but it sure does get into quite a bit of detail. Like Trurl stated, these plugins are user written. So it's at the user's end to decide whether or not they wish to venture outside of the published unRAID OS and add plugins. Plugins that are distributed knowingly without manuals and actual support from unRAID itself. To me this would be like adding a user generated script to SABNZBD and then being less than thrilled when it doesn't do exactly what it's supposed to, then trying to get the SAB authors to clean it up and write documentation for it.

Yes, the red X after the preclear next to the drive, that when you hover over it says "Clear Drive Stats".

Clear the stats from the drive, and you can preclear again.

No that's what I was after. Thanks very much! I guess I just misread or botched the replacement process. I keep a precleared warm spare for myself and this was for someone else, so I honestly haven't been through this actual process in a while. I was just curious as to expected behavior in each scenario. That makes it perfectly clear. So in the future, do not remove the bad drive from the array, add the new one and preclear, and reassign. The only question I would still have is, when a drive is failing and you try to watch video stored on it, it tends to buffer and skip. Which would be expected. Removing the failing drive first at least corrects that error while waiting for another drive. I guess the only way to achieve that would be doing it how I did, and just forcing a preclear via the command line? Thanks very much again for the quick response.

That's what I would expect as well, but it's not what happened. The drive failed, the array was stopped and the drive was removed, and the array restarted. All the data was visible. The new drive was added and the machine was rebooted. Listing the drives available for preclear showed no available drives, via the CLI or the unRAID UI. If the array was stopped, I was able to assign the new drive to the array. However, the only way I was able to clear this drive was preclear_disk.sh /dev/sdc So the script did not show the drive as available for a preclear, but it started anyway via the command line. The reason I was reporting it here, was because with the addition of this plugin, this was the last reason I had to regularly use the CLI and I thought that's a huge leap forward for less than experienced users who were possibly unfamiliar with the command line. So either this is expected behavior and unRAID does somehow claim the drive when there is a failed drive in the array, or there is a bug in the preclear script or plugin, that says a drive is not unavailable for preclear when indeed it is.

So is that correct? If there is a failed drive in the array, unRAID "claims it" and you are not able to preclear the drive normally? This doesn't seem like what would be expected behavior unless I'm really missing something. Since unRAID places a fairly high value on pre-clearing drives before they're added to the array (which is one of the nicest features imho) you'd think that it would want you to pre-clear any drive added. Is this a bug in the preclear script or is that actually expected behavior?

So you're saying it's not possible to preclear a drive with another drive missing from the array?

This is in response to fantax's post above. I have been helping him with his new unRAID system. He had a drive die. The array was stopped, the drive was removed from the array, and the array was started again. This seemed to work just fine, and the data on the missing drive was still accessible. He then replaced the drive. The new drive took the assignment of /dev/sdc . This was not shown under the unassigned devices plugin, so the option to preclear it wasn't there either. Stopping the array gives him the option of being able to assign it to the array and begin a parity rebuild of the data, but nothing from the preclear script, or plugin was reported. From the CLI, doing a "preclear_disk.sh -l" produced no results, it did not show the new drive. However, doing a "preclear_disk.sh /dev/sdc" began the preclear process just fine. He's running unRAID 6.1.3 . He has 5 drives attached to his motherboard, and 3 drives attached to an IBM M1015 flashed to IT mode. The drive in question is one of the ones connected to the controller card. I wasn't exactly sure where to post this, but it seems as though both this plugin, and the preclear script, were unable to see this new drive for whatever reason. Any help would be greatly appreciated.

I have recently noticed that Samba just seems to stop. At times my shares are invisible to Kodi and all other devices on the network. In mid play, a video will just stop. Attached is a copy of my log file. I was going to post in the unassigned devices plugin area, as I have that plugin, and also saw some other people having that issue, but I'm not 100% sure it's related and was wondering if someone can tell me where to go about diagnosing the issue?

That definitely sounds like a superior method, but that was for me one of the more tricky parts. Getting the right fstab entry from within the VM was driving me insane, so once it worked I just left it alone I could mount it no problem via the CLI and the normal mount command, but that was the only way I got it to work with fstab. Thanks for that very much. I've edited my post to point towards your instructions as well. Although I've heard that as well and toyed with the idea of changing it, as stated I do 6 streams no problem, which is the amount of tuners I have. So it at least works "that much". Thanks again dmacias

I had a user message me and ask me how to accomplish this, and I've actually also helped two other people do this personally so I thought I would go about posting my experience, and how I run Mythbuntu 14.04 in KVM on unRAID. I'm sure it's not the most perfect method, and I'm sure that there are other better ways to do some of the things in this post, so I'm all open to suggestions. During my setup I did find through my forum posting that doing a 9p type filesystem in fstab to mount an unRAID share from within the VM isn't necessarily the best way to do it, but it works so I honestly just left things as they are and they've been running very well. So you can choose to do that however you like, I believe it should be just a normal samba share. Again this is just my experience and it works, but very open to any changes or suggestions. I'm also aware of SparklyBalls MythTV docker which I encourage everyone to try before all of this. I personally wanted more control over my setup and chose a VM over a pre-baked docker but if you're looking for simplicity that's the way to go. I wouldn't have even thought a working docker with Mythbuntu / working MythWeb etc. could be built, and I wouldn't even have remotely tried to attempt it myself. So MAD respect to SparklyBalls and in no way am I trying to discourage people from going that route. Having said all of that.... I didn't want to run MythBuntu and have it store it's absolutely giant recordings it generates on the array or my relatively small SSD cache drive. For those of you that don't have experience with MythTV the output MPEG files are HUGE. Even several people casually watching Live TV ( of which it stores for 24 hours by default for FF/REW purposes, and this can be adjusted if you're hurting for disk space) can create quite a bit of video data, so for this reason I opted for adding one 1TB outside the array HDD to store both the ISO for MythBuntu to load from since it doesn't affect adversely performance, and the recordings that MythTV generates. The VM itself runs on my SSD cache drive. 1. Install the "Unassigned Devices Plugin" here: https://raw.githubusercontent.com/gfjardim/unRAID-plugins/master/plugins/unassigned.devices.plg 2. (Optional) Install the Pre-Clear plugin. It's still in beta, but I've had great success with it so far. It was also one of the last things I actually found myself dropping to command line with any frequency so I think it's a nice step forward for unRAID in general. Note, you also must follow the instructions for where to actually place the preclear_disk.sh script itself on the flash drive where the plugin will look for it : https://raw.githubusercontent.com/gfjardim/unRAID-plugins/master/plugins/preclear.disk.plg 3. Preclear the HDD you plan to use, clear the statistics from it through the unassigned devices plugin. Format the drive to XFS also via the Unassigned Devices Plugin. 4. From the unassigned devices plugin on the unRAID MAIN screen, click the small "+" icon below the drive (make sure it is unmounted) 5. Click somewhere inside of the mount path, and change it to read "/mnt/disks/mythtv" **note the extra "s". That is intentional and I use it as a way to separate my array disks from my non array disks when I'm banging around in the CLI. So it's not a typo. 6. Press ENTER inside of the text box to save your text changes, and mount the share. It is now accessible via the CLI at that location. 7. SSH into the server, and create two directories on this new HDD. One called "ISO" and one called "recordings". The final paths should be /mnt/disks/mythtv/ISO and /mnt/disks/mythtv/recordings mkdir /mnt/disks/mythtv/ISO mkdir /mnt/disks/mythtv/recordings 8. Download your Mythbuntu 14.04 ISO from their website, and place it in the directory /mnt/disks/mythtv/ISO . If you're having trouble getting the ISO there, from the CLI cd /mnt/disks/mythtv/ISO wget http://cdimage.ubuntu.com/mythbuntu/releases/14.04.3/release/mythbuntu-14.04-desktop-amd64.iso For 64 bit systems only. Please refer to Mythbuntu's download page if you need links to another version for some reason : http://cdimage.ubuntu.com/mythbuntu/releases/14.04.3/release/ 9. From the unRAID UI, go to Settings >> Network Settings and create a bridge adapter called br0. I chose to specify an IP address automatically as I use my router's DHCP server to assign IP's via MAC address but you can specify one or use DHCP and your router. Either will achieve the desired effect of giving this VM a static LAN IP. 10. From the CLI, create a directory on your cache drive called /mnt/cache/vmdata/MythTVKVM mkdir /mnt/cache/vmdata mkdir /mnt/cache/vmdata/MythTVKVM This is where the actual VM will run its' operations. It's worth taking note that the difference in running a VM from an HDD and an SSD, especially in the context of what MythBuntu is doing, which requires quite a bit of I/O, is EXTREMELY significant. I'm honestly not 100% sure it wasn't something to do with my hardware, but running this in the same way in a VM from an HDD, I could BARELY pull 3 HD 1080i streams without it choking, no matter how many cores or how much RAM I gave it. Running it from a proper SSD cache drive made a HUGE difference in performance. I can now do 6 1080i streams with 1 core and very little devoted RAM as mentioned below and in the attached settings screenshot. 11. Now we need to create the VM itself. Open your unRAID KVM manager, and create a new KVM. Attached are a screenshot of my settings. You can create any MAC you like, but make sure you are using the newly created br0 network interface. 12. Create the VM, and check on your router that the VM has indeed popped up, is on the network, and has the correct LAN IP address. You can see from the attached image I gave my MythBuntu VM 2 cores (although it did run fine pulling 6 HD streams with 1, I wanted to give it some buffer) and very little memory. It works wonderfully to watch or record 6 simultaneous 1080 HD streams and I don't have some great hardware running it. See my sig for my specs. 13. From the VM tab in the unRAID UI, click on your newly created and running VM icon, then click "VNC REMOTE". This will drop you into the VM in X-Windows. 14. You will be prompted to install MythBuntu. Follow all of the prompts, install the updates etc to finish you installation. Reboot your VM, and get yourself back to the VNC Remote. 15. From the menu, open a terminal window and type: sudo nano /etc/fstab You need to add this line to the bottom of this file: shares /mnt 9p trans=virtio,version=9p2000.L,_netdev,rw 0 0 EDIT: Please refer to the next post down for another (more correct) way of editing your fstab file inside of your VM to mount this remote unRAID share Now, save fstab, reboot the VM. From The CLI within your VM /mnt should have mounted /mnt/user/disks/mythtv Check via the CLI on your MythBuntu system to be sure cd /mnt/ ls -la as long as you see your directories you've created earlier "ISO" and "recordings", your remote unraid share is mounted. 16. Run the MythTV configure through the X-Windows menu, and setup your tuners, episode guide provider, and storage directories for MythTV (I prefer schedulesdirect as you can trim down all the channels you care nothing for). Your recordings directory for both "recordings" and "LiveTV" should reside within the /mnt/recordings directory from within the VM. So all of my scheduled recordings end up in /mnt/disks/mythtv/recordings/ and all of the stuff that's deleted daily for live tv is in /mnt/disks/mythtv/recordings/livetv . Just remember, that from within your VM, this directory is just called /mnt. But from the rest of your actual unRAID box, the former path is where they reside, which should also already be shared via samba as "mythtv" so you can access your recordings through a SAMBA share to manipulate them however you like. From the MythTV configuration menu, you want to change all of your "Directories" settings to be somewhere in /mnt . 17. For future ease of use in manipulating recordings etc. I would recommend mapping a network drive samba share (through windows of whichever OS flavor you choose) to \\your.server.IP.address\mythtv so you can grab, edit, copy, and delete your recordings without having to drop into the actual VM itself. Well, that's it You're now all setup and can get to tweaking all of your mythtv settings the way you like them with commercial editing, transcoding with handbrake, etc. I don't have ANY information of commercial editing, transcoding, or anything of the like. I primarily use this for NHL and NFL, and to record Jeopardy daily. I don't save any of these recordings or edit them to store in my library (that's what Usenet is for again, see my sig). It works like a charm, the guides in kodi are gorgeous (especially with Aeon Nox) and the box is quick and responsive to tune channels. From my small experience with Myth's commercial flagging, it depends on the channel entirely and works ~ 50% of the time. So for my purposes, this is for watching sports live. Yes, I'm a sad sop who pays for cable for that only.... I hope this has helped some users, and again, I'm open to any comments or suggestions. EDIT: 9/26/16 after upgrading to unRAID version 6.2 this fstab entry no longer works (as it probably shouldn't have in the 1st place as mentioned above). Here is a link to the correct fstab entry to automount an outside of the array share from within an Mythbuntu 14.04 KVM : http://lime-technology.com/forum/index.php?topic=52313.0

I'm still unable to get the temp sensors plugin working after getting PERL installed. Any other options until lmsensors.org is back up?

Thanks very much, posted my question over there

Is it possible to change the default webserver to chorus or chorus2? There's a streaming option, like yatse, as well as a whole "tool box" of JSON commands you can send to your kodi box. Apologies if I don't quite wrap my head around why doing a clean library or update from the headless box itself would cause everything to vanish. Wouldn't the "initial" scan to create a DB be done on the docker kodi ? So I guess I don't understand why updating it there would cause an issue.

Sorry for the ignorance, but do you mean initially there is some lag when it's each client first gets something added, or initially at startup each time? So it would just take some time to sync, but the day to day has no noticeable difference? Sounds great.

Is it possible to change the webserver to chorus, or possibly chorus2, and ports etc ? They've apparently gotten pretty good at the streaming of stuff from through chorus (although I''m not sure how many formats are supported). So people who don't need transcoding would have another streaming option outside of emby and plex, or using yatse. Their default webserver is GOD awful, but it does have the update and clean library buttons which is nice, especially for a centralized DB. Chorus is much nicer but it lacks the "clean" button, in an easily accessible menu. You can access a tools menu through chorus that allows you to very easily send a whole ton of JSON commands (including all the "clean library" functions, so they are accessible through Chorus) in various ways to your kodi box. I would think it would be a much nicer option and infinitely more useful in your docker Sparkly, but obviously just a suggestion I'm thinking of installing this docker and moving to a shared DB, but I've never used a central DB. I have all giga networking and cat6, a more than capable home router, but I have a fairly large library and I am wondering how much noticeable lag people get when you just hold down the button on your remote and fly through your movies? Or if you try to play trailers, or bring up extended metadata is there any noticeable difference? This would definitely be superior to what I do now, which is have each kodi box update at startup, and I have notifications from all of my automation to all of my different kodi boxes as well and if one isn't on, it obviously doesn't update until it wakes back up. This would eliminate that as well. In my opinion going to all of the trouble to build a nice NAS and collection, and proper htpc's, I want the best possible media experience. Meaning that I don't want any noticeable lag at all when browsing just about whatever sized library. I have the newest stable Isengard and Aeon Nox 3 on all of my htpc's and all of the nicer 1080p fanart, etc. enabled. What's the experience like for you guys?

I would pull your flash drive and check to make sure it's not on it's way out if that were me. Sounds as though something didn't install correctly in the first place. But I'm no expert.

Maybe a reboot, or a hard browser refresh? (Ctrl-F5) ? You don't have this listed in your plugins? "Dynamix Cache Directories"?

Refused for me as well.

Installed PERL, hit detect, blank screen for a second, still detects nothing

I have installed system temp, and click detect, and nothing happens. What am I missing?