nagle3092 1 Posted October 16, 2017 Share Posted October 16, 2017 You have path full errors in your log. Did you run out of HDD space at any time? There is also an error in sharp web socket that the header of your image file is missing. This makes it looks like you have a 0kb image, or possibly several of them. @[member="Echo"] off set out=found.txt (for /r . %%F in (*.*) do (if %%~zF LSS 1 echo %%F))> %out% pause Save the code above as zeros.bat and put it in the library folders which have issues, then click it and let it run. Afterwards, check the found.txt file it creates for the 0kb files. Delete these files. If all of them appear to need deleting and manually this is a chore, change the line below: (for /r . %%F in (*.*) do (if %%~zF LSS 1 echo %%F))> %out% To look like this below, and re-run the bat file: for /r . %%F in (*.*) do (if %%~zF LSS 1 del %%F) Note: The roku uses a different size for images than other clients. The server will create a custom sized image for displaying on the roku in the /cache folder. This way the roku doesn't have to downscale images, that can be several MB in size that take significant time to download. Instead the server sends the roku a custom sized image from the /cache once it exists. This optimizes the bandwidth and pictures load quickly. If you ran out of HDD space on your server, but not the drives which store your media files, you need to run the BAT file on your emby /cache folder. Find the 0kb files inside and delete them. This will make the server recreate the cache images as you use the roku client and hopefully you have enough freespace to accomodate them this time. I still have 10.6GB free on the vm and never ran out of free space, the network share which houses the libraries still has 8TB free as well. I will give this a shot though. Since I have all the metadata stored on the vm and not in the library locations I would still run it there? Link to comment Share on other sites More sharing options...
speechles 1920 Posted October 16, 2017 Share Posted October 16, 2017 (edited) I still have 10.6GB free on the vm and never ran out of free space, the network share which houses the libraries still has 8TB free as well. I will give this a shot though. Since I have all the metadata stored on the vm and not in the library locations I would still run it there? Cache: C:\Users\speechles-i7\AppData\Roaming\Emby-Server\cache Wherever the cache is stored is where you would place the .bat file and then click it. It will scan every subfolder. You shouldnt need to do your libraries themselves if the images work on other clients. I just mentioned that as routine maintenance. Always check for and remove zero byte files. Images, subtitles, roku bif, etc... If the providers are slow or your bandwidth not optimal it might save 0byte files for certain things. Checking none exist is just a way to make a first step towards your issue. This might not be the issue, but will check the box, and we can move onto other areas if need be. Edited October 16, 2017 by speechles 1 Link to comment Share on other sites More sharing options...
nagle3092 1 Posted October 16, 2017 Share Posted October 16, 2017 (edited) Cache: C:\Users\speechles-i7\AppData\Roaming\Emby-Server\cache Wherever the cache is stored is where you would place the .bat file and then click it. It will scan every subfolder. You shouldnt need to do your libraries themselves if the images work on other clients. I just mentioned that as routine maintenance. Always check for and remove zero byte files. Images, subtitles, roku bif, etc... If the providers are slow or your bandwidth not optimal it might save 0byte files for certain things. Checking none exist is just a way to make a first step towards your issue. This might not be the issue, but will check the box, and we can move onto other areas if need be. Didnt work, it deleted a ton of files but I have the exact same result on a roku missing the same posters. Edit: I created a checkpoint of the vm and deleted everything in the cache folder and its fixed. Edited October 16, 2017 by nagle3092 Link to comment Share on other sites More sharing options...
speechles 1920 Posted October 16, 2017 Share Posted October 16, 2017 Didnt work, it deleted a ton of files but I have the exact same result on a roku missing the same posters. If it created a found.txt it didnt delete the files in case it deletes something it shouldn't. It will just create a list of drive-letter:\path\to\filename.ext To have the .bat file automatically delete you have to change the .bat file like I mentioned in that post above. Did you do this alteration? Link to comment Share on other sites More sharing options...
nagle3092 1 Posted October 16, 2017 Share Posted October 16, 2017 If it created a found.txt it didnt delete the files in case it deletes something it shouldn't. It will just create a list of drive-letter:\path\to\filename.ext To have the .bat file automatically delete you have to change the .bat file like I mentioned in that post above. Did you do this alteration? Yes I made the altercation and ran it (actually created 2, one with the delete line). That didnt work but since you said it created cached images of the art I figured I would just create a checkpoint of the vm and delete everything in the cache folder. Well that worked. Thanks for your help. Link to comment Share on other sites More sharing options...
speechles 1920 Posted October 16, 2017 Share Posted October 16, 2017 @@nagle3092 Awesome. Now lean back and enjoy. This thread has gone on far too long, thats why I thought I would jump in.. heh The best solution may not be the easiest. Since finding the offending 0kb and deleting would be best. But having the server regenerate cache images for every app is the quickest. But keep in mind this will make initial loading of images slower on each app until these images caches refresh. 1 Link to comment Share on other sites More sharing options...
nagle3092 1 Posted October 16, 2017 Share Posted October 16, 2017 @@nagle3092 Awesome. Now lean back and enjoy. This thread has gone on far too long, thats why I thought I would jump in.. heh The best solution may not be the easiest. Since finding the offending 0kb and deleting would be best. But having the server regenerate cache images for every app is the quickest. But keep in mind this will make initial loading of images slower on each app until these images caches refresh. My thought process behind nuking it was since it found a lot of files that were 0 byte maybe there were files that were created over 0 bytes but for some reason corrupted/unreadable for whatever reason and since it was a cache I figured I would give it a shot. Thanks again for all your help! Link to comment Share on other sites More sharing options...
ebr 14930 Posted October 16, 2017 Share Posted October 16, 2017 Thanks @@speechles. I think he was probably correct in that, at some point in time, your system drive ran out of space which caused some of the cached images to be corrupted. Link to comment Share on other sites More sharing options...
Jdiesel 1114 Posted October 16, 2017 Share Posted October 16, 2017 I will have to try this today as well. Could be related to my transcoding-temp folder not clearing after a stream had ended. 1 Link to comment Share on other sites More sharing options...
nagle3092 1 Posted October 16, 2017 Share Posted October 16, 2017 Thanks @@speechles. I think he was probably correct in that, at some point in time, your system drive ran out of space which caused some of the cached images to be corrupted. I dont think that happened, emby is the only thing on the vm besides zap2xml. I dont know how I could loose all the drive space without hyper-v alerting me and then to recover said 10gb without intervening? Either way I'm glad its resolved. Link to comment Share on other sites More sharing options...
ebr 14930 Posted October 16, 2017 Share Posted October 16, 2017 I dont think that happened, emby is the only thing on the vm besides zap2xml. I dont know how I could loose all the drive space without hyper-v alerting me and then to recover said 10gb without intervening? Either way I'm glad its resolved. The most likely way I can think of is a transcode or two. A 3 hour live TV session could easily eat up 10GB of space that would then be released. In any case, it definitely appears that your cache had become corrupted somehow. 1 Link to comment Share on other sites More sharing options...
nagle3092 1 Posted October 16, 2017 Share Posted October 16, 2017 The most likely way I can think of is a transcode or two. A 3 hour live TV session could easily eat up 10GB of space that would then be released. In any case, it definitely appears that your cache had become corrupted somehow. Really? I will need to expand the vm disk space in that case. I didnt think about that though it makes sense for say a 15mb/s transcode. Those clear after viewing is over correct? Link to comment Share on other sites More sharing options...
Happy2Play 8296 Posted October 16, 2017 Share Posted October 16, 2017 Yes transcode-temp does auto clear after stopping playback but there is a topic where a few people are having issues with transcode folder isn't clearing properly. But some fixes have been added in next release. https://emby.media/community/index.php?/topic/51705-transcoding-clean-up/ Link to comment Share on other sites More sharing options...
nagle3092 1 Posted October 16, 2017 Share Posted October 16, 2017 Yes transcode-temp does auto clear after stopping playback but there is a topic where a few people are having issues with transcode folder isn't clearing properly. But some fixes have been added in next release. https://emby.media/community/index.php?/topic/51705-transcoding-clean-up/ Thanks, I'll bump up my disk space a little and check out that thread. Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now