ksman75

SOLVED...sorta. I tried the "Check Filesystem" tests and everything tested good, but the two drives were still marked as bad. I went to "Tools" and chose "New Config", keeping all my settings. Both drives were being flagged with more errors than was possible. I created a "New Config" without keeping my assignments, with the same results. Errors! I'm using a Zalman MS-800 case three 5 in 3 drive cages/backplanes (two ICY-DOCKs and one Chenbro), two LSI 9207-8i HBAs and supermicro fan-out cables. From top to bottom, left to right, I have my two parity drives followed by my data drives, and my two cache drives on the bottom. To further eliminate power/cable problems, I reversed all my drives except for the cache drives. My new configuration is from bottom to top, right to left, with the two cache drives followed by the two parity drives, then my data drives. With a "New Config", I added/assigned one drive at a time, started the array, waited a minute or two, stopped the array, then started it again. Doing this with each drive in turn, it took a couple of hours, but every single drive came back online with no errors or loss of data. I finally added/assigned my two parity drives, and am now rebuilding parity. So far, I haven't encountered any problems or errors. Have I solved this? Only time will tell, but it's looking good for now!

Today, my parity drive went "red x" as well as one of my data drives. Parity drive first...WD Red that I bought black Friday 2019. I can access its smart page, and run the smart self test, but unRaid flags it with a red x. I've swapped it and my second parity drive and the problem follows the drive so I don't think it's my drive bay, or cable. Disk 2...WD Blue that I bought Christmas 2019. Just flaged with the red x. I can access its smart page, and run the smart self test, but unRaid flags it with a red x too. Reseated the drive and it's still showing bad. Now it says it's saying Unmountable: No file system. If I format disk 2, will I loose the data that was on it, or have I already lost those files? Should I run a parity check/rebuild? Not sure where to go from here. unraid-smart-20200130-1319-parity.zip unraid-diagnostics-20200130-1316.zip unraid-smart-20200130-1228-disk2.zip

I personally prefer the linuxserver plex container...

Only Windows (10) Security and I have AdBlock disabled for my domain. Also, I have the same problem when I go to the unRaid GUI on my server, and there's no ad blocker or anti virus installed on the server. EDIT: I forgot to mention in my original post, using MC, I can access everything on my server. All shares are still in /mnt/user. I can also FTP to my server and am currently copying my most critical files to my Windows box.

Yesterday, I was playing with permissions with one of my shares, and though connected to all my shares, I noticed the server "Shares" tab said "There are no exportable user shares". I have tried everything I can think of, including wiping my flash drive and installing and configuring unRaid from scratch. Today, I created a share called "Test". It looked like it created it, but when I go back to the "Shares" tab, I get the same "There are no exportable user shares". I'm at a total loss as to what's happening, and how to get my shares back! Help!!!!! unraid-diagnostics-20191213-1035.zip

Glad to help

I'm using the linuxserver.io container. You're using the Limetech container. Not a big difference. Limetech has less options to screw up, LOL. At the bottom of your screen you have the "Add another Path, Port, Variable, Label or Device" option. Click on that and you can add the path(s) you need, pointing to your actual media. It'll look like the next screenshot. The "Config Type" defaults to "Path", so leave it. "Name" can be anything you want. "Container Path" is what you'll see in Plex. Be sure to have the leading slash. "Host Path" is a drop down you can use to drill down to the path on your array path. If it helps, it's very similar to mapping a network drive in windows. You want to map your host path (/mnt/user/whatever) to a local path (/mymedia in this example) that Plex will see.

You don't want to use /mnt. When you install Plex you have the following screen... Notice the three in red? These are the paths you want to use in Plex. You want /Movies for your movies, /TV for your TV shows, /Music for your music, etc. I add a path "/user1" that points to /mnt/user/media where all my media folders are. On my system, Plex looks like... On my system, it looks like this... I hope this helps

Above you show 'split level' set to 'auto split any directory' - do you think it was like that before, when you had appdata existing only on array? Also, do you know if appdata/plex directory specifically existed on multiple array disks? Yes, that's how all my shares are configured. Yes, appdata was spread across all 14 of my array drives. What I'd like to ask you to do is leave all config exactly the way it is, except, change the Plex appdata mapping from: /mnt/cache/appdata/plex to /mnt/user/appdata/plex Sorry . It's not that I don't want to, but I've finally got it working, and don't want to invest any more downtime, in the event it starts failing again. I hope you can understand this from my perspective.

My original configuration didn't use or have cache at all. Plex Config was set to /mnt/user/appdata/plex. Appdata (and all other shares) was configured to not use cache. After updating unRaid to 6.7.0, the database corruption started. While trying to find a fix, I found this link: https://github.com/binhex/documentation/blob/master/docker/faq/unraid.md. Q/A 7 seemed to describe the problem...see below. I rolled unRaid back to 6.6.7, added a single 240GB SSD cache drive, set appdata to "Cache Only", deleted the Plex folder in appdata, and reinstalled Plex. I ran this configuration for two weeks to make sure it was stable, then updated unRaid to 6.7.0 on June 13th. It has been one week today, running under 6.7.0, and I haven't has a single problem. I'm running several of the "Community Applications" plug-ins, and several more of the "Dynamix" plug-ins. I haven't made any changes to the plug-ins. I use "CA Backup / Restore" to backup my appdata daily, and haven't made any changes to it either. Hope this helps... Q7. I am seeing corruption with the applications database or issues when attempting to run Lidarr/Radarr/Sonarr/Plex/PlexPass, what could be the cause? A7. Certain applications are not compatible with FUSE, which is part of unRAID, FUSE is used to join or fuse (thus the name) multiple disks together to create a logical view of your media. The FUSE system is seen when you browse any share with /mnt/user/ in the path, such as /mnt/user/appdata. Due to this limitation certain apps need to have their configuration files defined on a non FUSE share, to do this follow the procedure below:- Stop the docker container that has the issues Left click the icon in the unRAID Web UI and select 'edit' Click on the toggle to switch to 'Advanced View' (top right) Scroll down to the path defined for /config and change it:- from /mnt/user/appdata/<container name> to (if you have a cache drive):- /mnt/cache/appdata/<container name> or (if you do NOT have a cache drive):- /mnt/disk<number>/appdata/<container name> Click on 'apply' to save the change Check the application is running. Note:- If you do NOT have a cache drive and you switch to a specific disk then you may need to reconfigure the application from scratch.

The original Plex config was /mnt/user/appdata/Plex. The new config is /mnt/cache/appdata/Plex. I also changed appdata to be cache only. I won't say this is the solution for everyone, but it is working for me. It's been a week, and I haven't had a single problem. Before making these changes, it would corrupt within hours, if not minutes.

The default, btrfs...

So, it's been one full week since my last post, and everything is still running smoothly.It seems that moving appdata to cache/a single dedicated drive may be the fix for Plex corruption.

I rolled back to 6.6.7 on May 31st, and haven't had any problems since then, which should come as no surprise, since the problem seems to follow 6.7.0. I had read that spreading Plex over the array could cause the corruption problem, so I installed a 240gb ssd when I rolled back and set appdata to cache only plus I edited all my docker containers to use /mnt/cache/appdata. To get appdata on to the cache drive, I used MC (midnight commander) to move appdata from all my shares to cache. I scoff at Triskaidekaphobia so on the 13th, I upgraded to 6.7.0 to see if forcing appdata onto the cache drive would eliminate the corruption problems. So far, I've been lucky, and haven't had any problems. I just hope it lasts, lol. I'll post an update when it crashes again, or in a couple weeks to update my success

Just a suggestion, and I'm no expert by any stretch of the imagination... Assuming you're using Docker (you didn't say), and your plex folder is named 'Plex'. Stop plex. Rename or move the 'plex' folder in appdata. Uninstall plex. Reinstall plex. See if everything is working. If it is then you have two choices... Sit back and let plex do it's thing. Yeah, I know, a PITA, but this would be my suggestion, or... Stop plex. Delete (no need keeping it at this point) the 'plex' folder. Rename or move your new 'plex' folder, the one you created in step 4 above, to its original name. Restart plex. You should be back to where you were in step 5 above. Good Luck

Cool! That's all I really needed to know, Safety over Speed any day

My real question is... If I replace one drive (5TB to 6TB) at a time, will the new parity drive(s) be limited to 5TB, or will I get the full 6TB after I install the second drive. I'm not sure how that works, and that's my main concern. I don't want to be stuck with 6TB drives only having access to 5TB.

I'm running two 5TB 7200 RPM parity drives and want to swap them with two 6TB 5400 RPM drives due to heat issues. What's the recommended route I should take? Is it recommended to replace them both at the same time, or one at a time, if that's even possible? Seems to me that both together would be the best route to take, but I would like some feedback before I do it. TIA

iilied posted his logs from plex. Haven't seen anyone post the syslog or Docker logs. I haven't checked my syslog, but my Docker log wasn't very helpful. Plex either starts normally, or it goes into an infinite loop, something like "Starting Plex Media Server", but Plex never starts.

Downshot is after 3 days error free, then crashing, I'm bats%#t crazy, and afraid to update unRaid in fear that the next update will have the same problems. As a side note, last night, I went back to unRaid 6.6.7, but...I deleted the share, appdata, installed a SSD (cache) drive, and configured appdata to use cache only. I restored appdata from a backup, and rebuilt PLEX from scratch. Now I have the conundrum of staying with 6.6.7 or updating to 6.7.0.

When this first happened to me on May 12th, I spent the following week trying to find a fix, and it happened with the plexinc, linuxserver.io and limetech containers. I didn't try the binhex container, but I would assume it has the same problem.

Sadly, my Plex died again. May 30, 2019 17:09:57.041 [0x14b38a9f4700] ERROR - SQLITE3:(nil), 11, database corruption at line 65066 of [bf8c1b2b7a] I'm at a loss as to what to do. I'm going back to unRaid 6.6.7.

And I can accept incompatibility issues. I've been around too long to take issue with that. I think there are too many people like me, that don't use cache, and go with the default "/mnt/user/appdata/..." and then end up spending a lot of time trying to figure out what went wrong, when an incompatibility pops up. I lucked out and found your unRaid Docker FAQ (https://github.com/binhex/documentation/blob/master/docker/faq/unraid.md), and it set me on the right path. I just hope posting my findings and the link to your FAQ in this thread and a couple of others I've posted to will help others before the damage is done

Thanks! But I already knew about the backups. Be sure to not delete your newest uncorrupted database. I you restore a corrupted database, you'll be sorry! Between my movies, music and tv episodes, I have about 10,000 media files. I use the trak.tv plugin in Plex to keep my watched movies and tv episodes in check, so my biggest problem is just the time it takes for Plex to scan my libraries, and to rebuild the metadata and preview/chapter thumbnails. If this fixes my problems, and it's been 3 days now, I'm willing to take that time and have everything back in working order. I mucked around with my database backups so many times that I don't have very much confidence on their integrity at this point. Limetech support asked about the problem and I replied with a brief account on what I've seen. Hopefully this will all go away soon, and everyone can go back to being clueless and happy, like I was until 3 weeks ago, lol.

This particular problem seems to be unique to unRaid 6.7.0. After updating from unRaid 6.6.7 to 6.7.0, a lot of people are seeing this error in Plex: "There was an unexpected error loading this library". In the Plex log there will be an entry like this: "May 25, 2019 14:57:19.398 [0x7f7fa8f2b700] ERROR - SQLITE3:(nil), 11, database corruption at line 65066 of [bf8c1b2b7a]", and Plex is pretty much dead in the water. This happened to me within the first few minutes after updating unRaid from 6.6.7 to 6.7.0, back on May 12th. I fought with it for a full week and finally went back to 6.6.7. During the following week I did more research and was led to believe that the corruption problem occurred with the Plex database when the Plex /config path was on /mnt/user/appdata/plex. By configuring /config to be on a single drive (/mnt/disknumber/appdata/plex) rather than in the share file system (/mnt/user/appdata/plex), the problem wasn't happening. I can't say that this is a problem with the unRaid 6.7.0 update, or something in Plex. I do know that I've been running unRaid 6.7.0 and Plex for the past 3 days and haven't seen the problem since. I made the change as I outlined above. Specifically, I configured my Plex /config fr4om /mnt/user/appdata/plex to /mnt/disk1/appdata/plex, and I haven't had any more problems. I hope this helps, and am glad to see someone from Limetech taking interest.