Server Version#: 1.16.4.1469
Seemingly unknown pattern of days that it happens, but it feels like every few days my server will crash just past midnight local time. Start the server back up and everything is just fine until the next night, or a few nights later, it’ll crash again. Not sure what’s causing it.Plex Media Server Logs_2019-08-11_01-22-06.zip (5.8 MB)
Just noting that this has been happening for quite some time, but never noticed the outages until I set up Tautulli to email me when plex went down. Our TV watching habits are very sporadic, so I would never notice that plex was down until we went to watch something.
If you would be kind enough to please go back and turn VERBOSE logging back off.
Verbose is counter productive in 99% of the cases.
In this case, it is extremely counter productive as we can only see 2 minutes of elapsed time.
The entire event is gone.
With VERBOSE off and DEBUG on, Please capture again as soon as you spot it.
If you wish, and are willing to do a little shell command line work, we can increase the number of log files retained to increase your capture window.
Ok, will do. Sorry, I had verbose on for something else. I can happily do the shell work too, whatever you need, I can send it along. May be a day or two before I get anything. Some days it’s every night, others it takes a few days.
I’m curious as a QNAP user while you wait for this to happen again, have you tried to clean your bundles and optimize your database? It sort of sounds like a scheduled event.
I think my QNAP does QTiering just after midnight as the only significant cron job,
but metadata updates later around 2 or 3am.
Good luck!
…have you specified when your scheduled maintenance can start/stop?
I’ve optimized my plex DB quite a few times, no change. I’ve never bothered with clean bundles as it’s already a scheduled task according to the UI. I have had this issue for a long time. I only have a single tier of storage so QTier shouldn’t be doing anything. This is just a 4bay NAS.
My scheduled tasks are set to start at 2am and stop at 7am, so there’s “no” correlation there
I’ve also had issues saving changes to my library config - such as I have two folders in my photos library, but when i try to remove one of the folders, it pauses for quite some time then tells me " Your changes could not be saved." which is super helpful. My plex db is old so I’m sort of hesitant to recreate it, but maybe that’s the best thing to do at this point.
Right on. My NAS does a lot of tasks just after midnight that are not from Plex.
I wonder if your crontab has a job going then. Smart status and fsck nominal?
I have the TVS 1282 with the latest firmware and Plex version.
I too am having random "stops’
Mine are around 10.30pm but not always.
I just thought that maybe that’s the time when it has been working it’s hardest and just gives up!!
In my situation Plex app remains on in the NAS.
I just stop it and start again and all good …
For the OP could something be “filling up” or a
strange cache setting?
Cache location moved/full.
Of course it hasn’t crashed yet. I don’t seem to have much happening at midnight in cron that’s of any importance, other than the smart simple scan and a few other things that are 100% default on the NAS. Certainly nothing I’ve added in there that’s of any volume.
But if it magically fixed itself, fine by me, I just feel like it’ll happen again soon.
Oh - sorry, I don’t use any cache settings, tiering, etc. Just a plane ol’ NAS with a bunch of 8tb disks that only really do plex and time machine. My use case for the NAS is basically Plex and bulk storage.
Interesting.
I am running 4.4.1.0998 (yes, the beta) and not experiencing any difficulties now on the TVS 1282 which runs constantly.
In QTS 4.4.1.0998, QNAP fixed a problem with the Virtual Switch. This was causing major interruption by declaring the primary adapter to be flagged as “offline”. When doing development work as this occurred, my sessions would be terminated.
Perhaps this is also impacting your systems?
I’m actually running 4.4.1.0998 as well on my TS-453A but I upgraded it a bit ago (before some of the crashes). I do recall messing with my virtual switch as it kept doing some weird stuff with default gateway selections. I run two of my 4 ports in a port channel though. haven’t had issues with the adapters going offline.
I have the 4 ports as one channel.
I have the 10GbE on the other.
The v-switches went offline (renders adapters offline as well).
Root cause here was the v-switch code.
Finally crashed last night. I thought it was finally happy.Plex Media Server Logs_2019-08-23_12-25-35.zip (4.2 MB)
Do you have an earlier version handy?
I see PMS scanning something it should (AFAIK) be ignoring.
librarySectionID=5&metadataItemID=517248&metadataType=13&state=-1&mediaState=idle (15 live) GZIP 5ms 166 bytes
Aug 23, 2019 00:33:49.509 [0x7faa1c855700] DEBUG - Checking by hash to see if we can find a match with f65d8157019ac3b1fcb513724fbe45a0c6e927f5 (display offset: 0, not part -1)
Aug 23, 2019 00:33:49.510 [0x7faa1c855700] DEBUG - We found a hash match for [/share/CACHEDEV1_DATA/Photos/@Recently-Snapshot/Replication-Schedule0/2018/School_2018_Logan/MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 017 c 1QTY.jpg] which was [/share/CACHEDEV1_DATA/Photos/2018/School_2018_Logan/MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 017 c 1QTY.jpg].
Aug 23, 2019 00:33:49.511 [0x7faa45dcb700] DEBUG - Completed: [127.0.0.1:56288] 200 GET /library/changestamp (15 live) GZIP 4ms 469 bytes
Aug 23, 2019 00:33:49.515 [0x7faa1c855700] WARN - Duplicate media part detected [/share/CACHEDEV1_DATA/Photos/2018/School_2018_Logan/MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 017 c 1QTY.jpg], but we'll scan it anyway.
Aug 23, 2019 00:33:49.818 [0x7fa9c3159700] DEBUG - Activity: updated activity 6bd3cd7f-12d6-47b9-bd3a-8c0e8611fd71 - completed 34.000000% - Generating video preview thumbnails
Aug 23, 2019 00:33:49.912 [0x7faa1c855700] DEBUG - Added new metadata item (MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 017 c 1QTY) with ID 517266
Aug 23, 2019 00:33:49.957 [0x7faa1c855700] DEBUG - * Scanning
Aug 23, 2019 00:33:49.957 [0x7faa1c855700] DEBUG - Looking for path match for [/share/CACHEDEV1_DATA/Photos/@Recently-Snapshot/Replication-Schedule0/2018/School_2018_Logan/MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 019 c 1QTY.jpg]
Aug 23, 2019 00:33:49.965 [0x7faa1c855700] DEBUG - Checking by hash to see if we can find a match with c12b067e913b33fce2aaa282ffafe0aadce669f0 (display offset: 0, not part -1)
Aug 23, 2019 00:33:49.965 [0x7faa1c855700] DEBUG - We found a hash match for [/share/CACHEDEV1_DATA/Photos/@Recently-Snapshot/Replication-Schedule0/2018/School_2018_Logan/MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 019 c 1QTY.jpg] which was [/share/CACHEDEV1_DATA/Photos/2018/School_2018_Logan/MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 019 c 1QTY.jpg].
Aug 23, 2019 00:33:49.969 [0x7fa9c3735700] DEBUG - Updating metadata item (save) (MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 017 c 1QTY) with ID 517266
Aug 23, 2019 00:33:49.971 [0x7faa1c855700] WARN - Duplicate media part detected [/share/CACHEDEV1_DATA/Photos/2018/School_2018_Logan/MYS 10x14 webdl Stubblefield010219115508623 ISKECLCQPJYNi 019 c 1QTY.jpg], but we'll scan it anyway.
I just downloaded 1.16.5.1554 because plexpy (or whatever they call it now) told me there was a new version, but generally i don’t keep earlier versions around. I assume it’s available online somewhere if you’re going to ask me to downgrade?
Given the timing of when it dies, I’m going to also suggest the Geeky thing.
Clean it out. PMS does its heaviest work after midnight (Scheduled Tasks), generating the most heat. QTS could kill it and/or the whole system could reset & reboot if it gets too hot.
True, i can try that too. I do have scheduled tasks to start at 2AM, so not entirely sure what would start at midnight other than anything the QNAP is doing itself.
Funny that this is about pictures though - i have a folder defined in my Photos library (I have 2 defined) that I want to remove, but every time I try, i get the whole “your changes could not be saved” thing. Just tried it again and uploading a fresh log set. Plex Media Server Logs_2019-08-23_13-53-32.zip (4.5 MB)
Wondering if the two things are related? I’ve tried to optimize my db a few times.
