dchamb

Frank1940, Thanks! It is now set if this happens again. I've attached the syslog from the first and only time it failed. Not sure what to make of it except I see a general protection fault (which to me usually indicates a memory leak) and an error on the SSD drive, but nothing further shows up on the SSD. Ran SMART tests and it shows fine. Dale tower-diagnostics-20190712-1413.zip

That's the problem. This has been the only time it happened so I have no basis to determine when it might happen again. If it does, I believe there is close to a zero chance that getting a Diagnostics file saved since everything appeared to be down before the reboot. There was a syslog from before and after the crash though.

I upgraded to 6.7.2 from 6.6.6 about a week ago and all seems well. Before that I replaced the Marvell SATA controllers with an LSI Broadcom SAS HBA. Disk access times are better too. I did have one anomaly. Sometime during the early morning I could not access shared from any computer on my network. The webgui was still running but show zero CPU utilization. I tried to shutdown, but all I got were the standard shutdown warnings and nothing happened. After about 45 minutes I forced it down and restarted. Everything came back up but it had to do a parity check. I've never had this happen before. Dale

Here is one I am looking at currently LSI Broadcom SAS 9300-8i 8-port 12Gb/s SATA+SAS PCI-Express 3.0 Low Profile Host Bus Adapter https://www.amazon.com/dp/B00DSURZYS/ref=cm_sw_r_cp_apa_i_3c3fDbFCRSQST

Thanks johnnie.black. I still am uncertain if I should scrap this controller card for a non-Marvell based one. I see SAS controllers out there but I'm not familiar with them. And if I change my controller will all the existing drive assignments be preserved?

Where would I find it?

I have an unRAID version 6.6.6 system with an I/O Crest 4 Port SATA III PCI-e 2.0 x1 Controller Card Marvell 9215 (based on the Marvell 88SE9215). When I tried to upgrade to version 6.7.0 I lost 3 drives which were attached to this controller card. I reverted back to version 6.6.6 and the drives reappeared on the array. I am reluctant to try going to 6.7.2 because I don't know if this problem still exists, plus I've been aware of the posts on database corruption issues. My question is this: should I replace the controller with a non-Marvell based controller and try again? If so, what are the recommended controllers for SATA drives that should not have this problem with 6.7.x? Thanks Dale

Do you have any Marvell SATA controllers? I have one where two of my array disks are attached and had the same problem going from 6.6.6 to 6.7.0. I rolled back and had no problems. I'm guessing based on your experience the problem did not go away with 6.7.2.

I agree 100%. It is impossible for anyone to test for every condition and they have responded very quickly to address problems as they surfaced. unRAID OS is the best thing going!

Are you using a Marvell controller? I upgraded my system from 6.6.6 to 6.7.0 and two of my array drives went missing. They are attached to a Marvell controller. I reverted back to 6.6.6 and everything was fine. I'm waiting to see if 6.7.2 resolves this problem.

Does 6.7.2 solve the Marvell missing disks problem?

I recently moved my Plex system to use the linuxserver.io plex docker. All seems well except occasionally I see a half second blip on screen during playback. I thought it might be related to the transcode folder, since it looks like it is going to the array. I tried to move it to tmp, but that made the problem worse, even causing some content to not play at all. I also noticed that using Plex.tv webgui it tends to skip ahead 5 seconds or so several times. Any thoughts on what is causing this? I had no problems like this running Plex under the PhaZe plugin. Dale

I had a bad experience going to 6.7.0. Two of my array drives showed up as missing. I rolled back to 6.6.6 and all was well. I will not be updating to 6.7.0. I would encourage you to roll back as well. There's a good chance nothing was corrupted. But in the event there was, it will be a matter of rebuilding the Plex appdata and not the media files. I realize this may not be a small feat if your library is as large as mine, but it beats the alternative. I do wish they would test these releases more diligently before making them available. Dale

Server Version#: 1.14.1.5488-cc260c476 I know the PhAzE Plex plug has been deprecated but does anyone know if it will still run under Unraid 6.6.6? Seems that the Plex server keeps crashing and I’m wondering if its now time to make the move to the Docker container. Also, I haven’t been able to get to the Plex logs. Dale

Monthly parity check shows no errors! Event: unRAID Parity check Subject: Notice [TOWER] - Parity check finished (0 errors) Description: Duration: 23 hours, 49 minutes, 14 seconds. Average speed: 139.9 MB/s Importance: normal Plus that 12TB WD Gold drive has improved the elapsed time of the parity check. I've been very happy with those drives. Dale