Honest, I'm trying to find the resolution to each of my issue before posting but some of them are real head scratchers. I have an interesting problem going on where randomly the server will become unreachable. I can log onto the console without an issue but I can't even ping the box. It does seem to happen when I'm configuring my Plex container. I did have issues installing Plex but after a container removal and deletion of the associated application files, I was able to successfully install.

 

This is an interesting problem. Perhaps there is an internal crash occurring effecting the network interface but I'm not sure how I would go about determining that and is it possible that such a network failure could be related to a docker container?

Please see Need help? Read me first!, and attach the diagnostics zip.

Appreciate that and have taken the first paragraph to heart. I'm breathing and will post up diagnostics before a restart after the next crash.

Well, I thought things had stabilized as I was able to run through the server and applications, configuring things quite smoothly for about an hour then the network connection dropped again. Attached are a syslog.zip and diagnostics.zip; hoping someone can help shed some light.

 

For what it's worth, I ran through the diagnostics gathering via 'powerdown' at the console then reset the server. I attached over the network to grab the files from the flash drive and it almost immediately crashed again. I powered off and pulled the USB stick to grab the files.

syslog.zip

tower-diagnostics-20160904-1312.zip

Diagnostics includes the syslog so there is no need to post both.

Thanks, noted for next time. A quick update, I am wondering if it is related to the shares in some way. Booting around the server, looking at configurations and fine tuning the installed applications seems to be pretty stable. It's almost as soon as I access the file system in some way that the problem occurs. For example, I've been changing the location of my TV episodes in SickBeard. That took about 45 minutes and everything seemed fine. I had a downloaded episode that was not post processed  (I think there's some configuration missing relative to the script?) so I manually processed from within Sickbeard. As soon as I selected the folder to browse the problem appeared; the server is now inaccessible.

Lots of this in your syslog: DMAR:[fault reason 06]

 

Yeah, that's what I could make out as well. I did find that same thread just not sure what my next steps should be. I suppose I could start with a motherboard bios update.

To close the loop on this one. I started to zero in on the network adapter as the server seemed stable even after what I thought was a crash, it just wasn't accessible over the network.  I went through a number of potential issues including bios version and network adapter drivers; that led me to the PCI adapter. I had issues with the onboard adapter a couple of years ago. I installed a PCI adapter and all has been running fine on v5.x ever since. It appears that v.6.x is much more sensitive to the number of adapters installed. After removing the PCI adapter and plugging back into the onboard NIC all was stable for a couple of hours of Plex this evening. I'll try disabling the onboard and reinstalling the PCI adapter tomorrow.

 

 

To close the loop on this one. I started to zero in on the network adapter as the server seemed stable even after what I thought was a crash, it just wasn't accessible over the network.  I went through a number of potential issues including bios version and network adapter drivers; that led me to the PCI adapter. I had issues with the onboard adapter a couple of years ago. I installed a PCI adapter and all has been running fine on v5.x ever since. It appears that v.6.x is much more sensitive to the number of adapters installed. After removing the PCI adapter and plugging back into the onboard NIC all was stable for a couple of hours of Plex this evening. I'll try disabling the onboard and reinstalling the PCI adapter tomorrow.

Typically, you have to disable the onboard adapter in the BIOS to make sure unRAID uses the other one.