My Unraid server had been running for months without these issues until recently. Sometimes when I go to the web interface the response is that there is nothing there at that address. I have tried using both http://tower and the IP address directly. Both had the same results. Sometimes I will just let it sit there and the page will eventually load and other times it will not. I get a connection refused error.

 

On top of the web interface my network shares will also not be reachable. There seems to be no rhyme or reason as to when it happens. I could be actively using the web interface or a network share and the next click will result in a connection error. When I plug in a monitor and look at the server itself it appears to be operating without issue though I haven't looked at any logs. I don't actually know where I would find the relevant logs for this.

 

I use the server very occasionally and I couldn't tell if the upgrade to 6.7.2 had anything to do with these issues or not.

I have also had issues with some Docker containers that stopped working correctly but I'm not sure if that's related.

 

What do I need to provide to get a clearer picture of what is going wrong?

The first thing would be to provide the system diagnostics zip file (obtained via Tools->Diagnostics) that includes the period when you are experiencing such a problem (ideally also giving a time so that anyone looking through the logs knows which part to concentrate on). 
 

it is also quite possible that this is nothing to do with Unraid but is something external like your router that is playing up.

tower-diagnostics-20190730-0344.zip

I have included the zip file. I know it was happening today (the 29th) around 3:15pm and 3:48pm. 

9 hours ago, itimpi said:

The first thing would be to provide the system diagnostics zip file (obtained via Tools->Diagnostics) that includes the period when you are experiencing such a problem (ideally also giving a time so that anyone looking through the logs knows which part to concentrate on). 
 

it is also quite possible that this is nothing to do with Unraid but is something external like your router that is playing up.

 

It seems like the issues is on my server than the router. Other devices on the network don't seem to have problems like these. They and the internet are all reachable. I have tried other ports on the router, other ports on the server, and even different cables.

There is nothing of note on 29 Jul. However, on previous days you have a lot of these:

Jul 25 02:48:43 Tower kernel: bond0: link status definitely down for interface eth0, disabling it
Jul 25 02:48:43 Tower kernel: bond0: making interface eth1 the new active one
Jul 25 02:48:43 Tower kernel: device eth0 left promiscuous mode
Jul 25 02:48:43 Tower kernel: device eth1 entered promiscuous mode
Jul 25 02:48:43 Tower kernel: e1000e: eth1 NIC Link is Down
Jul 25 02:48:43 Tower kernel: bond0: link status definitely down for interface eth1, disabling it
Jul 25 02:48:43 Tower kernel: device eth1 left promiscuous mode
Jul 25 02:48:43 Tower kernel: bond0: now running without any active interface!

That suggests NIC problem and/or router problem and/or cable problem.

 

Given you said there were issues on 29 Jul but there was nothing seen in syslog, I would assume it's more likely a router / cable problem.

 

 

 

 

The server has two onboard ethernet ports and and additional four ports in a PCIE card. I have tried different cables in different ports and it still happens. I have tested with bound ports and unbound ports. Still happens. 

 

Though I agree that the router could be a problem I would prefer having a more definitive test to confirm that. One of the things I wanted to do with the server was run PFSense on it but I'm not going to do that if I am not certain the server isn't having some kind of problem. I can't think of any other tests or logs to look at on the server though. I may be able to try something with an old router I have laying around to test if the router is causing any trouble.

Be sure to reboot the router and any switches used in the network.  Sometimes, 'Green' switches can have problems with dropping connections.  Any possible issues with network cable length?  If it is a long run, consider using a CAT6 cable.   Another point, not using your router as a switch will improve its longevity and reliability.  (Heat is the concern here.  The more active ports, the hotter your router will get.) 

I pulled out an old router and started testing. I tried all six available network ports on the server and different network cables. It really does look like the router I used was causing the unresponsive web interface and the spontaneous loss of network shares.