Hi everyone, I'm almost at the end of my tether with a recent issue which I haven't been able to solve in 3 days of trying. I'm hoping some kind genius will be able to help me.

 

Unraid specs:

• Intel Xeon E3-1245 v2
• 4x8GB DDR3
• 1x 500GB Samsung Evo (Cache)
• 2x 6TB Toshiba HDDs (Parity & Disk1)
• 1x 2TB Seagate Barracuda HDD (Disk2)
• Asrock Z77 Extreme6 motherboard
• No other expansion cards
•  

Symptoms:

• Connections to the Unraid GUI timeout after 15 - 20 seconds of doing anything but a page refresh will immediately bring it back up. This means practically docker updates seemingly stop part way through but a page refresh shows the command was still successful, the webterminal drops connection but immediately reconnects (repeat endlessly), Binhex-krusader VNC portal freezing up until a page refresh which brings it back to life, Unbalance plugin freezing during a transfer but a refresh brings it back after ~5 seconds, etc.
• At the same time, SMB shares sometimes do not load for what feels like 20 - 30 seconds before eventually connecting. This occurs on both my Pop_OS! (Ubuntu) desktop and laptop.
• No other devices on the same network are experiencing any issues with connecting to one another or the internet. I can load pages and browse with no issues.
• Pinging the server hostname or IP address during these connection issues show 0% packet loss.
• All of this started around 1 month ago and did not coincide with any hardware changes. Prior I've had no similar issues on the same box which has been running for 24+ months.

 

Troubleshooting steps I've already tried with no change:

• Swapping the ethernet cable between Unraid and switch with a new cable.
• Temporarily adding a USB NIC and connected to Unraid through it instead.
• Reinstalling the whole router VM (pfSense) on my proxmox server from scratch and restoring pfsense from a backup config.
• Running all 4 memory DIMMs through MEMTEST for 1 pass resulting in 0 errors. Swapping them out for a spare set of 2x4GB DDR3 also did nothing

 

The closest I've come is the issue being somehow related to fact my pfSense router is virtualised as a VM under Proxmox and this is screwing with the NTP server. The reason is Unraid's diagnostic syslog.txt file had this line:

 

ntpd[1908]: kernel reports TIME_ERROR: 0x41: Clock Unsynchronized

 

Attached are the full diagnostic logs. Note that there was a freeze around 17:06 before these logs were taken but nothing is showing from that time.

 

And my pfSense console also showing some errors related to NTP. But Unraid is set to use the default Google servers as it's timeserver, and could a screwed up system time really lead to these issues?

 

Any help is appreciated!

sollyserver-diagnostics-20211126-1706.zip

I almost hate to jump into one of these types of problems...  (But I am sometimes a gluten for punishment!)

 

Go to the Dashboard tab of the GUI.  Make sure that the 'Processor' section is visible.   Now look at the graphs.  They should move continuously. 

Edited November 26, 20214 yr by Frank1940

It's one of those inexplicable problems that seem to always end up having a cause and solution that are so obvious, but only once you find it..

 

The processor graphs on the dashboard move for around 30 seconds before freezing and resume after a page refresh. If it makes any difference, disk2 (2TB Seagate Barracuda) is showing a Reallocated sector count of 399 in it's SMART report plus the disk logs have this seemingly unrelated error:

Nov 26 20:39:38 Sollyserver wsdd[12881]: set_multicast: Failed to set IPv4 multicast
Nov 26 20:39:38 Sollyserver wsdd[12881]: Failed to add multicast for WSDD: Address already in use
Nov 26 20:39:38 Sollyserver wsdd[12881]: set_multicast: Failed to set IPv4 multicast

 

However I have move all data from the disk in preparation of removing it so I can't see how that would impact webGUI in this way. I don't plan on replacing the disk as there is more than enough space on the remaining 6TB.

Googling your issue, I would surmise that you have a networking issue. 

 

I would go first to the      Settings   >>>     Network Settings         and get a screen print or a screen capture of the current settings. 

 

What i would do next would be to rename these two files on your flash/boot drive  (example: network.cfg.bak).

 

/boot/config/network.cfg

/boot/config/network-rules.cfg

 

(The second one may not exist...)  Then reboot the computer.  Now set things back up again and see if that fixes the problem.  Note if you had a static IP address, it may have changed as the server will now be using DHCP to get its address and, depending on how your router is set up, it could be different.

Edited November 26, 20214 yr by Frank1940
EDIT: don't setup any IPv6 stuff unless you know you really require it.

I really appreciate you helping me out. Since my last reply I think I've found the cause and solution and you were correct that it's network related. My Unraid server's network settings had every VLANs available in my network configured (6 in total) just in case I needed to assign them to anything e.g. a VM or docker container. I removed 4 unused VLAN configs leaving just 1 for my single Win10 VM and 1 in use with ~5 docker containers. Thankfully my issue seems to have gone away but I will come back and update if not.

 

It's probably unrelated, but in the process of troubleshooting I read somewhere that the Recycle Bin plugin could cause some similar issues so I deleted that, along with a handful of other unused plugins (e.g. a WOL for VMs plugin). The Recycle Bin plugin wasn't even enabled which is why I doubt it's related but I wanted to mention it anyway.