After uprade to 6.3.5 unraid gui become unresponsive

[thatnovaguy]

Recently upgraded from 6.3.4 to 6.3.5 and all seemed perfectly fine for the first few hours, then the gui suddenly became unresponsive. Everything else (dockers, active vms, smb) seems to be working fine. I've tried accessing the gui from multiple browsers and devices with no avail as to rule out secondary devices. Currently the only way I can regain access to my gui is to reboot the server from ssh which I hate to do as it always forces a parity check.

Is there a way to pull down my log file without the gui? I've had unraid for a long while but I'm pretty green when it comes to dealing with system failures.

Also, is there a way to reboot from terminal without causing a parity check?

Sorry for the newbie questions

Thanks

-Andrew

[JorgeB]

6 minutes ago, thatnovaguy said:

Is there a way to pull down my log file without the gui?

 

On the console/SSH type diagnostics

[thatnovaguy]

Well, I ran the diagnostics command and it's been sitting with "Starting diagnostics collection... " for about an hour now. I'm guessing that's not normal?

[JorgeB]

8 minutes ago, thatnovaguy said:

I'm guessing that's not normal?

 

No, try typing reboot but you'll probably need to reset the server.

[thatnovaguy]

Reboot didn't work, though oddly enough, the mobo speaker beeped like it did work. Powerdown did work, though. I have the system log and am seeing mce errors. Where exactly are they logged? I have the plugin installed through the nerdpack plugin. Thanks again for all your help!

htpc-syslog-20170530-1509.zip

[Squid]

This is the mce error:

May 30 15:07:58 HTPC root: Fix Common Problems: Error: Machine Check Events detected on your server
May 30 15:07:58 HTPC root: Hardware event. This is not a software error.
May 30 15:07:58 HTPC root: MCE 0
May 30 15:07:58 HTPC root: CPU 8 BANK 12 
May 30 15:07:58 HTPC root: MISC 4937e004086 ADDR 89daa6700 
May 30 15:07:58 HTPC root: TIME 1496156803 Tue May 30 11:06:43 2017
May 30 15:07:58 HTPC root: MCG status:
May 30 15:07:58 HTPC root: MCi status:
May 30 15:07:58 HTPC root: Corrected error
May 30 15:07:58 HTPC root: MCi_MISC register valid
May 30 15:07:58 HTPC root: MCi_ADDR register valid
May 30 15:07:58 HTPC root: Threshold based error status: green
May 30 15:07:58 HTPC root: MCA: Generic CACHE Level-2 Eviction Error
May 30 15:07:58 HTPC root: STATUS 8c2000400007017a MCGSTATUS 0
May 30 15:07:58 HTPC root: MCGCAP 1000c14 APICID 20 SOCKETID 1 
May 30 15:07:58 HTPC root: CPUID Vendor Intel Family 6 Model 45

It was corrected automatically, but not much to go on as to what it exactly was, but appears to me to be something related to the L2 cache on the cpu.  (And it happened prior to the start of the syslog (probably during the initial boot screen)

[thatnovaguy]

Well I guess that doesn't explain the gui problem, then. What should I try next?

[JorgeB]

Boot in safe mode to see if it helps.

[thatnovaguy]

Ok, I'm going to wait until the gui becomes unresponsive again to do so. I'm guessing that would determine whether or not the problem is being caused by a plugin/docker etc?

[JorgeB]

Yep, but don't forget the MCE means there's a hardware problem somewhere, it may or not be causing your problems, but at least you can rule some things out.

[thatnovaguy]

Just a quick update, as of right now the GUI is still functioning normally. I have no idea what has changed.