System unresponsive, what next?

[superloopy1]

Hi, i'm in the middle of a parity sync and i've lost the webgui, argh!

This has happened a few time on this particular SM server and i've never known how to sort it and try to regain control, nearly always ending up pulling the plug.

Is there an agreed 'method' to try and identify whether the system is still running or not, this server runs headless.

I'm loathe to do that in this case as it's been running for well over a day with about, i guess, another 8hrs to go, hence the question.

[superloopy1]

Happened again!

I'm in the middle of doing a disk 'clear' and same system again unresponsive. I'll let it be for now but is there anything (?) i can do to try and identify the cause?

Since last hangup i now have a monitor and keyboard hooked up so with a bit of guidance i can do basic cmd line, i suppose ... any logs i can look at from cmd?

[superloopy1]

And again!! System crashed overnight during/following parity check to correct previous lockup. I THINK this related to 6.7.2 upgrade recently. Nothing else has changed. How do I regress safely? Also I have noticed that my dockers, used infrequently I must add, can no longer connect within my home network. I have spent a lot of time checking firewalls etc and can see no changes there. Router hasn't been fiddled with at all so I"m guessing that something had been changed here too that I'm not aware of?? So ... maybe going back a release will fix things but how do I do this safely? What needs to be secured beforehand ... 

[JorgeB]

If the update was done with the GUI you can downgrade to the previous release going to Tools -> OS Update

[superloopy1]

Yes it was done via GUI. So nothing to save, settings etc, just go back? Busy running memtest atm but I'll regress as my next course if action. If no joy then I'll bang in a bigger PSU although 650 should be more than capable if running 8 disks, 3 fans and a couple of data cards (could remove one, do they suck much power?)

 

[Vr2Io]

2 hours ago, superloopy1 said:

do they suck much power?

Negative. Suggest running Unraid in safe mode after memtest.

 

I suspect problem on some disk which cause hang. During hang, does telnet/SSH access still available ? Pls attach diagnostic or check all disk SMART.

[superloopy1]

Thanks!

Ran memtest for 24hrs without problems.

I've done a couple of things since ... SMART tested all drives, ok

                                                   ... XFS checked all filesystems, ok

                                                   ... setup a syslog server on separate server, should help

                                                   ... removed unassigned devices AND preclear plugin (read that they are incompatible, is this correct?, may not have been helping!)

Server HAS stayed up since so not able to telnet/ssh to check but each time the system has crashed previously the hardware has remained running.

 

I have another issue that my dockers, which previously ran without issues, now will not connect to their corresponding webui ..s

On each attempt i'm getting 'can’t establish a connection to the server at 192.168.1.15:xxxx, please try again messages' where xxxx is relevant to each individual docker.

What has happened recently on the networking front that i need to reset/amend??

I've checked as much as i know how to check, router, firewall settings etc and cant see a problem.

 

Edited July 26, 2019 by superloopy1

[superloopy1]

Update ... system has just crashed again so i am able to say that ssh/telnet isnt accessible. The server IS still running although console is unresponsive so i'm guessing it's yet another parity check it'll need. I'm sure that this isnt going to end well and that sooner or later disks are going to be hit so need another 10TB parity to slot in. At the time it went awol i was editing docker setups so am still no clearer on what is causing this? I think i will rollack to previous 6.7 version if no other ideas.

 

I have attached the syslog server eof if anyone can take a look see if its flagging any problems ..

syslog-sessionbeforecrash.log

Edited July 26, 2019 by superloopy1
attach syslog server files

[superloopy1]

Can i assume that there's nothing to interest anyone in this diagnostics file/

I am now about to regress to 6.6.7 as what i'm experiencing, unbeknown to me, server crashes, dockers not starting, mo access to internet from docker, appears to be an ongoing problem which manifested itself around the 6.7 release mark. I DO read the forum posts but somehow this one needs to be made more visible in my opinion.

So ... in advance of downgrading I've deleted ALL of my dockers but see that appdata retains all of the container information, how do i get rid of this and clean out fully? Tried to delete from within a windows share and it doesnt allow me due to permissions. I'm not a linux guy so can some kind soul advise me of the linux commands needed to fully tidy out my appdata share and start again!! I cant see an early resolutioin to all of this and need to be back prior to 6.7 when everything worked just as expected for my sanity!

Edited July 28, 2019 by superloopy1