Crashing Issues

[Bersquack]

Hi all,

 

Very new user here.  I installed Unraid last night on a newly built server. I followed all the instructions I could find online on initial setup.  Server seems to work fine only to find that I cannot connect to it through the web browser after it's been running for a few hours.  I attached a picture of the screen I have every time it crashes.  I really have no idea where to go from there.  Any help is tremendously appreciated.

 

I have a 6 HDD setup (2 for parity) all the same model, with 2 nvme drives for cache. Brand new mobo, CPU, RAM, etc.

[trurl]

Have you done memtest (on the boot menu)?

 

Before it crashes, go to Tools - Diagnostics and attach the complete diagnostics zip file to your NEXT post.

 

Also, see here for more on how to get us information leading up to the crash:

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=781601

 

[Bersquack]

Thanks.  I set up the first option for system log and will report as soon as it happens again.

[trurl]

12 minutes ago, trurl said:

Have you done memtest

 

[trurl]

And would still like to get complete diagnostics. It has information about your configuration that we can't get from syslog.

[Bersquack]

I tried to run memtest86 and it immediately went back to the motherboard "load screen", so I'll have to try again.  I downloaded the diagnostic zip file and attached it here.

yoda-diagnostics-20191222-0055.zip

[JorgeB]

8 hours ago, Bersquack said:

I tried to run memtest86 and it immediately went back to the motherboard "load screen"

Memtest will only work if booting legacy, not UEFI, also Ryzen on Linux can lock up due to issues with c-states, make sure bios is up to date, then look for "Power Supply Idle Control" (or similar) and set it to "typical current idle" (or similar), or completely disable C-sates.

 

More info here:
https://forums.unraid.net/bug-reports/prereleases/670-rc1-system-hard-lock-r354/

[Bersquack]

Thanks.  I just rebooted this morning. Will see how long it stays up and then try the power supply idle control.  I downloaded my syslog from last night just in case. Reading a bunch of posts you pointed to it does sound like Ryzen is a piece of work for unraid, but I'm glad there's a solution.

 

Thank you all again for all your help. I'll post later when I get back from watching Rise of Skywalker with the lady!

[Bersquack]

Well, it looks like to power supply setting did not fix my problem.  I woke up this morning to another crashed server.

 

The system log has a number of error lines like this: CPU: 3 PID: 11481 Comm: awk Tainted: G B D W 4.19.56-Unraid #1

 

From the dashboard the memory utilization is at 100% for the Log. I'm not sure what that means, but it doesn't look right.

 

What should I do next?

[trurl]

1 hour ago, Bersquack said:

The system log has

1 hour ago, Bersquack said:

What should I do next?

post that system log

 

[Bersquack]

OK guys.  I went through all the step in the post linked by ti-ti jorge, redid everything and it seems to be stable now.  Been running a bit over two days straight, so I think I'm good new.  However, it looks like I might be having BTRFS issues with my nvme drive, so I'll go read into that. 

 

I'll mark this post as solved for now.

[trurl]

50 minutes ago, Bersquack said:

However, it looks like I might be having BTRFS issues with my nvme drive, so I'll go read into that. 

post diagnostics

[Bersquack]

Here it is.

yoda-diagnostics-20191226-2103.zip

[JorgeB]

There are some checksum errors, run a scrub on the cache pool, make sure there are no uncorrectable errors.

[Bersquack]

9 hours ago, johnnie.black said:

There are some checksum errors, run a scrub on the cache pool, make sure there are no uncorrectable errors.

I just ran a scrumb and there are 2 uncorrectable errors.  I do not know if it's related, but there are files "stuck" in the cache that do not move when the mover runs. my guess is that I was transferring those when the server crashed a few days ago.

[JorgeB]

Yes, most likely those files are corrupt, btrfs doesn't allow copy of known corrupt data, you can see in the syslog, corrupt files will be identified by the scrub, you can delete and restore those files from backups (or re-download)

 

 

 

[Bersquack]

I stopped and started the array again with the cache (backed up appdata first), pre-cleared the cache drives, restored appdata and it looks like I'm all set.

[ramjam293]

I'm new to Unraid too and think I'm having similar issues this person had.  My system will run fine while doing a parity check.  And it will run fine as long as I'm using it.  But when it's not doing anything, it crashes.  The light on the outside of the case blinks.  I did the memtest and everything came back okay.  Attached is my diagnostics file.  I'm hoping someone can please help.  Thanks in advance.  

 

tower-diagnostics-20210207-1632.zip

[JonathanM]

46 minutes ago, ramjam293 said:

I'm having similar issues this person had.

I'm guessing this is probably the issue.

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=819173

If not, you would be better to start your own thread instead of posting into someone else's already solved thread.