6.12.1 Dashboard issue -- Related to ZFS cache pool error?

[tiny-e]

 

CPU section stopped updating.   I  (thinking it might be browser issue) cleared the browser cache and restarted the browser (chrome).  Now when I load up the dash this is what it shows me... 

Not sure if this is bug-report worthy.   But curious to know what might be causing this. 

Edited June 26, 2023 by tiny-e

[xyzeratul]

I had this bug a few times before:

 

• Once was in 6.11.2, the cause is the http proxy setting mess up, clean the proxy setting fix this.
• The recent one is when I switch to 10GB SFP+ connecting, maybe the NIC or the cable issue, I am still fixing it, but I do know this happen when network connecting is bad.

So I suggest you check your network hardware setup and network settings, to root out any setting issue, you best bet is make a new USB boot drive to see if the problem is still there, because use safe mode can still have this bug( bad proxy or network settings are still present in safe mode)

 

 

[tiny-e]

On 6/23/2023 at 8:34 PM, xyzeratul said:

I had this bug a few times before:

 

• Once was in 6.11.2, the cause is the http proxy setting mess up, clean the proxy setting fix this.
• The recent one is when I switch to 10GB SFP+ connecting, maybe the NIC or the cable issue, I am still fixing it, but I do know this happen when network connecting is bad.

So I suggest you check your network hardware setup and network settings, to root out any setting issue, you best bet is make a new USB boot drive to see if the problem is still there, because use safe mode can still have this bug( bad proxy or network settings are still present in safe mode)

 

 

Thanks -- I'm not using any proxy and running on the same ol' gig-e network that's been here virtually forever.   

 

[tiny-e]

So -- A reboot restores the CPU graph function to the dashboard

nimbus-diagnostics-20230626-1204.zip

 

I'm also seeing this error regarding one of my cache pool drives (sdl).  I recently just remade the cache as a ZFS pool (mirror) and I thought things went just fine.  All systems seem to be working as they should, but the error is concerning.

Jun 26 12:02:04 Nimbus kernel: critical medium error, dev sdl, sector 489213696 op 0x0:(READ) flags 0x700 phys_seg 2 prio class 2
Jun 26 12:02:04 Nimbus kernel: zio pool=cache vdev=/dev/sdl1 error=61 type=1 offset=250476363776 size=131072 flags=180880
Jun 26 12:02:04 Nimbus kernel: sd 1:0:5:0: [sdl] tag#1189 UNKNOWN(0x2003) Result: hostbyte=0x00 driverbyte=DRIVER_OK cmd_age=0s
Jun 26 12:02:04 Nimbus kernel: sd 1:0:5:0: [sdl] tag#1189 Sense Key : 0x3 [current] 
Jun 26 12:02:04 Nimbus kernel: sd 1:0:5:0: [sdl] tag#1189 ASC=0x11 ASCQ=0x0 
Jun 26 12:02:04 Nimbus kernel: sd 1:0:5:0: [sdl] tag#1189 CDB: opcode=0x28 28 00 1d 10 4f 00 00 01 00 00
Jun 26 12:02:04 Nimbus kernel: critical medium error, dev sdl, sector 487608064 op 0x0:(READ) flags 0x700 phys_seg 2 prio class 2
Jun 26 12:02:04 Nimbus kernel: zio pool=cache vdev=/dev/sdl1 error=61 type=1 offset=249654280192 size=131072 flags=180880

 

[JorgeB]

11 minutes ago, tiny-e said:

I'm also seeing this error regarding one of my cache pool drives (sdl).  I recently just remade the cache as a ZFS pool (mirror) and I thought things went just fine.  All systems seem to be working as they should, but the error is concerning.

It's logged as device error and that SSD shows a high number o read raw errors, other one is a zero, so likely not a good sign, consider replacing it if errors continue, you can also check here for better pool monitoring.

[tiny-e]

2 hours ago, JorgeB said:

It's logged as device error and that SSD shows a high number o read raw errors, other one is a zero, so likely not a good sign, consider replacing it if errors continue, you can also check here for better pool monitoring.

Thanks -- so from the log, its the sdl drive thats failing?   Just want to make sure I'm reading things correctly. 

 

 

[JorgeB]

Correct, first pool member.

[DeekB]

I upgraded to Unraid 6.12.3 recently and found that I also have the same problem with the graphs displaying no info.

 

I found by disabling the VM manager the graphs work\come back, I am not sure why as I dont even have any VM's set up, seems like a bug but as I dont use VM's I will just leave this disabled for now but thought it was worth mentioning as a google search brought me to a couple of forum posts (one where this issue was spotted in 6.12 RC6)

Edited July 17, 2023 by DeekB