Hello!
I had some troubles in the past, and one today too, with a pool made by SATA drives.
Eventually resolved, except for the today’s issue, but I noted that everytime an issue arise with one of the two pools hosted on the same machine, the TrueNAS Scale web pages > Dataset and > Storage Dshboard become totally unresponsive, entering in an infinite loop.
This prevent access/monitor/operations to the non affected pools.
Has anyone experienced such kind of behaviour?
Yes - but there are many reasons that the UI can stop responding. Being unresponsive is NOT necessarily an infinite loop - indeed most likely NOT one of these.
You need to look at IPMI (if you have it) or connect a monitor to the NAS to see what error messages are coming out on the console.
Probably, but I’m still investigating this point.
I disagree on this: the issue isn’t related to IPMI but TrueNAS itself: IPMI has nothing to do in managing the pools or the drives the pool is made of.
And the behaviour I reported happens only when a pool has an issue, aka degraded or a disk removed by TrueNAS.
I made a short clip on what’s going on; post here the link since I cannot upload the clip.
Yes - but that screen recording only shows that the UI is hung and gives absolutely no idea of why.
The point of IPMI is that it should show the hardware console error messages issued by Linux / TrueNAS that may not be logged to any log file before it crashes / reboots. I am not suggesting that this issue is caused by IPMI.
I see, good point.
The IPMI TrueNAS console shows this message
I’m googling the codes to have an idea what they means.
Moreover I checked the system load with htop, and the results are … interesting …
How it can be possible, with a CPU running at 2%, have a
Load average: 60.03 59.77 58.51?
Those numbers doesn’t sounds realistics to me.
