Error replacing drive with "The smartd service failed to start"

[eternal_cycle]

I'm attempting to replace a drive with a larger one via the GUI. I am currently in the process of simultaneously replacing 3 other drives in the same zvol. This is the final drive in the zvol, and I cannot begin the replace process on this drive.

All drives previously present in my machine are working, the goal is only to replace the drives with larger capacity drives.

/var/log/messages.log and /var/log/middlewared.log display nothing relevant besides perhaps "Got an orphan smartd process:" in the middleward.log. No new messages are present in /var/log/messages.log when this occurs. Restarting smartd via `service smartd onerestart` appears to work without error.

I have tried with 2 different drives, and got the same issue with each, so I don't believe this is an issue with the drives.

I'm on freenas 11.2-U3. I cannot find any obviously related bug reports in recent patch notes related to this. My next steps would be to upgrade and restart after my current resilvering completes.

 

[Chris Moore]

It is usually a good idea to tell us more about your hardware when you are having trouble. Here is a guide:

Forum Guidelines
https://www.ixsystems.com/community/threads/forum-guidelines.45124/

How many drives are you trying to replace simultaneously? What type of vdev are these drives in? Although it is possible, under the right circumstances, the usual recommendation is one drive replacement at a time.

 

[eternal_cycle]

To close the book here:

I waited for the resilver to finish, it replacing the last disk worked fine.

Started my next upgrade by replacing all the drives at once in the other vdev, also worked fine.

I'm unsure what my error was, or why I was able to replace all the disks at once the second time but not the first, but here we are.

Hopefully this is helpful to someone who finds it later!

 

[Chris Moore]

Hopefully this is helpful to someone who finds it later

Probably not, seeing as how you gave absolutely no details and didn't answer any questions.

 

[eternal_cycle]

I mean, I think that's kinda a rude thing to say.

I totally understand why you, as a person who is trying to help people debug would ask for a whole bunch of information that may or may not be relevant to the problem, because it reduces back-and-forth. I've been there.

However, as a person who doesn't necessarily have all the information requested in the guide you linked handy I balked at going and dredging it all up (I've had this for years, I don't know my mobo model, etc), plus, given I was going to be away for a few days anyway, so I just decided to let it run. This fixed the issue, even if I don't understand the root cause.

As a software engineer, seeing a thread with a rare, specific specific error and then having the person fix it and never come back is the worst damn thing. So, I came back and answered with the novice information that I had. That felt a lot better than radio silence to me.

The fact is, given the fact that I can't seem to find this anywhere else, there's a great chance that it's an issue some aspect of my local environment (parts, configuration, etc), and I felt as though coming back after I resolved an issue that I was unable to reproduce, to play 20 questions with the you or one of the other experts here was a waste of their time, because I couldn't even validate if any suggestions would have fixed my problem, because it was already fixed.

So, this might not be useful to you, and I get that, however, I would have been very relieved to see this thread if it had existed before I posted my issue. I'm sorry if I've wronged you or the community here, somehow, but I assure you I had the best intentions.

 

[Chris Moore]

If you would even tell what the number of drives in the vdev was or how many drives you were replacing at the same time it might actually be helpful to someone. You didn't really share anything.