Clear Critical HDD temp alert?

[EvanVanVan]

How do I clear HDD temp alerts? I've searched and only come up with "zpool clear volume1" but that didn't work (and I think it's because that's used for clearing a different set of errors?).

Last night I got emails about critical alerts on 3 of my drives, they reach 40-41 degrees. I cleared my fan filters off and the temps have all dropped back down to the mid to low 30s. I know another short smart test has run (I have it set to the default of every 30 minutes). The critical alerts haven't reoccurred in the log but the alert didn't clear either.

How do I clear the Alerts?

Thanks

 

[dlavigne]

Does unchecking the box for that alert clear it?

 

[EvanVanVan]

Unchecking those boxes makes the alert glow green, but when I open up the Alert window the Critical messages are still there. Will they clear themselves at some point?

Edit: I'm assuming unchecking them is just a good way of ignoring them (although in my case the problem has been rectified) and isn't as good as actually clearing them?

 

[DifferentStrokes]

Once your log reaches 1000 lines it will prune it out.

 

[EvanVanVan]

When you say prune it out, are you talking just about the log or both the log and the Alert System window?

Since 6 am this morning I've only had 4 lines added to the log after the temperature alerts, so I'm thinking reaching 1000 lines will take a while (I've got log entries from June 10th still for instance).

If by chance the temperature warning comes back in and I've unchecked the alerts in the Alert System window so that it glows green, will I get new emails about it? Or does unchecking it cause it to be ignored in the future?

 

[DifferentStrokes]

I don't know exactly what is going on behind the scenes. From my experience, however, I've noticed that the log gets check each night and clears itself after it reaches 1000 lines. So your right it will take you a while at the rate you are going. Rebooting clears the log as well. This may not be any option though.

A new entry and email is triggered each time it see the temp out of range. What do you have the high temp set at?

 

[EvanVanVan]

40C, and I've been checking it a bunch today and it's been fine so I'm not really that worried about it. I just have filter screens in front of my fans to try and keep as much dust and cat hair out of case, and it had been a while since I had wiped them clean...

 

[DifferentStrokes]

I guess it all boils down to rebooting or just living with it.

I just you could edit the log manually at the CLI, but I'm not should if that is something you'd want to do.

Maybe someone else that knows more than me can chime in with an easy fix.

 

[saurav]

Is this "Critical alert about HDD temperature" new in 9.3? My HDDs also cross 40C in summer (I live in a tropical country), but I don't get those alerts. Perhaps because I'm still on 9.2.1.7?

 

[DifferentStrokes]

I'm not sure if this was in 9.2 or not. However, it isn't on by default. You have to set it up in the GUI.

 

[EvanVanVan]

Is this "Critical alert about HDD temperature" new in 9.3? My HDDs also cross 40C in summer (I live in a tropical country), but I don't get those alerts. Perhaps because I'm still on 9.2.1.7?

Idk when it was added, I don't think it was a new addition though. It's in Services > S.M.A.R.T. > Critical.

Without further information or suggestions I'll assume that dlavigne's comment on unchecking them is the best solution, and DifferentStrokes is correct that they will auto prune, eventually.

 

[hugovsky]

I'll assume that dlavigne's comment on unchecking them is the best solution

No. The best solution is to set it properly.

 

[EvanVanVan]

No. The best solution is to set it properly.

Thank you for the dismissive unhelpful reply, but what are you talking about? I have it set properly... @ 40C. What does your link to the docs have to do with Clearing the Alert System window stuff?

 

[hugovsky]

The alert is there to protect the drives. If you are ok with higher temps, just set it higher. To clear it, just wait or reboot.

 

[EvanVanVan]

Ok, all conclusions I've already come to in this thread. As I've said, my temps are back in the mid 30s, which is why dlavigne's comment on unchecking [the alert window messages] is the best solution. And like you quoted me, I'll wait for the logs to autoprune. Thank you for your input.

 

[silversword]

Note: Testing this in 11.1-U6, looks like it'll only clear one alert at a time.

Submitting as bug.

 

[learnsomemore]

Also note, reboot does not change anything. It will still be in the alert window after reboot. Tested 11.1-U6

 

[Chris Moore]

Also note, reboot does not change anything. It will still be in the alert window after reboot. Tested 11.1-U6

The question is how to reset the, "Alert System" so that it stops advising of an event that has been cleared. Nobody ever answered this except to say reboot the system. There should be a better way, that doesn't require a reboot.

 

[learnsomemore]

The question is how to reset the, "Alert System" so that it stops advising of an event that has been cleared. Nobody ever answered this except to say reboot the system. There should be a better way, that doesn't require a reboot.

It's interesting this should come up again today. I had to replace a bad drive in a FreeNAS box today. Same issue, no easy way to clear GUI error of degraded state after pool finished rebuilding. No errors in terminal but GUI still showing old errors. According to https://redmine.ixsystems.com/issues/11117, this will never be changed.

I manage both FreeNAS (45 Drives hardware) and Synology boxes and the list of reasons to steer people away from FreeNAS grow fast enough without the help of nuisance issues like this.

 

[Chris Moore]

I manage both FreeNAS (45 Drives hardware) and Synology boxes and the list of reasons to steer people away from FreeNAS grow fast enough without the help of nuisance issues like this.

I have several QNAP systems where I have replaced the original Linux based software with FreeNAS as well as I have one of the LX60Turbo units from 45Drives and some other Supermicro systems at work. I use FreeNAS everywhere I can because of the robust nature of ZFS and easy management but it does have some room for improvement. All my production systems (home and work) are on 11.1-U7 which is where this defect is still hanging out ever since, forever I guess, because that ticket you pointed out is not the only one. I have a test system configured with the the new 11.2 software, and there are bugs, but I have not had any hardware faults on it yet so I can see how it handles them. I guess that I need to induce a fault.

I had to replace a drive in one of my production systems today at work. The thing has 122 data drives among two storage pools. It is used all day, every day, and it is just inconvenient to reboot the system to clear an alert for something that was already corrected by hot-swapping a drive.
I don't expect it to clear up on it's own, but if it is still flashing at me Monday morning, it may jut get rebooted. I don't like that flashing red ball.