Wrong notification disk usage twice a day (for years)

[Keichi]

Hello,

 

I'm currently building a second UNRAID (backup) server and I'd like to solve a "bug" I have with the current one.

 

If the disk usage exceeds 70%, I receive an email, then, the next minute, a second email saying that in fact no, the limit has not been exceeded.

 

I know this thread exists:

But obviously, the last message was from me, more than a year ago, and the situation hasn't improved with UNRAID v6.12

 

In Setting / Disk Settings, i have disable notification :

 

For the disk concerned, also :

 

I've also checked the /boot/config/plugins/dynamix/dynamix.cfg file:

However, I still receive two e-mails, twice a day, informing me that the limit has been exceeded :

I've also tried setting 90% and 95%, but still get an e-mail as soon as it exceeds 70%.

 

Would it be possible for someone to help me investigate the problem?

Here the diagnostic :

yggdrasil-diagnostics-20230715-1515.zip

 

Thank you very much,

K.

 

[itimpi]

You have over-ridden the global settings for some of your SSDs.    Have you tried setting those to also not generate notifications?

 

it is quite possible  you have uncovered a bug as I suspect most people do not disable these notifications but merely set them to values that should not trigger notifications.

[Keichi]

Indeed disk1 (HDD) was set to 90/95.

I deleted the values, and now, all disks (8 HDDs array + 4 SSDs pool) are supposed to follow the general rule.

Here the new diag :

yggdrasil-diagnostics-20230715-1728.zip

 

But it is still not 70% (i did a lot of tries, the notification kick at 70% no matter what config i choose).

disk6 generating the notification is at 88% right now, so still under the 90/95 of disk1, and still do not follow the "do not send notification".

 

My goal is to keep notification 90/95%, not disabled it.

I just ended up trying to disable it, in order to not receive the false notification.

 

I will have the evening notification in few hours, if i did not receive it, i will try putting back my 90/95 settings.

 

But the main issue here is : notification kick at 70% no matter what threshold i define.

 

Thanks

 

 

--

EDIT

The issue continue, still receive notifications :

Edited July 15, 2023 by Keichi
Adding Edit

[itimpi]

You have also appear to have over-ridden the defaults for:

CT1000MX500SSD1_2114E594BDCD

CT480BX500SSD1_1832E15212E7

CT2000MX500SSD1_2308E6B10F78

although this may well not be relevant.

 

it looks like it might be treating the 0 for the warning level as assume the default of 70?

 

 

 

[Keichi]

That is pretty strange, in the GUI, the value is greyed, meaning i did not change it :

 

I changed yesterday from 90/95% to 0/0 (globally, did not touch per disk setting).

 

Nothing changed : I still received two notifications ("Disk 6 is high on usage (88%)" and after a minute " Disk 6 returned to normal utilization level") twice a day (at 6 in the morning and 10 in the evening, french time).

 

I can easily activate these notifications by putting a disk above 70%, even if the global setting is 90/95% or since yesterday 0/0.

 

 

So it seems to me, that UNRAID completely ignore my settings and stick to 70% (default setting ?).

 

If you tell me that in the diag, SSDs have custom limit, and from the GUI, i did not see that, maybe there is something to investigate.

 

K.

[Keichi]

Can i put a little up, as the issue persists ?

 

Thanks

 

K.

[Keichi]

No answer ? Even to say "no sorry, we don't know" from UNRAID's team ?

 

K.

[itimpi]

1 hour ago, Keichi said:

No answer ? Even to say "no sorry, we don't know" from UNRAID's team ?

 

K.

You should not assume that anyone from the Unraid team sees every post on this forum as it is primarily used for community support.    To get a formal answer I would expect you need to email Unraid support.

[Keichi]

Thanks itimpi, i will do that.

 

This forum is great because I always found solutions here, this is the first time i did not

 

K.

[Womabre]

I've also been having this issue for year now. Running 6.12.5 (upgrading to 6.12.6 after writing this, the update triggered me to search for a solution again).

Attached are my diagnostics.

breedveld-diagnostics-20231202-1236.zip