Email notifications for degraded pool?

Troubleshooting
1

Just starting out with Rockstor 5.1.0. Have mainly used Debian Linux systems previously and commercial NAS like QNAP.

I wanted to test out some scenarios for my 2 disk RAID1 setup, including one of the two disks ‘disappearing’. I did this by creating a pool (with the degraded mount option to ensure continued accessibility), copying some data to it, then disconnecting one of the disks to simulate a disk failure.

Before I did this I configured the E-mail alerts and sent a test email successfully. However I found that Rockstor did not send out any email. I did see the flashing red message about a degraded pool in the web interface however.

I want to deploy this software for a family member and I won’t be checking the web interface regularly. I still want them to be able to access their data if one of the drives fails (yes I will be backing it up elsewhere regularly) hence the degraded mount option.

Is there really no option to have an email alert sent out notifying me if this type of failure occurs? This seems like a pretty major oversight for a NAS-like system.

Secondly, I find it strange that there does not seem to be a way to install Rock-Ons on the System drive. Am I missing something? I did not try importing the System Pool as I am just starting out and this is clearly indicated for ‘advanced users’

Thanks!

2

Hi @plutoz and welcome to the forum.

What I remember there is no e-mail notification from Rockstor itself for failing pools, but you can easily configure the smartd to send e-mail notification - the setup can be done through the Rockstor web-UI.

I have explained it here, this is from another thread:

Cheers Simon

2 Likes
3

Thanks for the quick response. I implemented the smartd configuration however it doesn’t seem to help with completely disconnected / failed drives that are no longer being picked up by the system (I tested it)

I’m also seeing a ‘DEV ERRORS DETECTED’ / Pool Device Errors Alert message on my single (non-raid) pool that I assigned to Rock-Ons which seems to be stopping that service from starting, but the disk passes self-test and doesn’t seem to be generating any SMART errors so no e-mail notifying me of this either :melting_face: I guess I’ll try swapping out that disk anyway and see if the errors continue but at this point I’m not sure I can put this system into production where it can’t be continuously monitored.

2 Likes
4

You could also take a look at a proposal that was submitted to the btrfsmaintenance repo:

(but I don’t think it was ever merged) that checks the btrfs array and sends an email (not using smartd):

https://github.com/kdave/btrfsmaintenance/issues/88

5

Thanks, I had a look and while there are definitely some useful tools there I couldn’t find anything related to email notifications?

6

This is the PR under the issue I linked to:

where it is proposed to add an email service in case of issues. Since it’s not been merged you would have to create those files yourself I think.

1 Like