[SOLVED] PVE notifications - how to?

[Razva]

Hello,

Right now I've made these settings:

• backup notifications (via GUI)
• root notifications (via /etc/aliases)
• zfs-zed is sending messages to root by default, so I guess that I don't need to do any other settings for this

I've also installed Zabbix.

Are there any other notifications I should set?

Thank you

 

[Dunuin]

You could setup smartctl to run SMART tests once a week/month and mail you if a smart test fails.

 

[Razva]

You could setup smartctl to run SMART tests once a week/month and mail you if a smart test fails.

Is SMART enabled for SSD/NVMe drives? I forgot to mention this aspect.

 

[Dunuin]

Is SMART enabled for SSD/NVMe drives?

Jup. Especially there you should check the SSD wear regularily as ZFS can kill consumer SSDs in months.

 

[Razva]

Jup. Especially there you should check the SSD wear regularily as ZFS can kill consumer SSDs in months.

Yes, you are correct, thank you! Any idea where do SMART notifications go? It's enabled.

 

[Dunuin]

Maybe this is helpful. Read that not long ago: https://forum.proxmox.com/threads/hardware-raid-notification-status.104585/post-450446

 

[Razva]

Maybe this is helpful. Read that not long ago: https://forum.proxmox.com/threads/hardware-raid-notification-status.104585/post-450446

Got it, thank you. It seems that's already there "by default".

Do you have any idea how to send weekly/monthly reports? I would feel way more comfy knowing how my hw is behaving and preparing upfront.

 

[Dunuin]

For advanced features like monthly reports and analyzing alot of metrics to see problems early you might want to install a monitoring software like Zabbix, checkMK or something similar.

 

[Razva]

For advanced features like monthly reports and analyzing alot of metrics to see problems early you might want to install a monitoring software like Zabbix, checkMK or something similar.

I already installed Zabbix, so at this point all my nodes have:

• backup notifications
• root notifications
• SMART notifications
• Zabbix notifications

Is there anything else I could do to improve this chapter?

 

[Dunuin]

I think alerts for failing fans are also useful. And alerts before running out of RAM or disk space.

 

[Razva]

I think alerts for failing fans are also useful. And alerts before running out of RAM or disk space.

RAM and disk space is covered by Zabbix. Any hint about the fans?

 

[Dunuin]

RAM and disk space is covered by Zabbix. Any hint about the fans?

Here zabbix is monitoring them too. Found templates for all my mainboards BMCs which are controlling the fans. So zabbix can connect using IPMI and read fans, voltages and temperatures. Maybe you find some templates too that can read the mainboards sensors.