I'm running ProxMox 6.0 VE on 4 HP Proliants with SP Smart Array controllers. I would like to figure out how to edit the existing SMART gui to monitor my drives. I have the cli command, smartctl -a -d cciss,0 /dev/sd[x], which gives me a significant amount of data but I would prefer to have it through the gui.
SMART monitoring SAS drives on HP Proliant DL360/380
- Thread starter coreyjohnson75
- Start date
Most other users use external monitoring for their servers (e.g Nagios/Icinga, CheckMK, Zabbix, etc...) which includes what you want.
Oh and we also use the hpssacli tool directly to also detect failures. In fact, we only rely on them and not directly on smart. The tools (also for other hardware) report smart errors as well.
A more simple soiution would be to setup the smartd daemon to run short and long tests and have it email to you any inconsistencies that it may find. Having an output with counters will need more work, but having the daemon check those counters and inform you when a {short,long} check fail is more helpful.
@LnxBil are you just checking the raid status for disk failures or are you actually getting the reports for the disks and parsing them?
@LnxBil are you just checking the raid status for disk failures or are you actually getting the reports for the disks and parsing them?
Actually both, but the reports are manual. According to the mails we got, there are also different type of errors occurring, e.g. media, predicted failure, other, etc. If there is an error detected (after patrol scan) that is not just a simple "disk dropped" message, I will get the controller event log output and parse it. It also works for MegaCli stuff. The log can be very huge and sometime cluttered with a lot of the same messages. I never bothered to "intelligently analyse" the files and build reporting around that.