[BUG] - Notifications not sending

[eperdeme]

Describe the bug A smart attribute is marked as FAILED but I'm not seeing any notifications. Test notifications work via

curl -X POST http://localhost:8080/api/health/notify

Expected behavior I'd expect a notification

Screenshots

Log Files

ime="2021-12-01T00:00:15Z" level=info msg="Collecting smartctl results for sdg\n" type=metrics time="2021-12-01T00:00:15Z" level=info msg="Executing command: smartctl -x -j /dev/sdg" type=metrics time="2021-12-01T00:01:19Z" level=error msg="smartctl returned an error code (64) while processing sdg\n" type=metrics time="2021-12-01T00:01:19Z" level=error msg="smartctl detected a error log with errors" type=metrics time="2021-12-01T00:01:19Z" level=info msg="Publishing smartctl results for 0x500003975be81771\n" type=metrics

kernel logs suggesting the drives also sad to backup the smart failure.

[52713.869412] sd 3:0:6:0: [sdg] tag#2512 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE cmd_age=3s [52713.869439] sd 3:0:6:0: [sdg] tag#2512 Sense Key : Medium Error [current] [52713.869445] sd 3:0:6:0: [sdg] tag#2512 Add. Sense: Unrecovered read error [52713.869451] sd 3:0:6:0: [sdg] tag#2512 CDB: Read(16) 88 00 00 00 00 00 00 ff 52 08 00 00 00 08 00 00 [52713.869455] blk_update_request: critical medium error, dev sdg, sector 16732680 op 0x0:(READ) flags 0x0 phys_seg 1 prio class 0

[FingerlessGlov3s]

Sounds similar to my bug https://github.com/AnalogJ/scrutiny/issues/203

[joe-eklund]

Does smart actually report the disk as failing? Is it marked as failed on the main dashboard? If not, it will not send a notification I think.

I asked about this here https://github.com/AnalogJ/scrutiny/issues/195, which turned out to be a duplicate of https://github.com/AnalogJ/scrutiny/issues/72.