Opened 8 years ago
Last modified 5 years ago
#695 new enhancement
Incorrect serial number in email report after drive replacement
| Reported by: | alan_nv | Owned by: | |
|---|---|---|---|
| Priority: | minor | Milestone: | unscheduled |
| Component: | smartd | Version: | 6.3 |
| Keywords: | Cc: | Nathan Stratton Treadway |
Description
Using the 6.3+svn4002-2+b2 version included in Debian Jessie.
After replacing a hard drive, smartd detected a smart error on the new drive, and sent an email with details. The serial number of the drive in the email was incorrect - it was the serial number of the previous drive in that slot.
Restarting the smartmontools service resulted in an email report with the correct serial number - presumably the old one was cached somewhere rather than queried.
The disks in question are in a Dell MD1000 enclosure attached to a Dell H800 raid controller - rebadged LSI SAS 2108. I haven't specified any +dsat,megaraid devices in /etc/smartd.conf, the devices seem to be noticed correctly by DEVICESCAN now, although perhaps that has some impact on the error I've encountered.
Change History (2)
comment:1 by , 8 years ago
| Component: | all → smartd |
|---|---|
| Milestone: | → unscheduled |
| Type: | defect → enhancement |
comment:2 by , 5 years ago
| Cc: | added |
|---|
This is as expected. Smartd does not rescan device list on hotplug or re-issue IDENTIFY/INQUIRY commands before each device check. After disk replacement, it is required to manually restart smartd or request a rescan (SIGHUP).
Related: tickets #60 and #218.
BTW: Yes, devices behind LSI SAS controllers are included in DEVICESCAN since r3734.