"new serial number received" but I have not changed the disk

[runen]

I have experienced 4 identical problem events stating that my disks have got new serial numbers. But I have not changed any hardware or rebooted the server when Zabbix alerted the problem. Should I ignore this issue or what should I do?



I have a Dell PowerEdge r630 server with 4 Intel SSD disks in raid controlled by a raid controller (PERC 730P Mini).
I'm running Zabbix 5.2.6 in a Debian 10 VM on my XCP-NG host.

Any help is highly appreciated

[johndoe2374]

Hello.

I guess you're using template "Server Dell iDRAC SNMP". The key is system.hw.serialnumber, but if you check MIB file for OID, it says that it's systemServiceTag, which doesn't sound like hardware serial number for me. Also from item's description:

MIB: IDRAC-MIB-SMIv2
This attribute defines the service tag of the system.

So, there might be a template creator's mistake in item name or something. I think you can ask Dell for better explaination of that OID. Also the RAID controller may be somehow involved, so it changes the values. Check for actual SSD's serial number and compare it with those being collected from history or using terminal directly like so (change SNMP version and community string to your situation):

snmpwalk -v 2c -c public <yourserver> 1.3.6.1.4.1.674.10892.5.1.3.2.0

Key physicalDiskSerialNo with OID 1.3.6.1.4.1.674.10892.5.5.1.20.130.4.1.7 from the same MIB sounds more like true. You can also try to contact template's creator for some explainations.

[elyograg]

I am running into this problem also. I get the message for two disks at the same time. I am not sure what the frequency is, but I get these two problems on a regular basis:

Physical Disk 0:1:12: Disk has been replaced (new serial number received)
Physical Disk 0:1:13: Disk has been replaced (new serial number received)

The hardware is R720xd, it has two 2.5 inch disk bays in the rear, which I have set up as a RAID1 for the OS. I have a RAID10 volume on the 12 front drive bays.

Checking the history of both keys, I see that they will swap serial numbers. These are a disk serial numbers, not the tag id that johndoe2374 mentioned.

I think this means that idrac is misbehaving. I doubt there will be a new idrac version for this motherboard, as it is very old hardware.

For other things there is a macro that disables the test. For example, on one of my hosts I defined the {$IFCONTROL:"virbr0"} macro and set it to 0, so now I don't get notified when that network interface goes down -- which it will do when the QEMU virtual machine I have defined on that host is shut down. I think that capability should be added to the idrac template.

[digdilem2]

Old post. but google brought me here when I encountered the same error after a reboot.

I got four copies of this from the Smart template for a 4-disk mdadm raid 10, reporting sdc/sdd etc

I am certain this is because /dev/sdc mounts randomly by the OS. It makes no difference to the RAID and it does not indicate a problem.

I've simply turned those triggers off, since they will fire each reboot.