Lately I came accross with an issue updating disk firmware on NetApp filer. With the new releases of Data ONTAP updating disk FW is piece of cake, right?

1. Check the current Disk Qualification Package right here – notice what is the date
2. Check the installed Disk Qualification Package on your NetApp

   rdfile /etc/qual_devices_v3

   you are looking for a line

   # Datecode: YYYYMMDD - one is at the beginning, one is at the end

3. If you have an out-dated version follow the given instruction to install the current one
4. Check the current Disk Firmware installed, and types of disk. For that I use the trick, sending request from Linux host:

   ssh -l root netapp01 storage show disk -x | awk -F" " '{print $6,$7}' | sort | uniq -c

   Exmple output:

      1 ----
      1 MODEL REV
      8 X287_HVIPC288A15 NA01
      4 X287_HVIPC288A15 NA02

Now all you have to do is to go http://mysupport.netapp.com/NOW/download/tools/diskfw/ compare your version with the current one. Check all the per-requsits, download the new one and go with the given instruction of installation.

Disk Firmware update issue

Once you have copied the newest version to /etc/disk_fw/ drive,  and options raid.background_disk_fw_update.enable is set to on automatic, background and non-disruptive update will start.

In /etc/messages you should see message like:

Wed Dec 28 15:06:59 EST [raid.disk.offline:notice]: Marking Disk /my_aggr/plex0/rg0/0c.00.3 Shelf 0 Bay 3 [NETAPP   X287_HVIPC288A15 NA01] S/N [XXX] offline.
Wed Dec 28 15:06:59 EST [bdfu.selected:info]: Disk 0c.00.3 [NETAPP   X287_HVIPC288A15 NA01] S/N [XXX] selected for background disk firmware update.
Wed Dec 28 15:06:59 EST [dfu.firmwareDownloading:info]: Now downloading firmware file /etc/disk_fw/X287_HVIPC288A15.NA02.LOD on 1 disk(s) of plex [Pool0]...
Wed Dec 28 15:07:18 EST [raid.disk.online:notice]: Onlining Disk /my_aggr/plex0/rg0/0c.00.3 Shelf 0 Bay 3 [NETAPP   X287_HVIPC288A15 NA02] S/N [XXX].

Which basically states the correct installation.  But what if you have a case like:

Wed Dec 28 15:06:59 EST [raid.disk.offline:notice]: Marking Disk /my_aggr/plex0/rg0/0c.00.3 Shelf 0 Bay 3 [NETAPP   X287_HVIPC288A15 NA01] S/N [XXX] offline.
Wed Dec 28 15:06:59 EST [bdfu.selected:info]: Disk 0c.00.3 [NETAPP   X287_HVIPC288A15 NA01] S/N [XXX] selected for background disk firmware update.
Wed Dec 28 15:06:59 EST [raid.disk.online:notice]: Onlining Disk /my_aggr/plex0/rg0/0c.00.3 Shelf 0 Bay 3 [NETAPP   X287_HVIPC288A15 NA01] S/N [XXX].

Pay attention. All operation took ~1 second, and at the end the disk is still within the same firmware. In my case the solution was:

1. I deleted from /etc/disk_fw/ older version of the same disk type (for example X287_HVIPC288A15.NA01.LOD), leaving only the newest version (for example  X287_HVIPC288A15.NA02.LOD)
2. I disabled option raid.background_disk_fw_update.enable
3. Waited couple of minutes
4. I re-enabled option raid.background_disk_fw_update.enable

And as simple as that – it helped. Other trick would be to manually force the update with disk_fw_update command, but check out this manual entry: https://library.netapp.com/ecmdocs/ECMP1511538/html/man1/na_disk_fw_update.1.html. As you can see this option might have a consequence for your customer using the array.

If you came across with such situation I strongly recommend opening a case to NetApp before you try any changes.