FMC Health Alert not clearable (Interface Status (Modified))

[Legusol]

We just upgrade to FMC v7.0.1 and afterwards we seeing an issue with a Health Alert that we cannot clear. The following is showing for one of our FTDs. At the interfaces shown are disabled interfaces on the FTD that are not in use which they should be since we don't use them. We cannot figure out how to clear this alert. If you click the alert, it takes us to the Health Monitoring page where both the FMC and FTD are showing Green without and alerts.

Prior to raising a TAC case I figured I check here.

[Marvin Rhoads · Accepted solution]

Here is the fix to delete the messages. The following commands need to be run as root from the FMC cli (expert mode):

OmniQuery.pl -db mdb -e "select status,category,hex(uuid),body from notification where status=11;"
OmniQuery.pl -db mdb -e 'delete from notification where uuid=unhex("<insert the uuid that the first command returns here>");'
OmniQuery.pl -db mdb -e "select status,category,hex(uuid),body from notification where status=11;"

repeat line 2 as necessary if there is more than one entry to be removed. The output after line 2 should be "Command returned no results." That is normal. Line 3 (same as line 1) should return "0 rows in set" if you removed the messages successfully.

The following bugID applies: https://bst.cloudapps.cisco.com/bugsearch/bug/CSCwa14518
(internal only as of this writing since it is still under investigation with only a work around in place and not a proper fix)

[Legusol]

Thank you! That worked perfectly!

[tjmann]

Just worked for me. I had to remember to do a sudo SU first, which gives you root user access. Then run the commands.

Thank you

[mdelvecc]

Hi Marvin,

I am running image 7.0.2-88 on my 4600 FMC, and tried running the commands you listed above. For some reason, they just won't clear. I ran all lines several times, but still display in notification window. It never returns a "0 rows" after running the 3rd line. Any other suggestions?

Thanks,

Mark

[Marvin Rhoads]

@mdelvecc is your FMC in HA by chance? If so be sure to run the commands on the active FMC.

If it's stand alone, then you may have some other root cause and would need to open a TAC case for resolution.

[mdelvecc]

Yes, I am running H/A pair, but ran it on the active. I see the table populated with the warnings, after I run the commands. I ran the same commands on the standby FMC and it returned "0 rows", which I would expect.

[Priit Kosmopoliit]

The solution worked, but we are running FMC 7.1.0.1 (build 28) and in the bug tool it says:

Known Fixed Releases (2 of 2)

7.1.0

7.0.1

Which seems to be not true.

[ste.ant]

Worked with FMCv v6.7.0.3 as well! Thank you.

[Martin Ostberg]

Running a newly installed 7.0.1 here and I'm getting this as well.

Running 

OmniQuery.pl -db mdb -e "select status,category,hex(uuid),body from notification where status=11;" 

 Doesn't return any results though.

admin@DJ-FW01:~$ sudo OmniQuery.pl -db mdb -e "select status,category,hex(uuid),body from notification where status=11;"
Password:
+--------+----------+-----------+------+
| status | category | hex(uuid) | body |
+--------+----------+-----------+------+

0 rows in set (0.000796 seconds)

Edit: my bad, I was in FTD, not  FMC.

This worked!