Question

Clear critical alerts with a later date

  • 17 January 2020
  • 7 replies
  • 1882 views

Badge +1

Hi,

I had a problem with the NTP server configured on the Nutanix cluster, unintentionally changed the date to 09012020 instead of 01092020,  the point is that after solving that problem, now the cluster reports to me Critical alerts dated September 1, 2020. The date of the NTP is already corrected, but Web Console  shows some critical alerts, but those alerts are not shown with ncli, but  are shown with Alert tool. I have try to aknowledge and  resolved them, but they are NOT eliminated from alert_list.

Does anyone know what I can do so that the web console no longer shows them?


This topic has been closed for comments

7 replies

Badge +1

I forgot to mention that I followed the nutanix procedure that instructs to run this command:

nutanix@CVM:$ ncli alerts resolve ids=1598992889808070:11:"00059574-18e4-4672-5df8-ac1f6bc18571"

Error: Could not find all the given alerts.

 

as I mentioned above, ncli cannot see these alerts. : (

Userlevel 3
Badge +5

Hi

Follow the below KBs, maybe you can get more:-

https://portal.nutanix.com/#/page/kbs/details?targetId=kA0600000008cOXCAY

https://portal.nutanix.com/#/page/kbs/details?targetId=kA0600000008iHbCAI

Badge +1

Hi Ramakrisnha

I appreciate your quick answer, but I have already been tested  both procedures and none worked, the critical alerts continue showing on prism, : (

Thanks

 

Userlevel 3
Badge +5

Hi

Ask Customer Technical Support :- https://portal.nutanix.com/

Then Create a New Case.

 

Userlevel 3
Badge +4
Description

Alerts in Prism might not be cleared after the alerts have been acknowledged and resolved.
This can also cause false alert emails to be sent out. To resolve this issue, follow the steps in the Solution section to rebuilt the Alert index.

Solution
Rebuilding the alert index database fixes this problem.
  1. Open two ssh sessions to the same CVM.
  2. In one of the ssh sessions type the following command. 
    nutanix@cvm$ tail -F ~/data/logs/alert_manager.INFO  | grep Done
  3. On the other CVM type the following command. 
    nutanix@cvm$ allssh "genesis stop alert_manager"
  4. From the same CVM type the following command. 
    nutanix@cvm$ ~/bin/alert_manager --alert_index_force_rebuild=true

    In the CVM where you have the tail command running, the logging Done applying xx alert index updates messages mean that the alert index is now rebuilt. You can stop the tail command initiated in Step 2 (ctrl-c).

  5. To start the alert manager back again type the following command.  
    nutanix@cvm$ cluster start

 

Note: I suggest to open a case and comment this procedure to the assigned SRE.

Userlevel 4
Badge +2

Hello @Nora,

See if KB-2453 helps you. The description of your issue and this article matches.

Userlevel 3
Badge +4

Normal methods to clear alerts would not work here since Prism, and NCLI which is just CLI interface to the same system, cannot list these alerts.

I see a case was created for this. I think that is the best course of action here. The solution requires some manual editing that SREs normally would not do unassisted.

Possibly a software fix can be produced from this so similar issues are not encountered in the future.