OneView: Mass clear warnings

[vanz]

Hi Everyone,

I am strugling with a lot of old warnings on numerous servers in my appliance (version 2.00.02). I am wondering if you know a way to mass celar warnings in OneView. Maybe using Powershell or something else. I have more than 1000 activities and clearing them manually will be nightmare.

Thanks in advance!

[ChrisLynch · Accepted solution]

Your only option really is to use the REST API.  Yes, you can use the PowerShell library to retreive alerts and then to clear them:

Get-HPOVAlert -State Active | Set-HPOVAlert -Cleared

[Syed_UAE]

Hello Chris,

Could you eloborate more on this, I have a similar problem, lots of hardware errors in one view, I want to manually clear all the events. The command you have given, will it clears all the alerts or is there soem other way to clear the events.

[ChrisLynch]

The command I provided above will retrieve all Active alerts.  Then they will be passed to the command that will update the state of the alert.  If the Alert is not Locked, then it will be cleared.  Otherwise, an error will be generated that it cannot be cleared.  For Locked alerts, you need to fix the physical issue of the resource in question first, which OneView should then automatically clear the Locked alert.

[CristianRomero]

there is the way with powershell i/o othe rmethod to release these locked events that appear in oneview 4.0 (synergy)

I managed to remove the alerts with the command Get-HPOVAlert -State Active

[Steve_Tippett]

I know one of my primary frustrations with OneView 4.1 in Monitored mode (about 200 c7000 enclosures and several hundred rack mount servers) is the blizzard of alerts that are not very useful.  It's a lot of noise.  Examples are:  "Remote Insight/Integrated Lights-Out self-test error"  "Integrated Lights-Out detected more than unauthorized login attempts"  "Unable to establish trusted communication with the server".

So I wrote a Powershell script to retrieve all the Active alerts, parse through them, and clear the ones with the text above.   I run the script to eliminate the bogus, non-useful alerts, and then go logon to the OneView GUI interface so I can focus on the realistic alerts.

[Henry_B]

I found the "Remote Insight/Integrated Lights-Out self-test error" usually was an indicator for a bad NAND in my cases. Not always, but enough times to check that first.

[Casty]

To actually REMOVE the alerts, which is what I wanted to do, vs. just clearing them, run this Powershell command:

#Reminder- run the following commands using HPE OneView 4.00 Powershell Library

Connect-HPOVMgmt 

#When prompted, enter the FQDN of your OneView Appliance and the Administrator credentials

#Run the following to REMOVE ALL Cleared alerts

# Note- you can only remove Cleared alerts.  If you like to remove Active alerts, first Clear them using the Clear command below then remove the,.  

Get-HPOVAlert -State Cleared | Remove-HPOVAlert

#Run the following if you simply just want to CLEAR All Alerts

 Get-HPOVAlert -State Active | Set-HPOVAlert -Cleared

[Steve_Tippett]

Henry, I agree with your thoughts.  I apparently have several servers with the NAND error.  The lengthy fix documentation (advisory c04996097 & advisory a00048622en_us) lists several steps, but with the final disclaimer "A server AC power removal may be required prior to the NAND format".   I've tried everything short of the AC power removal (so difficult to arrange permission for that in a large Prod data center environment), and nothing has fixed the issue.

[R-G]

For removing cleared alerts you might want to add -Confirm:$false to the end of your command so it doesn't prompt for confirmation every time.
e.g. Get-HPOVAlert -State Cleared -Count 1000 | Remove-HPOVAlert -Confirm:$false

[Casty]

It's also helpful to add a -count value so as not to return thousands upon thousands of alerts which will take powershell a long, long time.  For instance, run the following to retun 100 alerts and purge them:

Get-HPOVAlert -State Cleared -Count 100 | Remove-HPOVAlert

Or you can specify a date range:

Get-HPOVAlert –AlertState Cleared -Start '2018-06-01' -End '2018-08-31' | Remove-HPOVAlert

[cla1801]

I've succeded to clear the error online by doing the NAND flash format, it only reboots the ILO. 
But it does clear the Intelligent Provisioning which you need to reinstall at a time when you can reboot the server. 

for for clearing the error i was able to to it online. IP is only used when rebooting anyway, so it can wait.

[cla1801]

How can you clear old locked errors?? I've tried the rest api solution and i do get the sesssion and everything, but the response comes up with a.. 

Used this to no avail.
https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-a00048315en_us

{"errorCode":"AlertAuthorizationException","message":"Only trusted Resource Managers can make this request.","details":"","recommendedActions":["Provide a valid trusted component token."],"errorSource":null,"nestedErrors":[],"data":{}}

I am using the "administrator" user for this. And the errors are old "Leaf certificate with alias name serial number is expired". These systems have recreated valid certificates but i can't remove these old errors from OneView :(

[Mainecoon]

I have the same request and from what i have found so far is that we do not have permission to remove these so a ticket is required.

[ChrisLynch]

@Mainecoon, correct.  The error message is not a helpful message, and is quite misleading.  Currently, Locked alerts cannot be deleted without opening a support case.  I have reached out to some folks internally to find out how we are going to handle this type of locked alert better.

Can you please tell us what version of HPE OneView you are using?

[Mainecoon]

Hi Chris,

Thank you for looking into this, i'm currently one release behind running 4.10.03-0364293 and have about 30+ alerts stuck from March till July last year that i cannot clear although the certs are updated already. Would appriciate if there is an easy way to get them deleted so i'm one step closer to an "all green" status. Also one comment about the alert filtering that i'm working on, when you set a line that is not working due to typo or wrong statement/wording, the complete filter is kind of blocked in sence that you don't receive anything anymore. Had this a few times now while working on this, would be nice there was some verification that would stop you from saving the rules if it is wrong.

[cla1801]

The same here Chris, though my vesion is the very last one -

4.10.04-0370820

Old alerts, and this is a monitored solution with no subscription. But we are to get some Synergy frames and i would like for this issue to be fixed without opening a case.

[cla1801]

Well an update to OneView  4.20.01-0380241 did not solve this for me :( Again quite annoying not to be able to delete those errors. My OCD is getting hammered here ;)

Any news on this Chris?

[fatfreddyscat]

Hi Chris,

currently on 5.20.00-0419867

I need suggestions on how to unlock locked alerts as below with powershell, or get OneView to ignore certificate errors! Or do I STILL need to raise a support ticket to unlock them?

thanks for your time.

Steve

PS C:\>
PS C:\> Get-HPOVAlert -State locked | grep Leaf
Warning  Security                     Tue May  5 16:25:02 2020 Tue May  5 16:25:02 2020 Locked                Leaf certificate with alias name 35383534-3736-5A43-4338-...............
PS C:\>
PS C:\>
PS C:\>