Azure Application Gateway
An Azure service that provides a platform-managed, scalable, and highly available application delivery controller as a service.
605 questions
Azure Application Gateway Unavailability and generate Alert
Mishra, Arpit
1
Reputation point
Hi Team,
My Use case is, as soon as App Gateway gets down due to any reason an alert should have generate and follow its action.
However when i tried to put a resource health alert in Azure, it seems Azure Doesn't provide the Unavailable Event for it. Hence, when app gateway went down it didn't generate any alert but when it back to available state, alert generated consisting its current health status and can have previous health status as well.
App Gateway Metrics, i didn't find much which exactly fix my use case. and same with kusto queries.
I have also tried with Azure Service Health Alert but that wont help me as it comes at region level.
Can you please help me is there anyway by which i can fix my use case.
{count} votes
-
KapilAnanth-MSFT 10,981 Reputation points Microsoft Employee2023-03-03T10:54:54.2733333+00:00 Welcome to the Microsoft Q&A Platform. Thank you for reaching out & I hope you are doing well.
I understand that you would like to monitor Application gateway and trigger an alert.
What I believe can be a good metric here is to use the "Response status (4xx, 5xx) crosses threshold" .
Refer : App Gateway Alerts
You should observe the gateway in production to determine static threshold or use dynamic threshold for the alert.
Please let me know if this helps or you require further information on this.
Cheers,
Kapil
@Mishra, Arpit
Can you please update us if the action plan provided was helpful?
Should there be any follow-up questions or concerns, please let us know and we shall try to address them.
Thanks,
Kapil
Hi @KapilAnanth-MSFT pls let me try today and will update.
Hi @KapilAnanth-MSFT , i have tried the configuring 5xx status metrics but it didnt throw an alert when i manually stop the app gateway. pasting screenshots.
can you please help me if anything i'm missing here
@Mishra, Arpit
The behavior is expected when you stop the App Gateway.
App Gateway, when it's stopped, won't give any 5xx responses and instead will time out.
Had you stopped the backends, the App gateway would have given 5xx responses to the clients, and this will be logged and the Alert would have worked.
For platform related issues, you can configure "Azure Resource Health".
Again, I believe stopping the App gateway will bot be reflected in the "Resource Health"
Cheers,
Kapil
@Mishra, Arpit
Could you please provide an update on this post?
Kindly let us know if the below helps or you need further assistance on this issue.
Thanks,
Kapil
Sign in to comment
1 answer
Sort by: Most helpful
@Mishra, Arpit
Welcome to the Microsoft Q&A Platform. Thank you for reaching out & I hope you are doing well.
I understand that you would like to monitor Application gateway and trigger an alert.
For platform related errors, you can configure "Azure Resource Health".
For Backend related errors (where client gets a 4XX or 5XX error)
The metric to use here is "Response status (4xx, 5xx) crosses threshold" and create an alert when the client gets a abnormally more number of such responses
Refer : App Gateway Alerts
You should observe the gateway in production to determine static threshold or use dynamic threshold for the alert.
Kindly let us know if the above helps or you need further assistance on this issue.
Thanks,
Kapil
Please don’t forget to close the thread by clicking "Accept the answer" wherever the information provided helps you, as this can be beneficial to other community members.
resource alerts doesnt have Unavailability events .. but let me check the backend status as well. will update it soon
Sign in to comment
Activity