ServicePulse only allows you to delete 50 error messages at a time.
Is there a way we can automatically delete error messages that are older than a month for example.
Currently our CPU is spiking at 100% and we think it’s in response to the large number of errors we have.
We’ve just cleared the errror messages that are older than a month to see if that helps.
Hi,
There are couple of angles that I will try to cover here.
- ‘ServicePulse only allows to delete 50 error message’ - that is partially true, when you select individual messages that is correct, but if you want to delete large amount of messages you probably should delete whole groups: Failed Message Monitoring • ServicePulse • Particular Docs
There are different ways to classify messages (by endpoint, stacktrace, message type etc.). There you can delete all messages matching given filtering
- If you want to delete old messages, in ServiceControl Managment you can specify: Automatic Expiration of Data • ServiceControl • Particular Docs expiration thresholds. All messages older than the set value would be automatically removed.
- CPU spiking at 100% - I assume it is spiking on the box with ServiceControl. That means that ServiceControl is most likely busy doing things. Large amount of data may have an impact on Raven indexes which could consume CPU, but I would also suggest checking the ServiceControl database sizes: RavenDB Embedded Location • ServiceControl • Particular Docs (for both error and audit) as well as queue sizes, it could be that the number of messages it to big for ServiceControl to process.
Thanks so much for this detailed response Michal!
We’ll try out deleting error messages older than a month automatically via Service Control.