Remove build retry attempt

[squakez]

The build should just fail and report this information back to the user ASAP, there is no reason to retry a build that is going to likely fail again for the same reason. I guess when it was introduced it was because there was some maven build intermittent issue.

Originally posted by @squakez in #6582 (review)

[Evynglais]

Hi,

"there is no reason to retry a build that is going to likely fail again for the same reason"

This is not always true; the example that prompted me to work on the PR that raised this was for builds that were failing to due missing (temporarily) knative service endpoints. If the service isn't available during the build, then the operator will fail my integration; if it retried, then it would be successful once the service has started. This change would fix that issue by allowing the operator to retry until the external service is available -- not to say that there wouldn't be another (and perhaps, better) way of fixing this specific issue. Happy to take suggestions/pointers if you have any ideas. There are likely other valid reasons for temporary build failures that would be resolved with a retry.

I do think that for retrying it would be good to be able to distinguish between different types of failure (non-resolvable build issues, vs temporary/external dependency issues), as I agree that there are build failures that can't be rectified, and we shouldn't necessarily waste resources retrying those.

Thanks,
Evyn

[Evynglais]

This seems to be what we need:

#camel-k > Integration Builder Recovery @ 💬

In that case, the deployment just fail and until there is any change in the Integration or the related custom resource, there won't be any reconciliation cycle to verify again that the resource is now available. In theory we should think how we can introduce some reconciliation cycle after a while the Integration is in error state in order to automatically recover a race condition like that one.

[squakez]

Yes. So, beside removing the rebuild attempt, we can include a "retry after" on an errored integration in order to kick off a monitoring action every minute or so.

[Evynglais]

Sounds good - thanks for the discussion. I'll close the PR for the build retries.