Fix "query in loop" in InfraConversionThrottler

[ghost]

When implementing #18860, @Fryguy noticed that https://github.com/ManageIQ/manageiq/blob/master/lib/infra_conversion_throttler.rb#L4 feels like a terrible N+1 (query in a loop is bad).

This issue track the effort to get rid of this bad thing.

[Fryguy]

That entire method is N+1s nested in other N+1s. You have an outer loop which has multiple inner loops and those inner loops have more loops inside there, with individualized queries inside. I recommend doing a complete performance check on it.

cc @kbrock

[ghost]

@miq-bot assign @djberg96

[kbrock]

I had tried with #18876

But some caching got injected in there after a few runs.

It sounded like the consensus for the next version would:

• remove check_concurrent_tasks from eligibility. since this changes over time
• load the counts locally, and divvy out the tasks ignoring the counts in the database. Since the counts will only change by 1 and this method is the one that will change it

[miq-bot]

This issue has been automatically marked as stale because it has not been updated for at least 3 months.

If you can still reproduce this issue on the current release or on master, please reply with all of the information you have about it in order to keep the issue open.

Thank you for all your contributions! More information about the ManageIQ triage process can be found in the traige process documentation.

[Fryguy]

@fdupont-redhat Is this issue still relevant?

[ghost]

@Fryguy If we remove the V2V feature, it's not.

[Fryguy]

Removed in #21515

[miq-bot]

@ghost 'djberg96' is an invalid assignee, ignoring...