Don't delete gem files in our build process

ManageIQ / manageiq-rpm_build

Code to build RPMs for ManageIQ appliances and container images

Apache License 2.0

10 stars 25 forks source link

Don't delete gem files in our build process #207

Open kbrock opened 3 years ago

kbrock commented 3 years ago

Background

Our process (loosely)

fetches files (git / bundler)
Deletes some files that are not necessary <=== focusing here
tars those files together
builds an rpm

Request

Drop the delete process and have the tar process exclude those files.

Desired result

This will consolidate the logic for the excluded files.

This also gets us closer to being able to reuse docker containers that build rpms, at least for local development.

Currently we have to throw away a docker container even when we make minor changes to our source code costing us over half an hour per change. Or more accurately, making us hack environments instead of testing them in environments similar to our customers.

If we can speed up the building of rpms, then a final test of a patch could be done with pods that are build with our real production process.

Compiling delay

Fryguy commented 3 years ago

I was thinking about this change as well. I noticed we do both tar exclusion and we have that method for cleansing the gemset. Doesn't make sense when, I think, the tar exclusion can handle both.

Currently we have to throw away a docker container even when we make minor changes to our source code costing us over half an hour per change. Or more accurately, making us hack environments instead of testing them in environments similar to our customers.

I don't understand how this is related. The gem install / tar build doesn't happen in a docker layer, so there's nothing to cache.

If you want to have a docker container that caches the built rpms layer, you can literally do that now by just introducing a new layer in the rpm build Dockerfile that does the actual RPM build.

Fryguy commented 3 years ago

@kbrock Unless I'm mistaken, this issue should be in the manageiq-rpm_build repo, right?

kbrock commented 3 years ago

@kbrock Unless I'm mistaken, this issue should be in the manageiq-rpm_build repo, right?

sigh. correct

miq-bot commented 1 year ago

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 triage process documentation.

miq-bot commented 1 year ago

This issue has been automatically closed because it has not been updated for at least 3 months.

Feel free to reopen this issue if this issue is still valid.

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

miq-bot commented 1 year ago

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 triage process documentation.

miq-bot commented 9 months ago

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.

miq-bot commented 6 months ago

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.

miq-bot commented 3 months ago

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.