Bug 1758467
| Summary: | RFE: set package manager options for buildroot | ||
|---|---|---|---|
| Product: | [Community] Copr | Reporter: | Michael Mráka <mmraka> |
| Component: | frontend | Assignee: | Copr Team <copr-team> |
| Status: | CLOSED UPSTREAM | QA Contact: | |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | unspecified | CC: | amoralej, praiskup, toracat |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2020-12-14 14:37:25 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Michael Mráka
2019-10-04 08:46:15 UTC
Do you have work-around for plain mock? Yes, an easy one: # cat >>/etc/mock/rhel-8-x86_64.cfg <<EOF config_opts['yum.conf'] += """ [main] best=False """ EOF Ah, this. I'd expect per-build knob? IMHO per-chroot is enough. I'm also hitting issues with centos-stream missing best=0 parameter. Any plan to implement this feature in copr short-term? > Any plan to implement this feature in copr short-term?
No. If it causes pain for you, then please increase a push on Modularity and DNF team. Fedora devel mailing list is likely the best place.
(In reply to Miroslav Suchý from comment #6) > > Any plan to implement this feature in copr short-term? > > No. If it causes pain for you, then please increase a push on Modularity and > DNF team. Fedora devel mailing list is likely the best place. I see epel8 build root has best=false because of https://bugzilla.redhat.com/show_bug.cgi?id=1756681. Could the same approach be applied in centos-stream config? (In reply to Alfredo Moralejo from comment #7) > I see epel8 build root has best=false because of > https://bugzilla.redhat.com/show_bug.cgi?id=1756681. Could the same approach > be applied in centos-stream config? Thanks, valid point: https://infrastructure.fedoraproject.org/cgit/ansible.git/commit/?id=6fa5f05650050a7b3e4eebf68edea24255d37f4c There's no reason to work-around epel-8 chroots, and keep centos-stream in non-working state. Though I agree with Mirek that this is to be fixed in modularity. (In reply to Pavel Raiskup from comment #8) > (In reply to Alfredo Moralejo from comment #7) > > I see epel8 build root has best=false because of > > https://bugzilla.redhat.com/show_bug.cgi?id=1756681. Could the same approach > > be applied in centos-stream config? > > Thanks, valid point: > https://infrastructure.fedoraproject.org/cgit/ansible.git/commit/ > ?id=6fa5f05650050a7b3e4eebf68edea24255d37f4c > > There's no reason to work-around epel-8 chroots, and keep centos-stream > in non-working state. Though I agree with Mirek that this is to be fixed > in modularity. Apparently that commit broke builds on centos-streams: https://copr-be.cloud.fedoraproject.org/results/@openstack-sig/centos8-deps/centos-stream-x86_64/01120405-erlang-eflame/builder-live.log.gz comparing with previous configs it seems the include for the template should be full path: include('/etc/mock/templates/centos-stream.tpl') From a previous successful run: https://copr-be.cloud.fedoraproject.org/results/%40openstack-sig/centos8-deps/centos-stream-x86_64/01120132-erlang-cuttlefish/configs/centos-stream-x86_64.cfg Meh, sorry. Fixed here: https://infrastructure.fedoraproject.org/cgit/ansible.git/commit/?id=b9543141c93f2954bd7b13113cbaa6d885368bb5 (In reply to Alfredo Moralejo from comment #9) > (In reply to Pavel Raiskup from comment #8) > > (In reply to Alfredo Moralejo from comment #7) > > > I see epel8 build root has best=false because of > > > https://bugzilla.redhat.com/show_bug.cgi?id=1756681. Could the same approach > > > be applied in centos-stream config? > > > > Thanks, valid point: > > https://infrastructure.fedoraproject.org/cgit/ansible.git/commit/ > > ?id=6fa5f05650050a7b3e4eebf68edea24255d37f4c > > > > There's no reason to work-around epel-8 chroots, and keep centos-stream > > in non-working state. Though I agree with Mirek that this is to be fixed > > in modularity. > Thanks for fixing it so fast! And yes, i also think there is a problem with modularity that should be fixed there. IIUC there is ongoing discussions about existing issues with modularity but i guess it will take some time to get it fixed. There's an alternative proposal for: ``` # cat >>/etc/mock/rhel-8-x86_64.cfg <<EOF config_opts['yum.conf'] += """ [main] best=False """ EOF ``` https://pagure.io/copr/copr/issue/1336 Feel free to reopen if you think the RFE isn't matching your usecase. FTR, today I disabled the best=0 work-around. The default mock-core-configs without this hack doesn't seem to mistakenly install modular packages anymore into EL 8 chroots. Let me know if this still breaks something. https://pagure.io/fedora-infra/ansible/c/b9d52f1b6517901e0694aec67d6bf44960e2dff5?branch=main |