postgresql-jdbc-9.4.1209-5.fc25 fails to build from source. This affects Base Runtime, Rawhide, and F25. The reason for this is a failing test, org.postgresql.test.jdbc2.ConnectTimeoutTest. The ConnectTimeoutTest test makes poor assumptions about the non-routability of IP address 10.255.255.1 that is hard coded in the test. It turns out that IP is routable and reachable in the Fedora build environment. This issue has already been reported upstream: https://github.com/pgjdbc/pgjdbc/issues/556 Since there are no guarantees about the routability of any static IP address, the usefulness of the ConnectTimeoutTest test is questionable. The best workaround is to completely disable the test to prevent it from failing the build of the package.
Created attachment 1234524 [details] proposed patch to correct FTBFS issue Although the unpredictable test affects builds for all versions of Fedora, the attached patch to disable ConnectTimeoutTest is based from commit 9b77a0e3ea2b09b5d5771d172683bbd14312d9e2--the latest version of the package based on upstream pgjdbc 9.4.1209. (pgjdbc 9.4.1209 is the latest upstream version that currently has a package for Fedora 25.) If possible, Base Runtime would appreciate a new Fedora 25 based package based on pgjdbc 9.4.1209 to minimize the chances of other upstream changes in pgjdbc causing undesirable side effects.
Please apply the patch or respond with a reason it should not be applied within two weeks. After that time, a provenpackager will go ahead and apply it.
Thanks for the analysis and the patch! (In reply to Merlin Mathesius from comment #1) > If possible, Base Runtime would appreciate a new Fedora 25 based package > based on pgjdbc 9.4.1209 to minimize the chances of other upstream changes > in pgjdbc causing undesirable side effects. I don't understand this request, what side effects? (In reply to Merlin Mathesius from comment #2) > Please apply the patch or respond with a reason it should not be applied > within two weeks. After that time, a provenpackager will go ahead and apply > it. The proper fix is to disable this test upstream, or fix the test upstream. > It turns out that IP is routable and reachable in the Fedora build > environment. What is meant to be Fedora build environment? Both Koji and Copr build the pacakage fine. Because I agree that this deserves proper fix, I'm curious how burning this issue is for you ... is it OK to wait for proper upstream resolution? Note also the upstream CI: https://copr.fedorainfracloud.org/coprs/g/pgjdbc/pgjdbc-travis/builds/ (applying this in Fedora only -- not upstream -- will increase the the differences in the specfiles)
(In reply to Pavel Raiskup from comment #3) > Thanks for the analysis and the patch! > > (In reply to Merlin Mathesius from comment #1) > > If possible, Base Runtime would appreciate a new Fedora 25 based package > > based on pgjdbc 9.4.1209 to minimize the chances of other upstream changes > > in pgjdbc causing undesirable side effects. > > I don't understand this request, what side effects? > What we mean by side-effects is impact on other packages if this package was rebased to a newer upstream version. * If it grows dependencies on new packages or packages with a higher version than currently in base runtime, it means that we have to recursively address that. * If it changes API or ABI in any way, it may result in packages that depend on this package to fail to build or run. For the current phase of base runtime development, we are trying very hard to minimize differences from what shipped in the frozen package set as part of Fedora 25 (aka Fedora 25 with no updates). > (In reply to Merlin Mathesius from comment #2) > > Please apply the patch or respond with a reason it should not be applied > > within two weeks. After that time, a provenpackager will go ahead and apply > > it. > > The proper fix is to disable this test upstream, or fix the test upstream. > Yes, ideally this patch should make its way upstream. We would like a short-term solution in Fedora, however. > > It turns out that IP is routable and reachable in the Fedora build > > environment. > > What is meant to be Fedora build environment? Both Koji and Copr build the > pacakage fine. We're investigating this; it builds in the fedorainfracloud, which may have this IP routable. However, we just realized that this may be partly due to our testing which was done on machines within the RHT network, for which that IP is *definitely* routable. I'll update this ticket once I've figured that out. > Because I agree that this deserves proper fix, I'm curious > how burning this issue is for you ... is it OK to wait for proper > upstream resolution? > These FTBFS failures are blocking the base runtime effort, so if "wait for proper upstream resolution" means "shortly after the new year", that's probably fine. If it means "sometime this spring or later", that's going to be an issue we need resolved sooner. > Note also the upstream CI: > https://copr.fedorainfracloud.org/coprs/g/pgjdbc/pgjdbc-travis/builds/ > (applying this in Fedora only -- not upstream -- will increase the > the differences in the specfiles)
Ok, you want us to *not* rebase in F25. Afair, in newer pgjdbc there are some dependencies which are not available in F25; and rebase in stable was not planed anyway. Feel free to push the patch as a provenpackager then (as proposed) on deadline unless this gets resolved otherwise until that time. Thanks!
(In reply to Pavel Raiskup from comment #5) > Ok, you want us to *not* rebase in F25. Afair, in newer pgjdbc there are > some > dependencies which are not available in F25; and rebase in stable was not > planed anyway. > > Feel free to push the patch as a provenpackager then (as proposed) on > deadline > unless this gets resolved otherwise until that time. Thanks! If you're giving us the go-ahead, mind if we just push and build today? We can always revert it in favor of the upstream resolution later if it turns up.
Sorry for delay, feel free :)
I rebased the patch against F25 & rawhide and started builds. Please keep this bug opened until this is resolved upstream. Thanks again for the patch!
This message is a reminder that Fedora 25 is nearing its end of life. Approximately 4 (four) weeks from now Fedora will stop maintaining and issuing updates for Fedora 25. It is Fedora's policy to close all bug reports from releases that are no longer maintained. At that time this bug will be closed as EOL if it remains open with a Fedora 'version' of '25'. Package Maintainer: If you wish for this bug to remain open because you plan to fix it in a currently maintained version, simply change the 'version' to a later Fedora version. Thank you for reporting this issue and we are sorry that we were not able to fix it before Fedora 25 is end of life. If you would still like to see this bug fixed and are able to reproduce it against a later version of Fedora, you are encouraged change the 'version' to a later Fedora version prior this bug is closed as described in the policy above. Although we aim to fix as many bugs as possible during every release's lifetime, sometimes those efforts are overtaken by events. Often a more recent Fedora release includes newer upstream software that fixes bugs or makes them obsolete.
This issue seems to be already fixed in upstream and will be closed with rebase to higher version with #1810342.
FEDORA-2020-c787fa2551 has been submitted as an update to Fedora 32. https://bodhi.fedoraproject.org/updates/FEDORA-2020-c787fa2551
postgresql-jdbc-42.2.11-1.fc32 has been pushed to the Fedora 32 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2020-c787fa2551
postgresql-jdbc-42.2.11-1.fc31 has been pushed to the Fedora 31 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2020-9855e9a60f
Thank you for following up on this!
FEDORA-2020-c787fa2551 has been pushed to the Fedora 32 stable repository. If problem still persists, please make note of it in this bug report.
FEDORA-2020-9855e9a60f has been pushed to the Fedora 31 stable repository. If problem still persists, please make note of it in this bug report.