On 12/25/16 00:14, Patrick O'Callaghan wrote: > It seems unlikely that a specific package could be the cause of the bug > rather than dnf itself (including the plugin of course). Package > installation is often going to run scripts (complex ones in the case of > akmod) so it would be up to dnf to handle the locking correctly.
You are right that it shouldn't be BZ'd against akmods. akmodsposttrans is called by virtual of the postinst.d entry. This calls akmods which subsequently calls akmodsbuild and then dnf to install the generated rpm. It seems to me the major difference in the working/failing log is... Working: Running transaction check Waiting for process with pid 24889 to finish. Transaction check succeeded. Running transaction test Transaction test succeeded. Running transaction Failing: Running transaction check Transaction check succeeded. Running transaction test Transaction test succeeded. Running transaction For some reason there is no "Waiting for process" in the failing case. So, it would seem, either the PID being checked had already finished but another one had started and locked the database but somehow dnf is unaware of it and tries to move forward with the installation. -- You're Welcome Zachary Quinto _______________________________________________ users mailing list -- users@lists.fedoraproject.org To unsubscribe send an email to users-le...@lists.fedoraproject.org
