Hello, I have since 8.0 betas had problems with rpm. I originally submitted a bug report(#68056) about rpm hanging during beta. At the time the bug was reported fixed. Later RedHat 8.0 was released and the flood gates opened with a number of bug reports(#73097, #73415, #74726, #75553, #75647, #77480, #77562, #77988, #77997, #78366, #78679, #79272) with the same symptom came in. Most have been closed with the the result of WORKSFORME, which seems to me to be a poor choice. It may work for the maintainer, Jeff Johnson(jbj@redhat.com), but as the bug reports mentioned above show it sure isn't working for everyone else. There as been a test release made, which is mentioned in some of the bug reports above. It is 4.1-9 and is at ftp://people.redhat.com/jbj/test-4.1/ It doesn't fix the problem, it just makes it more rare. The especially odd thing in my opinion is why 4.1-9 or something like it hasn't been released as errata. Jeff seems to suggest it's use to many that complain of the bug and has never mentioned to my knowledge a reason not to use it. It has been months since my original bug report and months since RedHat 8.0's release, yet there hasn't been a fix made, or even a errata to help the problem. Mostly what Jeff seems to have done is try to ignore the problem while he Slowly tries to fix the problem. The best hint at any hope seems to be this comment from him in #77988. "AFAIAC the "real" solution is to use NPTL on /dev/futex, almost there." One of the bugs, #75647, was closed as rawhide. The funny thing is rawhide is now 4.2-* and doesn't seem to include the fix in 4.1-9, as 4.2-* locks up just as bad as 4.1-1.06. It would seem to me if a real workaround was to be until a complete fix was created it would be to release a set of rpms with all the necessary changes to disable concurrent write access, and hence revert rpm to a more RedHat 7.3 state. I have said before and have seen others mention that they won't migrate servers to RedHat 8.0 until this issue is fixed. I have also had a RedHat developer agree with me that this issue should have been at least somewhat addressed with an errata long ago. Why a bug that would stop customers from migrating to the latest release for months hasn't been fixed doesn't make sense to me. -- Psyche-list mailing list Psyche-list@redhat.com https://listman.redhat.com/mailman/listinfo/psyche-list
