Bug 797348
Summary: | ERROR with transaction check vs depsolve: | ||||||
---|---|---|---|---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | John Antony <john_antony40> | ||||
Component: | yum | Assignee: | Seth Vidal <skvidal> | ||||
Status: | CLOSED NOTABUG | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||
Severity: | unspecified | Docs Contact: | |||||
Priority: | unspecified | ||||||
Version: | rawhide | CC: | ffesti, james.antill, maxamillion, michal, pmatilai, tim.lauridsen, zpavlas | ||||
Target Milestone: | --- | ||||||
Target Release: | --- | ||||||
Hardware: | Unspecified | ||||||
OS: | Unspecified | ||||||
Whiteboard: | |||||||
Fixed In Version: | Doc Type: | Bug Fix | |||||
Doc Text: | Story Points: | --- | |||||
Clone Of: | Environment: | ||||||
Last Closed: | 2012-03-14 19:06:27 UTC | Type: | --- | ||||
Regression: | --- | Mount Type: | --- | ||||
Documentation: | --- | CRM: | |||||
Verified Versions: | Category: | --- | |||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
Cloudforms Team: | --- | Target Upstream Version: | |||||
Embargoed: | |||||||
Attachments: |
|
Description
John Antony
2012-02-25 02:00:38 UTC
What looks like the same issue hits with yum-3.4.3-22.fc18. An attempt to run yum update firefox 'xulrunner*' on a rawhide installations (versions to be updated to are firefox-11.0-1.fc18.x86_64, xulrunner-11.0-3.fc18.x86_64 and xulrunner-devel-11.0-3.fc18.x86_64) ends up with: ERROR with transaction check vs depsolve: gecko-libs(x86-64) = 11.0-3-1 is needed by firefox-11.0-1.fc18.x86_64 but "yum provides 'gecko-libs(x86-64) = 11.0-3-1'" query comes back with xulrunner-11.0-3.fc18.x86_64 so everything seems to be fine. Now what? A log from run with -d5 yum flag is attached. Created attachment 570065 [details]
an output from "yum update -d5 firefox 'xulrunner*'" on the current rawhide installation
There are at least 3 different problems in this BZ, the first: ERROR with transaction check vs depsolve: gnome-utils = 1:3.2.1-1.fc17 is needed by gnome-system-log-1:3.2.1-1.fc17.i686 ...means that yum created a transaction, but rpm refused to run it due to the above missing dep. This often means that yum assumes the dep. was there (but it isn't, due to rpm --force --nodeps usage). How reproducible: yum -y update gnome-utils* [...] --> Finished Dependency Resolution Error: Package: 1:gnome-system-log-3.2.1-1.fc17.i686 (@rawhide/17) Requires: gnome-utils = 1:3.2.1-1.fc17 Removing: 1:gnome-utils-3.2.1-1.fc17.i686 (@rawhide/17) gnome-utils = 1:3.2.1-1.fc17 Obsoleted By: gnome-screenshot-3.3.2-1.fc17.i686 (rawhide) Not found ...this is just a generic packaging bug in gnome-system-log or gnome-screenshot or something else. Yum is just telling you that you can't do the above due to it. The third looks like a bug in rpmbuild, and the package for firefox. The problem is that the package requires: ERROR with transaction check vs depsolve: gecko-libs(x86-64) = 11.0-3-1 is needed by firefox-11.0-1.fc18.x86_64 ...which can't be valid. The other package provides: xulrunner-11.0-3.fc18.x86_64 : XUL Runtime for Gecko Applications Repo : rawhide Matched from: Provides : gecko-libs(x86-64) = 11.0 ...the problem is that '-' is not a valid character in a version _or_ a release "number" ... yum is just splitting at the first '-' and so going with if v=11.0 and r=3-1 ... which is wrong, but works. rpm is presumably doing something different. But, as I said, rpmbuild should be throwing an error at build time because foo-11.0-3-1 would be taken as n=foo-11.0, v=3, r=1. That could be a problem with firefox and xulrunner after all despite of a yum claim that xulrunner-11.0-3.fc18 provides gecko-libs(x86-64) = 11.0-3-1. Looking directly at available packages I see: # rpm -qRp firefox-11.0-1.fc18.x86_64.rpm | grep gecko gecko-libs(x86-64) = 11.0-3-1 but # rpm -q --provides -p xulrunner-11.0-3.fc18.x86_64.rpm | \ grep gecko gecko-libs = 11.0 gecko-libs(x86-64) = 11.0 So why "yum provides ..." comes back with a suitable package? (In reply to comment #4) > So why "yum provides ..." comes back with a suitable package? Ah, ok. See comment #3. :-) Effects are quite confusing. |