Bug 473239
| Summary: | Package contains a summary which is a list, yum doesn't expect this and fail, probably "dbvis" | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Wilko Quak <c.w.quak> |
| Component: | rpm | Assignee: | Panu Matilainen <pmatilai> |
| Status: | CLOSED NEXTRELEASE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | rawhide | CC: | alauschke, aob.phantom, broadwing, c.cerbo, dimi, ffesti, james.antill, jnovy, jvromans, katzj, kazimieras.vaina, lf010, massoniana, mcepl, mcepl, me, pmatilai, randall, thad, tim.lauridsen, toivol |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | i386 | ||
| OS: | Linux | ||
| Whiteboard: | anaconda_trace_hash:79ca450b8cc40f6961a61eaf68977f29c24e4dc93c16eb0693c7eb699d21ca50 | ||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2009-01-07 09:20:59 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
Wilko Quak
2008-11-27 10:24:09 UTC
Created attachment 324851 [details]
Attached traceback automatically from anaconda.
Created attachment 324890 [details]
Attached traceback automatically from anaconda.
There is nothing to triage here. Switching to ASSIGNED so that developers have responsibility to do whatever they want to do with it. Comment on attachment 324851 [details]
Attached traceback automatically from anaconda.
When I dropped the dbvis package before upgrading the upgrade went fine.
*** Bug 473169 has been marked as a duplicate of this bug. *** *** Bug 473766 has been marked as a duplicate of this bug. *** *** Bug 474073 has been marked as a duplicate of this bug. *** Does everyone else from the DUPs also have dbvis installed? I reported one of the DUPs (bug 473169). I can confirm that dbvis caused the problem. I tried uninstalling dbvis before the upgrade, and it went smoothly. dbvis is/was not installed on my system and my yum logs don't show it either. 474073 was my bug entry. This is what I'm thinking of apply upstream, can anyone try it to make sure?
The packages.py file will be /usr/lib/python2.*/site-packages/yum/packages.py
diff --git a/yum/packages.py b/yum/packages.py
index 48dfc8d..71e0a75 100644
--- a/yum/packages.py
+++ b/yum/packages.py
@@ -978,7 +978,9 @@ class YumHeaderPackage(YumAvailablePackage):
self.ver = self.version
self.rel = self.release
self.pkgtup = (self.name, self.arch, self.epoch, self.version, self.rel
- self.summary = misc.share_data(self.hdr['summary'].replace('\n', ''))
+ # Summaries "can be" lists, see BZ 473239, *sigh*
+ self.summary = misc.to_str(self.hdr['summary']).replace('\n', '')
+ self.summary = misc.share_data(self.summary)
self.description = misc.share_data(self.hdr['description'])
self.pkgid = self.hdr[rpm.RPMTAG_SHA1HEADER]
if not self.pkgid:
I reported bug 473766 I had dbvis installed, and removing it before the upgrade allowed the upgrade to proceed smoothly. No, I did not have dbvis installed. My 'offender' is softsqueeze-3.5-1.i386. is this an old pkg? is it still possible to generate pkgs with summary as a list? Summary is certainly not supposed to be a list when given to python (it technically is an array type as it's an i18nstring tag). Can somebody add a link to one of these packages exhibiting this syndrome (dbvis or softsqueeze)? I'll try to restore the exact softsqueeze rpm from a backup (probably tomorrow). I grabbed a copy of the dbvis rpm from the dbvis website. It has a null summary field. It comes out as [] from the python interface and (none) when I query it with rpm from the cli. It must have been made on either an odd version of rpm, not made with rpm or made on an old ver of rpm. B/c no recent version will allow a missing or non-existent summary field in a spec file. Ahhah... I see. There's a dumb bug in rpm-python: attempting to retrieve non-existent tag doesn't honor the array/single value rules used elsewhere. I remember seeing that code and also remember leaving it alone for some reason, but whatever the "reason" was it doesn't seem very sensible now that I look at it... Rpm 4.6.x is supposed to behave identically to 4.4.x in case of non-existent (but otherwise legal) tag, the supposed compatibility might've been the "reason" for leaving the code the way it is. And it actually does, *except* for RPMTAG_OLDFILENAMES, RPMTAG_GROUP, RPMTAG_DESCRIPTION and RPMTAG_SUMMARY, which in rpm 4.4.x get a special treatment, but *only* when called from python. What fun :) Will fix. Done upstream, will push to rawhide later today hopefully. *** Bug 474889 has been marked as a duplicate of this bug. *** *** Bug 475874 has been marked as a duplicate of this bug. *** rpm-4.6.0-0.rc3.1.fc10 has been submitted as an update for Fedora 10. http://admin.fedoraproject.org/updates/rpm-4.6.0-0.rc3.1.fc10 *** Bug 476740 has been marked as a duplicate of this bug. *** rpm-4.6.0-0.rc3.1.fc10 has been pushed to the Fedora 10 testing repository. If problems still persist, please make note of it in this bug report. If you want to test the update, you can install it with su -c 'yum --enablerepo=updates-testing update rpm'. You can provide feedback for this update here: http://admin.fedoraproject.org/updates/F10/FEDORA-2008-11332 Created attachment 327807 [details]
Attached traceback automatically from anaconda.
rpm-4.6.0-0.rc3.1.fc10 has been pushed to the Fedora 10 stable repository. If problems still persist, please make note of it in this bug report. Created attachment 328971 [details]
Attached traceback automatically from anaconda.
Created attachment 329914 [details]
Attached traceback automatically from anaconda.
*** Bug 483813 has been marked as a duplicate of this bug. *** uninstalling dbvis package allowed the install to proceed. Created attachment 334462 [details]
Attached traceback automatically from anaconda.
|