Description of problem: When file permissions don't allow up2date to access a file, the directory repository backend will spit tons of python. error: read failed: Permission denied (13) Traceback (most recent call last): File "/usr/sbin/up2date", line 1188, in ? sys.exit(main() or 0) File "/usr/sbin/up2date", line 766, in main fullUpdate, dryRun=options.dry_run)) File "/usr/sbin/up2date", line 1051, in batchRun batch.run() File "up2dateBatch.py", line 58, in run File "up2dateBatch.py", line 96, in __findPackagesToUpdate File "up2dateBatch.py", line 96, in __findPackagesToUpdate File "packageList.py", line 117, in run File "rhnPackageInfo.py", line 315, in getAvailableAllArchPackageList File "rhnPackageInfo.py", line 153, in availablePackageList File "rpcServer.py", line 110, in doCall File "repoDirector.py", line 20, in listPackages File "rpmSource.py", line 226, in listPackages File "/usr/share/rhn/up2date_client/repoBackends/dirRepo.py", line 129, in lis tPackages hdr = rpmUtils.readHeaderBlob(hdrBuf.unload()) AttributeError: 'NoneType' object has no attribute 'unload' done The real error message tends to get lost amongst all the tracebacks. It'd be better to fail fail with a real error message explaing the problem and listing the name of the file. Version-Release number of selected component (if applicable): up2date-4.3.11-2.1.1 How reproducible: Always Steps to Reproduce: 1.Attempt to install a package off an dir type repository that the local root user doesn't have access to. For example, an NFS export. Actual results: up2date will spew tons of python, burying the import stuff amongst a bunch of tracebacks. Expected results: up2date should give an error message explaining that a file could not be accessed and specifying the name(s) of the file(s) Additional info:
Internal RFE bug #119630 entered; will be considered for future releases.
Thank you for the suggestion. It was passed along to product management, but not committed for a future release.
Closing this as notabug is *terrible*. * The message clearly is a bug. A Python traceback does not constitute an error message. * Having something like this in a supposedly finished OS look really unprofessional. * The bug shouldn't be difficult to fix.
Ping? Seriously, we should not consider crashing Python an adequate error message.
I think the bug was originally closed because it was marked as an RFE when it got imported into the feature request tracker.
REOPENED status has been deprecated. ASSIGNED with keyword of Reopened is preferred.
This bug is filed against RHEL 3, which is in maintenance phase. During the maintenance phase, only security errata and select mission critical bug fixes will be released for enterprise products. Since this bug does not meet that criteria, it is now being closed. For more information of the RHEL errata support policy, please visit: http://www.redhat.com/security/updates/errata/ If you feel this bug is indeed mission critical, please contact your support representative. You may be asked to provide detailed information on how this bug is affecting you.