real kerneloops and suspend/hibernate/resume bugs are hard to separate

Bug #349621 reported by Andy Whitcroft
4
Affects Status Importance Assigned to Milestone
apport (Ubuntu)
Fix Released
Medium
Martin Pitt
Jaunty
Fix Released
Medium
Martin Pitt

Bug Description

Binary package hint: apport

When submitting a real kerneloops we end up with a tags of <arch> apport-kerneloops, for suspend/hibernate/resume bugs we get <arch> apport-kernelopps resume <sleep type>. This is fine for the s/h/r bugs as they can be found with 'apport-kerneloops resume' but there is not way to exclude the resume bugs from the kerneloops list.

We are treating the kerneloops report type in apport as a kernelerror type and we should probabally therefore make the tags reflect that. One option might be to make the general tag apport-kernel and then manually add the apport-kerneloops only to the actual oopses that migth work better.

   oops: apport-kernel apport-kerneloops
   s/r/h: apport-kernel resume hibernate

The other might be to add a subsiduary tag for panics:

  oops: apport-kerneloops panic-oops
  s/r/h: apport-kerneloops resume hibernate

Changed in apport (Ubuntu Jaunty):
importance: Undecided → Medium
status: New → Triaged
Martin Pitt (pitti)
Changed in apport (Ubuntu Jaunty):
assignee: nobody → pitti
status: Triaged → In Progress
Revision history for this message
Martin Pitt (pitti) wrote :

I agree that "apport-kernel" would be nicer as a general tag. I'm just concerned that this breaks backwards compatibility wrt. to LP searches and tools which search for this, thus right now I tend to go with the second alternative. Is that okay for you?

Martin Pitt (pitti)
Changed in apport (Ubuntu Jaunty):
status: In Progress → Fix Committed
Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package apport - 0.148

---------------
apport (0.148) jaunty; urgency=low

  [ Matt Zimmerman ]
  * apport/hookutils.py: add attach_media_build to include information about
    the build of installation media in use (i.e. in a casper live CD
    environment)
  * general-hooks/ubuntu.py: use attach_media_build (LP: #351781)
  * bin/apportcheckresume: Use attach_file_if_exists rather than attach_file to
    avoid spurious error messages about non-existent log files (LP: #351973)

  [ Martin Pitt ]
  * debian/local/ubuntu-bug: Drop generic passthrough of apport-{cli,gtk,kde}
    options since this leads to too much confusion. Instead just support a
    single argument and check whether it is a pid, a package name, a .crash
    file, or a program path. This does the right thing when calling it with a
    .crash file (LP: #347392) and fixes the help output (LP: #344923) Update
    manpage accordingly.
  * apport/hookutils.py: Move attach_media_build() to
    general-hooks/ubuntu.py, since it is Ubuntu specific.
  * bin/apport-retrace: Fix KeyError crash on bugs with an ExecutablePath
    which does not exist any more. Close the bug as invalid instead.
    (LP: #352331)
  * bin/kernel_oops: Add "kernel-oops" tag. Since both bin/kernel_oops and
    bin/apportcheckresume use the "kerneloops" bug class, it previously was
    hard to filter out the bug reports which were real oopses. (LP: #349621)

 -- Martin Pitt <email address hidden> Wed, 01 Apr 2009 18:10:01 +0200

Changed in apport:
status: Fix Committed → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.