plainbox crashes while attempting to create session.next

Bug #1396532 reported by Zygmunt Krynicki
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
PlainBox (Toolkit)
Triaged
Low
Unassigned

Bug Description

I just looked up this crash from a comment in another bug:

2014-11-26 14:27:23 [pid:2392, thread:MainThread, reltime:13904ms] CRITICAL plainbox.crashes: Executable 'checkbox-launcher' invoked with Namespace(command=<checkbox_ng.commands.launcher.LauncherCommand object at 0x26cf650>, debug_console=False, debug_interrupt=False, dont_suppress_output=False, dry_run=False, launcher='/usr/bin/checkbox-cli', log_level=None, non_interactive=False, pdb=False, trace=[]) has crashed
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/plainbox/impl/clitools.py", line 529, in dispatch_and_catch_exceptions
    return self.dispatch_command(ns)
  File "/usr/lib/python3/dist-packages/plainbox/impl/clitools.py", line 525, in dispatch_command
    return ns.command.invoked(ns)
  File "/usr/lib/python3/dist-packages/checkbox_ng/commands/launcher.py", line 84, in invoked
    launcher).run()
  File "/usr/lib/python3/dist-packages/checkbox_ng/commands/newcli.py", line 113, in run
    return self.do_normal_sequence()
  File "/usr/lib/python3/dist-packages/checkbox_ng/commands/newcli.py", line 165, in do_normal_sequence
    self.run_all_selected_jobs()
  File "/usr/lib/python3/dist-packages/plainbox/impl/commands/inv_run.py", line 718, in run_all_selected_jobs
    self.run_single_job(job)
  File "/usr/lib/python3/dist-packages/plainbox/impl/commands/inv_run.py", line 722, in run_single_job
    self.run_single_job_with_ui(job, self.get_ui_for_job(job))
  File "/usr/lib/python3/dist-packages/plainbox/impl/commands/inv_run.py", line 739, in run_single_job_with_ui
    self.manager.checkpoint()
  File "/usr/lib/python3/dist-packages/plainbox/impl/session/manager.py", line 350, in checkpoint
    self.storage.save_checkpoint(data)
  File "/usr/lib/python3/dist-packages/plainbox/impl/session/storage.py", line 371, in save_checkpoint
    return self._save_checkpoint_unix_py32(data)
  File "/usr/lib/python3/dist-packages/plainbox/impl/session/storage.py", line 640, in _save_checkpoint_unix_py32
    os.O_WRONLY | os.O_CREAT | os.O_EXCL, 0o644)
OSError: [Errno 17] File exists: '/home/u/.cache/plainbox/sessions/pbox-54yfy1.session/session.next'

The full tarball is in
https://bugs.launchpad.net/plainbox/+bug/1393291/comments/8

Zygmunt Krynicki (zyga)
Changed in plainbox:
status: New → Confirmed
Zygmunt Krynicki (zyga)
Changed in plainbox:
importance: Undecided → Critical
Jerry Kao (jerry.kao)
tags: added: ce-qa-concern
Revision history for this message
Zygmunt Krynicki (zyga) wrote :

I'm demoting this to high because:
- this issue doesn't happen often
- it can always happen in principle (we need to handle this case, whatever that means)
- we have no idea why it happened (it _may_ happen though why is a big question)
- might (need to check) be possible to hit this if we open the same session twice somehow

Changed in plainbox:
status: Confirmed → Triaged
importance: Critical → High
Revision history for this message
Zygmunt Krynicki (zyga) wrote :
Changed in plainbox:
importance: High → Low
Revision history for this message
Zygmunt Krynicki (zyga) wrote :

09:02 <@spineau> zyga: did we merge a specific patch for https://bugs.launchpad.net/plainbox/+bug/1396532? I see a reference to https://code.launchpad.net/~zkrynicki/checkbox/storage-fixes/+merge/243883
09:02 <@spineau> zyga: it's all about session.next
09:03 <@spineau> zyga: I'll probably demote again the priority of this bug (High so far)
09:04 <@zyga> spineau: we did merge something that could cause that to happen
09:04 <@zyga> spineau: the error recovery path in session storage was wrong
09:05 <@spineau> zyga: does it mean that this bug can be closed?
09:05 <@zyga> spineau: the bug can be demoted low and left open as there's no recovery path to when that erorr (if ever) happens, we should think of some UI for that
09:05 <@zyga> spineau: I think that I'd leave it open as it can still (in theory) happen, I need to look at locking code in the storage class to be sure
09:05 <@spineau> zyga: ok to keep it open. Thanks
09:05 <@zyga> spineau: but let's demote it to low

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.