From Fedora Project Wiki

(clean up the start a bit)
(double grrrr)
Line 2: Line 2:
|description=This test case tests the functionality of the ABRT kerneloops feature. We will check that the {{command|abrt-dump-oops}} works and provides correct amount of crashes detected.
|description=This test case tests the functionality of the ABRT kerneloops feature. We will check that the {{command|abrt-dump-oops}} works and provides correct amount of crashes detected.
|actions=
|actions=
{{admon/warn|Syslog|This test abuses syslog a bit - don't forget to restart it at the end of the test case.}}
{{admon/note|Syslog|This test abuses syslog a bit - don't forget to restart it at the end of the test case.}}
# Download files:
# Download files:
#* [http://git.fedorahosted.org/git/?p=abrt.git;a=blob_plain;f=examples/oops1.test;h=e4f02d26c11cf613c522d577486ed9dfc597b1b7;hb=HEAD oops1.test]
#* [http://git.fedorahosted.org/git/?p=abrt.git;a=blob_plain;f=examples/oops1.test;h=e4f02d26c11cf613c522d577486ed9dfc597b1b7;hb=HEAD oops1.test]

Revision as of 16:49, 29 March 2011

Description

This test case tests the functionality of the ABRT kerneloops feature. We will check that the abrt-dump-oops works and provides correct amount of crashes detected.


How to test

Note.png
Syslog
This test abuses syslog a bit - don't forget to restart it at the end of the test case.
  1. Download files:
  2. Test that stand-alone detector tool works properly; Do as a root for both files:
    $ abrt-dump-oops -d -o oops1.test
    abrt-dump-oops: Found oopses: 1
    Version: 2.6.27.9-159.fc10.i686 #1
    BUG: unable to handle kernel NULL pointer dereference at 00000000
    ...
    EIP: [<f88dec25>] radeon_cp_init_ring_buffer+0x90/0x302 [radeon] SS:ESP 0068:f0a0cf08
    abrt-dump-oops: Creating dump directories
  3. Open abrt-gui and check that dump directory was created
  4. And now test if ABRT's scanning of /var/log/messages works, you'll need root privs for this
  5. Make sure /etc/abrt/abrt_event.conf contains EVENT=post-create analyzer=Kerneloops abrt-action-analyze-oops and EVENT=report_kerneloops.org analyzer=Kerneloops abrt-action-kerneloops. And /etc/abrt/abrt.conf contains abrt-dump-oops = abrt-dump-oops -drw /var/log/messages. If changed anything, restart abrtd: service abrtd restart.
  6. Concatenate oops1.test to /var/log/messages e.g.:
    cat oops1.test >> /var/log/messages
  7. If you are logged as a root: Wait for abrt-applet (make sure it's running) and notification to emerge to show the notification about the detected kerneloops. Otherwise open abrt-gui as a root instead.

Expected Results

  1. After running abrt-dump-oops on oops1.test crash should be created and visible via abrt-gui and in the terminal.
  2. Running abrt-dump-oops on not_oops3.test should not create crash case.
  3. ABRT should automatically detect OOPS in /var/log/messages and visible via abrt-gui.

Now restart rsyslog service so writing to /var/log/messages works correctly again.