Bug #17329
closedmodprobe test fails
0%
Description
Centos + ubuntu and latest jewel
Following test fails during modprobe test, is that line necessary to check?
2016-09-20T01:05:51.151 INFO:teuthology.orchestra.run.vpm005.stdout:pg_num: 12
2016-09-20T01:05:51.152 INFO:teuthology.run_tasks:Running task exec...
2016-09-20T01:05:51.191 INFO:teuthology.task.exec:Executing custom commands...
2016-09-20T01:05:51.387 INFO:teuthology.task.exec:Running commands on role client.0 host ubuntu@vpm083.front.sepia.ceph.com
2016-09-20T01:05:51.387 INFO:teuthology.orchestra.run.vpm083:Running: "sudo TESTDIR=/home/ubuntu/cephtest bash -c 'modprobe -r rbd'"
2016-09-20T01:05:51.450 INFO:teuthology.orchestra.run.vpm083:Running: "sudo TESTDIR=/home/ubuntu/cephtest bash -c 'modprobe --first-time rbd'"
2016-09-20T01:05:51.805 INFO:teuthology.orchestra.run.vpm083:Running: "sudo TESTDIR=/home/ubuntu/cephtest bash -c 'test ! -f /sys/module/rbd/parameters/single_major'"
2016-09-20T01:05:51.877 ERROR:teuthology.run_tasks:Saw exception from tasks.
Traceback (most recent call last):
File "/home/teuthworker/src/teuthology_master/teuthology/run_tasks.py", line 66, in run_tasks
manager = run_one_task(taskname, ctx=ctx, config=config)
File "/home/teuthworker/src/teuthology_master/teuthology/run_tasks.py", line 45, in run_one_task
return fn(**kwargs)
File "/home/teuthworker/src/teuthology_master/teuthology/task/exec.py", line 54, in task
c],
File "/home/teuthworker/src/teuthology_master/teuthology/orchestra/remote.py", line 192, in run
r = self._runner(client=self.ssh, name=self.shortname, **kwargs)
File "/home/teuthworker/src/teuthology_master/teuthology/orchestra/run.py", line 403, in run
r.wait()
File "/home/teuthworker/src/teuthology_master/teuthology/orchestra/run.py", line 166, in wait
label=self.label)
CommandFailedError: Command failed on vpm083 with status 1: "sudo TESTDIR=/home/ubuntu/cephtest bash -c 'test ! -f /sys/module/rbd/parameters/single_major'"
<pre>
Updated by Ilya Dryomov over 7 years ago
This test assumes that the suite is scheduled with -k <something>. If for some reason you wanted to test a distro kernel, that should have been -k distro, but normally krbd suite is scheduled with -k testing.
Updated by Vasu Kulkarni over 7 years ago
the default behaviour without -k option is distro, I mainly test on latest distro kernels and ignore known issues, I will comment out the lines after modprobe in yaml and retest.
Updated by Ilya Dryomov over 7 years ago
The default is to leave the kernel unchanged, or so it was the last time I checked. Even if that got changed and the default now is distro, krbd suite still needs to be scheduled with -k to force the kernel: snippet into the job definition.
This job tests rbd CLI against a specific (older) kernel. That assert is there to make sure that that kernel actually gets installed - this ticket just proved its usefulness ;)
Updated by Ilya Dryomov over 7 years ago
In any case, commenting out the assert coupled with scheduling without -k is not the way to go - that would defeat the purpose of the test.