Project

General

Profile

Actions

Bug #45012

closed

qa/cephfs/test_backtrace_repair: stuck forever when ran by vstart_runner.sh

Added by Xiubo Li about 4 years ago. Updated about 4 years ago.

Status:
Resolved
Priority:
Normal
Assignee:
Category:
-
Target version:
-
% Done:

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
Labels (FS):
qa
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

2020-04-09 05:20:30,207.207 INFO:__main__:I think my launching pid was 29093
2020-04-09 05:20:30,207.207 INFO:__main__:Running ['sudo', 'python3', '-c', '\nimport glob\nimport re\nimport os\nimport subprocess\n\ndef find_socket(client_name):\n        asok_path = "/tmp/ceph-asok.fV0BkU//client.0.29093.asok"\n        files = glob.glob(asok_path)\n\n        # Given a non-glob path, it better be there\n        if "*" not in asok_path:\n            assert(len(files) == 1)\n            return files[0]\n\n        for f in files:\n                pid = re.match(".*\\.(\\d+)\\.asok$", f).group(1)\n                if os.path.exists("/proc/{0}".format(pid)):\n                        return f\n        raise RuntimeError("Client socket {0} not found".format(client_name))\n\nprint(find_socket("client.0"))\n']
2020-04-09 05:20:30,416.416 INFO:tasks.cephfs.fuse_mount:Found client admin socket at /tmp/ceph-asok.fV0BkU//client.0.29093.asok
2020-04-09 05:20:30,895.895 INFO:__main__:Running ['stat', '--file-system', '--printf=%T\n', '--', '/tmp/tmp7q9P_t/mnt.0']
2020-04-09 05:20:30,927.927 INFO:tasks.cephfs.fuse_mount:ceph-fuse is mounted on /tmp/tmp7q9P_t/mnt.0
2020-04-09 05:20:30,929.929 INFO:__main__:Running ['sudo', 'chmod', '1777', '/tmp/tmp7q9P_t/mnt.0']
2020-04-09 05:20:31,078.078 INFO:__main__:Running ['mkdir', 'parent_a']
2020-04-09 05:20:31,148.148 INFO:__main__:Running ['touch', 'parent_a/alpha']
2020-04-09 05:20:31,208.208 INFO:__main__:Running ['python', '-c', '\nimport os\nimport stat\n\nprint(os.stat("/tmp/tmp7q9P_t/mnt.0/parent_a/alpha").st_ino)\n']
2020-04-09 05:20:31,267.267 INFO:__main__:Running ['./bin/ceph', 'fs', 'dump', '--format=json']
2020-04-09 05:20:31,996.996 INFO:__main__:Running ['./bin/ceph', 'daemon', 'mds.b', 'flush', 'journal']
2020-04-09 05:20:32,515.515 INFO:tasks.cephfs.filesystem:_json_asok output: {
    "message": "",
    "return_code": 0
}

It stuck here.

Actions #1

Updated by Xiubo Li about 4 years ago

  • Assignee set to Xiubo Li
Actions #2

Updated by Xiubo Li about 4 years ago

  • Status changed from New to In Progress
Actions #3

Updated by Xiubo Li about 4 years ago

It is because that the remote.sh in vstart_runner.sh will call teuthology/misc.py's sh(), which has no stdin/stdout paramter.

And for the "ceph-dencoder" command it will wait data input from stdin.

Actions #4

Updated by Xiubo Li about 4 years ago

  • Status changed from In Progress to Resolved

Fixed by:

commit 2cc0ee709c36eabe03311a00b72295da468bccf4
Author: Rishabh Dave <ridave@gmail.com>
Date:   Fri Mar 13 07:03:50 2020 +0000

    qa/vstart_runner: update vstart_runner.LocalRemote.sh

    Commit 9f6c764f10f replaces remote.run calls by remote.sh without
    updating the definition of vstart_runner.LocalRemote.sh which breaks the
    cephfs tests when executed locally.

    Fixes: https://tracker.ceph.com/issues/44579
    Signed-off-by: Rishabh Dave <ridave@redhat.com>

Actions

Also available in: Atom PDF