Description of problem:

  tests/bugs/bug-767095.t is failing in Rackspace, due to the VMs there having
  long hostnames, which the test isn't able to cope with.

    # tests/bugs/bug-767095.t
    1..19
    ok 1
    ok 2
    No volumes present
    ok 3
    ok 4
    ok 5
    ok 6
    ok 7
    ok 8
    ok 9
    ok 10
    ok 11
    ok 12
    kill: usage: kill [-s sigspec | -n signum | -sigspec] pid | jobspec ... or kill -l [sigspec]
    not ok 13 
    ok 14
    ok 15
    ok 16
    ok 17
    ok 18

  The failing piece of code in the test is this:

    TEST PID=`gluster volume status $V0 | grep patchy1 | awk {'print $5'}`;

  It's expecting output similar to this:

    Status of volume: patchy
    Gluster process						Port	Online	Pid
    ------------------------------------------------------------------------------
    Brick 127.1.1.1:/d/backends/1/patchy			49152	Y	21977
    Brick 127.1.1.2:/d/backends/2/patchy			49154	Y	21996
    Brick 127.1.1.3:/d/backends/3/patchy			49155	Y	22027

  When a hostname is fairly long though, the output from "gluster volume status $V0" wraps around to the next line, so the awk statement doesn't capture the PID.

    Status of volume: patchy
    Gluster process						Port	Online	Pid
    ------------------------------------------------------------------------------
    Brick jc-api-test-node0.cloud.gluster.org:/d/backends/p
atchy1							49160	Y	24658
    Brick jc-api-test-node0.cloud.gluster.org:/d/backends/p
atchy2							49161	Y	24670
    Brick jc-api-test-node0.cloud.gluster.org:/d/backends/p
atchy3							49162	Y	24681

  This can probably be fixed by using --xml mode or similar, which shouldn't have the same problem.

Version-Release number of selected component (if applicable):

  Upstream GlusterFS git master, as of Thu 3rd April 2014.
  commit 997c89b6172116557f981510a94232486ec526b0


How reproducible:

  Every time.