On 26.04.2012 [13:16:04 +0800], Amos Kong wrote: > On 26/04/12 13:02, Nishanth Aravamudan wrote: > >Hi Amos, > > > >On 26.04.2012 [12:29:38 +0800], Amos Kong wrote: > >>Define a SSHHost() functions if fail to import SSHHost module > >>from autotest.server > >> > >>--- > >>Tested in autotest server. > >> > >>Signed-off-by: Amos Kong<[email protected]> > >>--- > >> client/virt/tests/netperf.py | 15 ++++++++++++++- > >> 1 files changed, 14 insertions(+), 1 deletions(-) > >> > >>diff --git a/client/virt/tests/netperf.py b/client/virt/tests/netperf.py > >>index 5fcb2f1..be49154 100644 > >>--- a/client/virt/tests/netperf.py > >>+++ b/client/virt/tests/netperf.py > >>@@ -3,7 +3,20 @@ from autotest.client.shared import error > >> from autotest.client import utils > >> from autotest.client.virt import aexpect, virt_utils > >> from autotest.client.virt import virt_test_utils > >>-from autotest.server.hosts.ssh_host import SSHHost > >>+try: > >>+ from autotest.server.hosts.ssh_host import SSHHostsdf > > --^ > > > >Is that a typo? > > Oh, yes. I changed it for testing. > > >And in what context does that import fail? I guess it must fail on a > >client machine without the server bits? Would it be better to move the > >hosts definition to common? > > As we talked here [1], it's a temporary solution. > > [1] http://test.kernel.org/pipermail/autotest/2012-April/010285.html
Ah sorry, I missed that context -- maybe should be in the changelog in the future (or mention why it's necessary)? Thanks, Nish -- Nishanth Aravamudan <[email protected]> IBM Linux Technology Center _______________________________________________ Autotest mailing list [email protected] http://test.kernel.org/cgi-bin/mailman/listinfo/autotest
