err value was overwritten by thread return number which resulted in
incorrectly reported failed tests as pass.
Signed-off-by: NLukáš Doktor <ldoktor@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

That function was supposed to return a CmdResult object,
but it wasn't due to a rebase mistake.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

This is a legacy of the days where the kvm kernel modules would
be installed either by compiling the source code or via "kmod"
like packages, and the loading and unloading of kernel modules
would depend on that.

Nowadays, we can assume that the kvm modules always come from
the kernel packages.

One last piece of refiniment here would be to remove the module
loading/unloading feature from the base installer code.
Signed-off-by: NCleber Rosa <crosa@redhat.com>

Rather than just print the top SHA1, it is frequently
useful for quick reference to print the top commit
message.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NCleber Rosa <crosa@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

In order to make the code a little more robust, so
we can achieve better success rates in flaky networking
infrastructures, if we had a failure downloading a file
from Brew or Koji, retry during a timeout (30 seconds),
in fixed time intervals (3 seconds). If at the end of
this timeout we still didn't get it, give up and throw
an error.

As the code uses frequently the terminology 'get' for
this rather than 'download', I opted for an internal
method _get() that encapsulates all the process nicely.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

This is yet another library that we want to move to
autotest proper, since it is in fact already there.
So it needs to be in virttest.staging rather than virttest.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Following other tests (destroy, domid, reboot, setvcpus, undefine,
vncdisplay, dominfo, dommemstat, domstate) - adjust the defaults for
the local_ip to be "LOCAL.EXAMPLE.COM" and the remote_ip to be
"REMOTE.EXAMPLE.COM" in order to cause the test to be skipped if
the defaults weren't changed.

Following other tests (destroy, domid, reboot, setvcpus, undefine,
vncdisplay, dominfo, dommemstat, domstate) - adjust the defaults for
the local_ip to be "LOCAL.EXAMPLE.COM" and the remote_ip to be
"REMOTE.EXAMPLE.COM" in order to cause the test to be skipped if
the defaults weren't changed.
Signed-off-by: NJohn Ferlan <jferlan@redhat.com>

Running the libvirt tests requires configured guest in libvirt.
This introduce new options, the first one is '--install' which
will install the guest using import method.

There is also second new option '--remove' which will remove the
guest from libvirt without deleting the disk file.
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

Signed-off-by: NDongsheng Yang <yangds.fnst@cn.fujitsu.com>

Signed-off-by: NDongsheng Yang <yangds.fnst@cn.fujitsu.com>

Signed-off-by: NDongsheng Yang <yangds.fnst@cn.fujitsu.com>

Wrong path in snapshot_external_disk cause the error in cleanup.
Signed-off-by: Nyangdongsheng <yangds.fnst@cn.fujitsu.com>

Signed-off-by: NDongsheng Yang <yangds.fnst@cn.fujitsu.com>

Signed-off-by: NDongsheng Yang <yangds.fnst@cn.fujitsu.com>

Signed-off-by: Nyangdongsheng <yangds.fnst@cn.fujitsu.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Add kickstart and configuration files for the new version
of JeOS, based on Fedora 19.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NYu Mingfei <yumingfei@cn.fujitsu.com>

Signed-off-by: NAlex Jia <ajia@redhat.com>

Signed-off-by: NAlex Jia <ajia@redhat.com>

Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Sometimes, we might have qemu output like

'drive_image1: /usr/local/autotest/tests/virt/shared/data/images/win7-64-sp1.qcow2 (qcow2)\n
[not inserted]\n
ide0-cd0: /usr/local/autotest/tests/virt/shared/data/isos/windows/winutils.iso (raw, read-only)\n
Removable device: not locked, tray closed\n [not inserted](qemu) '

Breaking the assumption of a (qemu) prompt separated from
the rest of the output. So we have to be less exigent when
matching the prompt.
Signed-off-by: NLucas Meneghel Rodrigues <lmr@redhat.com>

Signed-off-by: NYu Mingfei <yumingfei@cn.fujitsu.com>

Signed-off-by: NYu Mingfei <yumingfei@cn.fujitsu.com>

Signed-off-by: NYu Mingfei <yumingfei@cn.fujitsu.com>

Currently, remote.handle_prompts(...) only support ssh and telnet.
To connect ESX server with "esx://IP/...", it should add prompts
to matched list for esx connection.
Signed-off-by: NYu Mingfei <yumingfei@cn.fujitsu.com>