- 05 7月, 2016 1 次提交
-
-
由 vivianQizhu 提交于
-
- 01 7月, 2016 1 次提交
-
-
由 Xu Han 提交于
Signed-off-by: NXu Han <xuhan@redhat.com>
-
- 29 6月, 2016 2 次提交
-
-
由 MeYang 提交于
-
由 Wei Jiangang 提交于
There's no attribute TestFailed. Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
- 27 6月, 2016 2 次提交
-
-
由 Lin Ma 提交于
First, The original relative path of floppy images are not exist. There is no 'data/images/' folders under 'shared' folder by default, It causes 'dd' command failures. If adding code into 'pre_command' of readonly_floppy.cfg to check/create 'shared/data/images/', Then the second issue jumps out: Because 'process_command' function in env_process.py will perform ... "cd %s; %s" % (test.bindir, pre_command) ... The test.bindir's value is assigned to '/usr/share/avocado-plugins-vt', So we got 2 fl images under /usr/share/avocado-plugins-vt/shared/data/images/. But in the final qemu command line parameters which generating by avocado, The path of floppy image is /usr/share/avocado/data/avocado-vt/images/... So I think that we insert 'cd /usr/share/avocado' to pre_command for ignoring the effect of 'cd /usr/share/avocado-plugins-vt' and change the original relative path info from 'shared/data/images/' to 'data/avocado-vt/images/' may be a better choice. Signed-off-by: NLin Ma <lma@suse.com>
-
由 Xu Tian 提交于
1. only online offline state memory 2. check memory after hotplug/unplug memory Signed-off-by: NXu Tian <xutian@redhat.com>
-
- 24 6月, 2016 3 次提交
-
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
由 Cong Li 提交于
1. vcpu_need_hotplug for max_cpu scenario can not be a fixed num, should be vcpu_need_hotplug = vcpu_maxcpus - vm.cpuinfo.smp 2. Rename param "id" to "cpuid", as there is same param in the object. 3. Enlarge waiting time after hotplug cpus, 60s is not enough for max cpu hotplug. Signed-off-by: NCong Li <coli@redhat.com>
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
- 22 6月, 2016 1 次提交
-
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
- 21 6月, 2016 2 次提交
-
-
由 Wei Jiangang 提交于
The variable is decimal, not percentage No need to be followed by percent sign, Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
由 Wei Jiangang 提交于
delete spaces between number and "%" Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
- 16 6月, 2016 3 次提交
-
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
由 Wei Jiangang 提交于
create_monitor() has chosen the proper monitor, We can use vm.monitor.protocol to distinguish between qmp and hmp. Not need to call qemu_has_option any more. Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
由 Wei Jiangang 提交于
If the param 'qemu_path' is default, The utils_misc.qemu_has_option() takes qemu-kvm as test object, Not the qemu specified by user. It's definition as follow, def qemu_has_option(option, qemu_path="/usr/bin/qemu-kvm"): .... so call get_qemu_binary() to get the right test object(qemu) before calling qemu_has_option(). Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
- 15 6月, 2016 1 次提交
-
-
由 Ping Li 提交于
As upstream commit 5be5b7764f83cf9a535a22ecbd33710daf1fe210 has been merged in the release qemu-kvm-rhev-2.6.0-1.el7, BlkdebugEvent are named with an underscore instead of a period. Update the related events according to new naming rules. Signed-off-by: Ping Li pingl@redhat.com
-
- 14 6月, 2016 3 次提交
-
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
由 vivianQizhu 提交于
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
- 13 6月, 2016 1 次提交
-
-
由 Xueqiang Wei 提交于
rhel7.3 guest report error message likes "Tray of device ... is not open" when 'change' command be executed in first time. Signed-off-by: NXueqiang Wei <xuwei@redhat.com>
-
- 12 6月, 2016 1 次提交
-
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
- 08 6月, 2016 1 次提交
-
-
由 Wei Jiangang 提交于
replace '%' with '%s' and remove unused variable. Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
- 07 6月, 2016 1 次提交
-
-
由 Suqin Huang 提交于
Add scenerios for virstor/vioscsi test e.g run iozone in system disk/data disk before/during/after sub tests (shutdown, reboot)
-
- 06 6月, 2016 1 次提交
-
-
由 Ping Li 提交于
Correct the CmdError exception to get stderr, which is usefull for the caller to parse and search specific messages. Signed-off-by: NPing Li <pingl@redhat.com>
-
- 03 6月, 2016 1 次提交
-
-
由 Wei Jiangang 提交于
s/fail_falg/fail_flag Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
- 01 6月, 2016 1 次提交
-
-
由 Cong Li 提交于
In serial sessions, frequently the kernel might print debug or error messages that make read_up_to_prompt to timeout. Let's try to be a little more robust and send a carriage return, to see if we can get to the prompt when safe=True. Signed-off-by: NCong Li <coli@redhat.com>
-
- 31 5月, 2016 1 次提交
-
-
由 Lin Ma 提交于
Signed-off-by: NLin Ma <lma@suse.com>
-
- 30 5月, 2016 3 次提交
-
-
由 Cong Li 提交于
1. force create image for block testing. 2. ignore errors in get_installation_logs in finally part, which will blocks real failed reason. 3. no need compare current driver version as expected if driver is uninstalled. Signed-off-by: NCong Li <coli@redhat.com>
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
由 Wei Jiangang 提交于
For example, 'session.sendline(params.get("shutdown_command"))' was discarded, "Shutdown command is sent, ..." is inapproprivate. Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
- 28 5月, 2016 1 次提交
-
-
由 Cong Li 提交于
Signed-off-by: NCong Li <coli@redhat.com>
-
- 27 5月, 2016 1 次提交
-
-
由 qzhang 提交于
-
- 25 5月, 2016 1 次提交
-
-
由 Wei Jiangang 提交于
utils_test.wait_for_login had been replaced with vm.wait_for_login You can refer to virt-test/feea4fd. Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
- 23 5月, 2016 1 次提交
-
-
由 Suqin Huang 提交于
suppress_exception always be True currently Signed-off-by: NSuqin Huang <shuang@redhat.com>
-
- 20 5月, 2016 1 次提交
-
-
由 Zhengtong 提交于
-
- 18 5月, 2016 3 次提交
-
-
由 Suqin Huang 提交于
Signed-off-by: NSuqin Huang <shuang@redhat.com>
-
由 Wei Jiangang 提交于
Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
由 Wei Jiangang 提交于
The dbd70861(2. add qtree info when test failed) introduced NameError about 'vm'. The reason is that no vm is passed to get_match_cdrom(), And the order of (qtree_info, serial_num) is wrong. Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-
- 17 5月, 2016 2 次提交
-
-
由 Zhengtong 提交于
-
由 Wei Jiangang 提交于
1) correct the comment and code format 2) remove unused variable 3) avoid call params.get() repeatly. Signed-off-by: NWei Jiangang <weijg.fnst@cn.fujitsu.com>
-