Shulou(Shulou.com)06/02 Report--

The server name is wh01 wh02 wh03.

Engine and all nodes set SELINUX=permissive firewall state: enable ssh mutual trust

Select the isolation policy for the cluster:

If the status of Gluster brick is UP, ignore the isolation operation

If the Gluster quorum is not satisfied, ignore the isolation operation

Properties of the virtual machine:

The system is a win7 memory 8192 4G solid state 30G system disk 10G hdd data disk

Migration mode: running automatic and manual migration

Enable highly available options to restore behavioral auto_resume watchdog i6300esb power outage

After the test results of the Win7 where the test content engine is located, the wh03 network cable Wh02Wh0330s is disconnected, it is unable to ping the engine. After 3 minutes and 15 seconds, the engine and platform returned to normal, the win7 was in an unknown state, and the wh03 did not respond.

After connecting the wh03 network cable Wh02Wh03win7 to the up,5 molecule, the wh03 returns to normal, but the terminal needs to restart to use the disconnected wh01 network cable Wh02Wh0320s, the ping engine cannot be used, and the virtual machine of the terminal becomes stuttered for a few seconds, and then it can be accessed normally. The engine and platform returned to normal after 6 minutes and 45 seconds. When the engine floats from wh02 to wh03, the terminal virtual machine crashes again and needs to be restarted again. Wh02's brick is also down, and the ovirt-ha-agent service is down, causing the Wh02 engine to float to wh03, and the recovery time is also longer. As for wh02's brick, why did it hang up? The wh01 resumes after connecting the wh01 network cable Wh03Wh0320 for many seconds, while the wh02 is unable to work, which may be related to the loss of the previous brick down, and the wh02 resumes after 6 minutes. And the brick down and wh01 brick of wh02 down for one time up forcibly starts all brick,brick in the volume to return to normal. However, it was found that wh01 could not work, and after manual activation, it returned to normal successfully. When the wh03 cable is disconnected, the Wh03Wh03 cannot ping the engine. After 2 minutes and 10 seconds, the engine resumes and the platform is normal. There is a delay in the host mentality and brick status of the win7 unknown Wh03. After a second delay in connecting to the wh03 cable, wh03 starts up, and wh03's brick goes through two restarts. The virtual machine state is paused. Gluster volume status brick has been normally disconnected wh02 network cable Wh03Wh01 engine and platform access is normal. Win7 is not responding. As expected, win7 should not be affected after wh02Wh03Wh01Wh02 fails to work, Win7 status still does not respond, win7 shuts down, shutdown fails, node2 status is not assigned, restart wh02,win7 closes successfully, wh02 returns to normal after 10 molecules, Wh01 accidentally restarts Wh03Wh02 engine and platform, terminal virtual machine session is not interrupted, the system can be used normally by virtual machine, it is not in the test range.

Network disconnection refers to unplugging both management and storage network cables, and there is no test to disconnect any of them separately. Connect the network cable is also connected to two, follow the last test content. And the test order is from top to bottom)

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.