Symptom
Background: After a physical machine is forcibly powered off and then powered on, the service container is not started. The direct cause is that a large number of files are damaged or lost. After the system is powered on, the fault persists.
The service container is not started.
Impact on Services
Service Containers Are Not Started After Power-on
Cause
The agent process of the ChinaSoft International solution is not started, and the corresponding installation program directory is lost. Hasen Data Loss
Identification Method
1. Log in to the CSP01 and CSP02 nodes as the paas user and switch to the root user.
2. Run the docker ps|grep mesos command. The command output is empty.
3. Run the docker ps|grep ICTS_CCN_MASTER command. In the command output, the container status is healthy.
4. Run the ps -ef|grep rm-agent|grep -v grep command. The rm-agent process does not exist. Check the /opt/csp/icts_ccn/software directory and find that the agent installation directory is lost.
Emergency Operations
Step 1 Delete the installation flag of the ChinaSoft International agent program. Otherwise, the installation script exits in advance.
cd /opt/csp/package/scripts/
rm icts_flag.log
Step 2 Run the bash software_agent_install.sh install yes command to execute the agent installation script. During the execution, the installation result logs of ChinaSoft International components are generated to ensure that the installation and startup are successful.
