Problem Symptom
After a cluster upgrade is complete, the next upgrade starts immediately. During the new upgrade, DeployManager fails to be upgraded. Displayed information:
[checkRemoteRepos][[Line:1816][INF0] Beggin to compare repos package in remote node with local size 143][checkRemoteRepos][[Line:1816][Error] Failed to check remote repos in remote node!][precheck][[Line:2186][Error] failed to check repo packages in remote nade]
Problem Diagnosis
Check whether [Error] Failed to check remote repos in remote node! is recorded in the log printed during the DeployManager upgrade.
Causes
Microservice packages in the software repository of the active and standby nodes are synchronized periodically using ha_client_tool. For a cluster that has just been upgraded, the upgrade package information in the software repository of the active node has not been synchronized to the standby node. During the pre-upgrade check for the DeployManager upgrade, the package information in the software repository of the active and standby nodes is inconsistent. As a result, DeployManager fails to be upgraded.
Solution
Run the /opt/dfv/oam/oam-u/ha/ha/module/hacom/tools/ha_client_tool --syncallfile command on the active node. ha_client_tool tool is triggered to synchronize files.
Run the find /opt/fusionstorage/repository/data/repos/ -name "*.tar.gz" -type f|wc -l command repeatedly on the active and standby nodes. If the numbers displayed in the command output are the same on the active and standby nodes, ha_client_tool has synchronized the software packages.
Run the sh upgrade.sh command on the original standby node to perform the upgrade again.
Check After Recovery
Check whether DeployManager is upgraded successfully.
Suggestion and Summary
N/A
Applicable Versions
All
