/storage/lifecycle/vmware-wcp/backup/archives directory on vCenter Server keeps filling up due to incomplete vSphere Supervisor Backups
search cancel

/storage/lifecycle/vmware-wcp/backup/archives directory on vCenter Server keeps filling up due to incomplete vSphere Supervisor Backups

book

Article ID: 421316

calendar_today

Updated On:

Products

VMware vSphere Kubernetes Service

Issue/Introduction

  • Disk space within the vCenter Server is being consumed rapidly as the /storage/lifecycle/vmware-wcp/backup/archives directory fills up. The directory contains multiple large backup files that are incomplete or abandoned. 
  • This accumulation coincides with failed vCenter Server backup jobs when the Supervisor Cluster is included in the backup configuration.
  • Analysis of the vCenter Server backup logs at /var/log/vmware/applmgmt/backup.log confirms that backup processes are failing intermittently:
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [Proc::GetProcsStatus:Proc.py:344] ERROR: Process returncode is -13, but expected exit codes are [0].
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [Proc::GetProcsStatus:Proc.py:328] ERROR: rc: 1, stderr: Traceback (most recent call last):
      File "/usr/lib/applmgmt/backup_restore/py/vmware/appliance/backup_restore/plugins/../util/Calculate.py", line 58, in <module>
        main(sys.argv[1], sys.argv[2], sys.argv[3])
      File "/usr/lib/applmgmt/backup_restore/py/vmware/appliance/backup_restore/plugins/../util/Calculate.py", line 45, in main
        stdout_obj.write(data)
    BrokenPipeError: [Errno 32] Broken pipe
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [Proc::GetProcsStatus:Proc.py:333] INFO: Skip to report the error.
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [Proc::GetProcsStatus:Proc.py:344] ERROR: Process returncode is 1, but expected exit codes are [0].
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [Proc::GetProcsStatus:Proc.py:361] ERROR: rc: 55, stderr: b'curl: (55) Send failure: Broken pipe\n'
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [Proc::UpdateExceptionStatus:Proc.py:384] ERROR: Checksum not generated at /dev/shm/backupRestoreSumFile-<File_Name>
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [Log::run:Log.py:64] ERROR: Failed to write backup content: [Errno 32] Broken pipeException ignored in: <_io.TextIOWrapp
    er name='<stdout>' mode='w' encoding='utf-8'>
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [Log::run:Log.py:64] ERROR: BrokenPipeError: [Errno 32] Broken pipe
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [ComponentScripts::ComponentScriptsBackup:ComponentScripts.py:106] ERROR: Component backup command "/etc/vmware/backup/component-scripts/wcp/supervisors_backup_restore.py --backup" failed 120.
    YYYY-MM-DDTHH:MM:SS [ComponentScriptsBackup:PID-######] [ComponentScripts::ComponentScriptsBackup:ComponentScripts.py:135] ERROR: Error during component supervisors backup

Environment

VMware vCenter Server

Cause

This issue is caused by the behavior of the Supervisor Cluster backup mechanism during a failure event. When the main vCenter Server backup process fails intermittently while the Supervisor Cluster is configured to be included, the failure triggers a fallback or an ad-hoc backup of the Supervisor Cluster which is the stored at the following directory: /storage/lifecycle/vmware-wcp/backup/archives

Resolution

To resolve the issue permanently, the root cause of the vCenter Server's backup failure needs to be determined and fixed. 
As a workaround, files within the /storage/lifecycle/vmware-wcp/backup/archives directory can be removed and purged by following the below steps:  

NOTE: Take a backup of the vCenter Server.

  • Open an SSH session to vCenter Server. 
  • Run the following commands: 
    • cd /storage/lifecycle/vmware-wcp/backup/archives
    • pwd (This is to ensure that present working directory is /storage/lifecycle/vmware-wcp/backup/archives)
    • rm -rf *
  • Run the following command to confirm if disk space is reduced: 
    • df -h

Additional Information

vCenter backup fails when the Supervisor is included.

vCenter backup fails when including Supervisor cluster

VAMI backups fail intermittently when Supervisor Control Plane is included.