Update Failed:Too Many Levels of Symbolic Links

Products

• VMware Tanzu Kubernetes Grid Integrated Edition

Issue/Introduction

When upgrading or making changes to clusters, the instance update may fail due to sym links that were never cleaned up during log collection.

 

tkgi cluster <cluster-name>

Last Action Description:
Instance update failed: There was a problem completing your request. Please contact your operations team providing the following information: service: p.pks, servi
e-instance-guid: #############, broker-request-id: a71eb7a0-28C8-4066-88f4-b7157b1bd776, task-id: 3544044, operation: update, error-message: Action Failed get_ task: T sk bf88df2a-a90c-47bb-4436-d7f9accde976 result: Copying filtered files to temp directory: Copying Files to Temp Dir: Getting file info for /var/vcap/sys/Log/custom-var log/Log/Log/1og/1og/1
g/ 10g/log/ 1og/ log/ 1og/ 1og/ Log/ 1og/ Log/ 1og/Log/10g/ Log/ Log/1og/1og/1og/ Log/1og/ 1og/ 1og/Log/ Log/ Log/ 1og/Log/ 1og/ 1og/log/ 1og/log/ 1og/log/log': stat /var/vcap/sys/ 1og/custom-var log/1og/ 10g/ 10g/
og/Log/log/Log/log/log/Log/1og/1og/log/Log/Log/log/1og/1og/1og/Log/Log/log/Log/log/Log/og/log/Log/log/Log/log/1og/1og/Log/Log/Log/1og/log/log: too many levels of symbolic links

Cause

The customer may have applied the below KB article and forgot to remove the symlinks.

https://knowledge.broadcom.com/external/article/327455/how-to-manually-collect-logs-and-detaile.html

 

Resolution

# Important Step - Remove symlink for cleanup and executing the script next time
bosh ssh -d "${BOSH_LOG_DEPLOYMENT}" master "sudo rm /var/vcap/sys/log/custom-varlog"
bosh ssh -d "${BOSH_LOG_DEPLOYMENT}" worker "sudo rm /var/vcap/sys/log/custom-varlog"

# Important Step - Remove symlink for cleanup and executing the script next time bosh ssh -d "${BOSH_LOG_DEPLOYMENT}" pivotal-container-service "sudo rm /var/vcap/sys/log/custom-varlog" bosh ssh -d "${BOSH_LOG_DEPLOYMENT}" pks-db "sudo rm /var/vcap/sys/log/custom-varlog"