How to Increase the Timeout on BOSH CPI Command Calls in Pivotal Cloud Foundry

Products

• VMware Tanzu Application Service for VMs

Issue/Introduction

A BOSH task process times out if it exceeds 60 minutes. This can occur through BOSH CLI or when applying changes in Operations Manager. The reason for this is the default tasks timeout is set to 60 minutes which causes a timeout similar to the following:

Failed updating job nfs_server-partition-d29a7e415aac7257b200 > nfs_server-partition-d29a7e415aac7257b200/0 (canary): Task exceeded 60 minutes, task properties: {:obj=>#<VimSdk::Vim::Task:0x0055daf7d7c030 @__mo_id__="task-49715",

Environment

Cause

BOSH CPI timeout is set to 60 minutes.

 

Resolution

Use the following procedure to increase the timeout as required:

1. SSH into the MicroBOSH Director.

2. Run the following command to find the client.rb file:

find /var/vcap/data/packages|grep 'vsphere/client.rb'

3. Edit the file:

vim vsphere/client.rb  

4. Change this line within the client.rb file: 

"Task exceeded 60 minutes, task properties: #{properties}" if duration > 3600 # 1 hour

To:

"Task exceeded 60 minutes, task properties: #{properties}" if duration > 21600 # 6 hours

5. Save the file and restart the directory by running the following command:

monit restart director

6. Run the following command to check if director restarted correctly:

watch monit summary

Once changes are made, continue with deployment.

Note: This issue has been fixed in 1.7+ https://github.com/cloudfoundry-incubator/bosh-vsphere-cpi-release/releases/tag/v20

Additional Information