ThousandEyes Agent Install fails with "Duplicate mount point"

[twylyghtcisco]

Issue appears to be related to https://bst.cisco.com/quickview/bug/CSCvz84369

When I upgraded to the Thousand Eyes (TE) agent to 5.1.1 to install on my Catalyst 9K switches, a number of my docker/agent configurations were altered.

Among these alterations were the removal of our token (for registering to the Thousand Eyes portal) and removal of our local domain name for our hostname variable.

Along with these reversions to the docker runtime options, I also removed the following to remediate bug CSCvz84369:

--mount type=tmpfs,destination=/var/tmp/te-agent,tmpfs-size=340m

Please note that I needed to remove both the agent on the switch as well as the registration on the TE portal prior to re-install.

---

[balaji.bandi]

is this installed manually or using DNAC ?

Older version works as expected and new Version having issue when you upgrade ?

what is the IOS XE code running on that Cat 9K

how about if you have test device, remove complete agent and install fresh 5.1 ?

[twylyghtcisco]

The deployment displaying the issue was from the GUI (ERR_AH_1000).  The focus was fixing the GUI for our T2 team to be able to leverage Catalyst Center for deployment and maintenance going forward. 

CLI was producing a DB error for the App ID stating the interface was not configured.

TE agent was previously 5.0.1.  I upgraded it to 5.1.1 to address a separate issue, but was thought to also address the NTP syslog issue as well.  Agents were deployed recently using 5.0.1 and displaying the NTP issue, so this issue was being used for further troubleshooting in our lab.

IOS XE code is 17.12.4