Data change counter not ready with tell-based protocol

This is seen in CSIT tests, relevant for Carbon-SR2 candidate validation.

After configuration, read of /restconf/operational/data-change-counter:data-change-counter returns [0] 404 with:
{"errors":{"error":[

{"error-type":"application","error-tag":"data-missing","error-message":"Request could not be completed because the relevant data model content does not exist "}

]}}

Note that the same suite is also running with ask-based protocol, when the read succeeds [1]. There is a hard reset between the suites.

I have been trying to reproduce this on Sandbox with a simplified testplan, both on all [2] and -only [3] job. But each run either passes, or fails on CONTROLLER-1645.

As on Releng, Verify_Data_Change_Counter_Ready test case if failing consistently [4], mu current hypothesis is that this failure is somehow caused by the previous suite, even though that suite occasionally passes [5].

More investigation is needed.

[0] https://logs.opendaylight.org/releng/jenkins092/bgpcep-csit-1node-periodic-bgp-ingest-all-carbon/418/log.html.gz#s1-s12-t6-k2-k1-k2
[1] https://logs.opendaylight.org/releng/jenkins092/bgpcep-csit-1node-periodic-bgp-ingest-all-carbon/418/log.html.gz#s1-s4-t6-k2-k1-k4-k1
[2] https://jenkins.opendaylight.org/sandbox/job/bgpcep-csit-1node-periodic-bgp-ingest-all-carbon/
[3] https://jenkins.opendaylight.org/sandbox/job/bgpcep-csit-1node-periodic-bgp-ingest-only-carbon/
[4] https://jenkins.opendaylight.org/releng/job/bgpcep-csit-1node-periodic-bgp-ingest-all-carbon/418/robot/bgpcep-bgp-ingest.txt/Singlepeer%20Changecount_1/Verify_Data_Change_Counter_Ready/
[5] https://jenkins.opendaylight.org/releng/job/bgpcep-csit-1node-periodic-bgp-ingest-all-carbon/418/robot/bgpcep-bgp-ingest.txt/Singlepeer%20Prefixcount_1/