Closed mikegorman-nf closed 1 year ago
Everything in the log excerpt looks normal to me.
The service (or its intercept configuration) was updated. The tSDK handles this as a remove/add, which explains the "service unavailable" and subsequent "ziti_dns_deregister_hostname()" logs.
Prior to the service update, the hostname "ebslwarncv29.ramco" was mapped to 100.64.0.93. This mapping was dropped when the "remove" step of the update occurred:
[2022-12-26T04:41:51.478Z] INFO tunnel-cbs:ziti_dns.c:375 ziti_dns_deregister_intercept() DNS mapping ebslwarncv29.ramco -> 100.64.0.93 is now inactive
Since that mapping was made, the dns server's internal counter that it uses for the next IP assignment was incremented a few times, so when the mapping is (re)made the IP is 100.64.0.153.
Note: the query for "ebslwarcnv29.ramco" at t=04:44:17.429Z and the associated record concerned me for a while, until I realized that it is a different hostname than the one in the service that was updated.
ebslwarncv29.ramco
ebslwarcnv29.ramco
Odd bug in a client. It looks to me like a new service was added that is a new port on the same address as other services. A new DNS entry was added, but the the old one remained, so the new service wasn't reachable because it wasn't attached properly.
44 min. ago