Fix CI fails on the "test_agent_over_ipv6" system test.
The test_agent_over_ipv6 system test unexpectedly started failing on the master and all MRs. It works properly on the local machine.
Logs:
server-1 | 2024-06-27T08:52:29.977147688Z time="2024-06-27 08:52:29" level="warning" msg="rpc error: code = Unavailable desc = connection error: desc = \"transport: Error while dialing: dial tcp [3000:db8:1:42::100]:8080: i/o timeout\"" file=" manager.go:124 " agent="[3000:db8:1:42::100]:8080"
server-1 | 2024-06-27T08:52:29.977203669Z time="2024-06-27 08:52:29" level="warning" msg="Failed to get state from the Stork agent; the agent is still not responding" file=" grpcli.go:329 " agent="[3000:db8:1:42::100]:8080"
server-1 | 2024-06-27T08:52:29.977210309Z time="2024-06-27 08:52:29" level="warning" msg="failed to get state from agent [3000:db8:1:42::100]:8080: grpc manager is unable to re-establish connection with the agent [3000:db8:1:42::100]:8080: rpc error: code = Unavailable desc = connection error: desc = \"transport: Error while dialing: dial tcp [3000:db8:1:42::100]:8080: i/o timeout\"" file=" statepuller.go:247 "It is an error thrown when the Stork server tries to make a GRPC call to the Stork agent to fetch the agent state. This single test makes the call over the IPv6 protocol. All other tests do it over IPv4, and they work properly.
The first failed job:
Other failed jobs:
Edited by Slawek Figiel