docker: v18.09.9

[greut]

Pin Docker to v18.09.9. /docker/docker and /moby/moby are now pulled from docker/engine which is properly tagged.

The goal is to be able to pin the "latest" version, reading 19.03.8+, which will involve deeper changes which I'm not willing to make right now. Let's see how far can I go with this.

[greut]

Is it a flaky test? I get it to run just fine locally.

$ sudo -E PATH="$GOPATH/bin:/usr/local/go/bin:$PATH" go test ./nomad -v -run TestRPC_Lim
its_Streaming

=== RUN   TestRPC_Limits_Streaming
=== PAUSE TestRPC_Limits_Streaming
=== CONT  TestRPC_Limits_Streaming
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.372+0100 [DEBUG] docker/driver.go:1500: plugin_loader.docker: using client connection initialized from environment: plugin_dir=
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.372+0100 [DEBUG] docker/driver.go:1500: plugin_loader.docker: using client connection initialized from environment: plugin_dir=
[INFO] freeport: detected ephemeral port range of [32768, 60999]
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.378+0100 [INFO]  raft/api.go:549: nomad.raft: initial configuration: index=1 servers="[{Suffrage:Voter ID:127.0.0.1:9501 Address:127.0.0.1:9501}]"
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.378+0100 [INFO]  raft/raft.go:152: nomad.raft: entering follower state: follower="Node at 127.0.0.1:9501 [Follower]" leader=
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.378+0100 [INFO]  go-hclog/stdlog.go:46: nomad: serf: EventMemberJoin: nomad-001.global 127.0.0.1
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.378+0100 [INFO]  nomad/server.go:1398: nomad: starting scheduling worker(s): num_workers=8 schedulers=[service, batch, system, noop, _core]
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.379+0100 [INFO]  nomad/serf.go:60: nomad: adding server: server="nomad-001.global (Addr: 127.0.0.1:9501) (DC: dc1)"
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.474+0100 [WARN]  raft/raft.go:214: nomad.raft: heartbeat timeout reached, starting election: last-leader=
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.474+0100 [INFO]  raft/raft.go:250: nomad.raft: entering candidate state: node="Node at 127.0.0.1:9501 [Candidate]" term=2
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.474+0100 [DEBUG] raft/raft.go:268: nomad.raft: votes: needed=1
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.474+0100 [DEBUG] raft/raft.go:287: nomad.raft: vote granted: from=127.0.0.1:9501 term=2 tally=1
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.474+0100 [INFO]  raft/raft.go:292: nomad.raft: election won: tally=1
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.474+0100 [INFO]  raft/raft.go:363: nomad.raft: entering leader state: leader="Node at 127.0.0.1:9501 [Leader]"
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.474+0100 [INFO]  nomad/leader.go:71: nomad: cluster leadership acquired
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.475+0100 [TRACE] nomad/fsm.go:292: nomad.fsm: ClusterSetMetadata: cluster_id=d21d1d38-041e-3226-94d3-348943565dec create_time=1584265243475565516
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.475+0100 [INFO]  nomad/leader.go:1415: nomad.core: established cluster id: cluster_id=d21d1d38-041e-3226-94d3-348943565dec create_time=1584265243475565516
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.476+0100 [TRACE] drainer/watch_jobs.go:145: nomad.drain.job_watcher: getting job allocs at index: index=1
    TestRPC_Limits_Streaming: rpc_test.go:965: expect connection to be rejected due to limit
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:43.933+0100 [ERROR] nomad/rpc.go:314: nomad.rpc: rejecting client for exceeding maximum streaming RPC connections: remote_addr=127.0.0.1:35080 stream_limit=80
    TestRPC_Limits_Streaming: rpc_test.go:989: expect streaming connection 0 to exit with error
    TestRPC_Limits_Streaming: rpc_test.go:934: connection 0 died with error: (*net.OpError) read tcp 127.0.0.1:34920->127.0.0.1:9501: use of closed network connection
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:45.936+0100 [INFO]  nomad/server.go:594: nomad: shutting down server
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:45.936+0100 [WARN]  go-hclog/stdlog.go:48: nomad: serf: Shutdown without a Leave
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:45.937+0100 [TRACE] drainer/watch_jobs.go:147: nomad.drain.job_watcher: retrieved allocs for draining jobs: num_allocs=0 index=0 error="context canceled"
    TestRPC_Limits_Streaming: testlog.go:34: 2020-03-15T10:40:45.937+0100 [TRACE] drainer/watch_jobs.go:153: nomad.drain.job_watcher: shutting down
--- PASS: TestRPC_Limits_Streaming (3.57s)
PASS
ok      github.com/hashicorp/nomad/nomad        (cached)

[notnoop]

Wow - this is amazing! Thank you so much! Resolving the versions that work together isn't so straightforward!

And yes, that test is flaky now.

[greut]

@notnoop wooot! I've got a feeling that the jump to 19.03.x will be a bigger leap, but we've got to fix the gopsutil thingy.

[github-actions]

I'm going to lock this pull request because it has been closed for 120 days ⏳. This helps our maintainers find and focus on the active contributions.
If you have found a problem that seems related to this change, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.