btest/cluster/telemetry: Add smoke testing for telemetry

This commit is contained in:
Arne Welzel 2025-06-20 14:41:14 +02:00
parent b28e5f261e
commit 0e1431eef4
10 changed files with 347 additions and 2 deletions

View file

@ -0,0 +1,9 @@
### BTest baseline data generated by btest-diff. Do not edit. Use "btest -U/-u" to update. Requires BTest >= 0.63.
### NOTE: This file has been sorted with diff-sort.
zeek, zeek_cluster_core_incoming_events_total, [node], [manager], 2.0
zeek, zeek_cluster_core_outgoing_events_total, [node], [manager], 3.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [Cluster::Backend::ZeroMQ::hello, manager, zeek.cluster.nodeid.__normalized__], 1.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [Cluster::hello, manager, zeek.cluster.nodeid.__normalized__], 1.0
zeek, zeek_cluster_core_verbose_outgoing_events_total, [handler, node, topic], [Cluster::Backend::ZeroMQ::hello, manager, zeek.cluster.nodeid.__normalized__], 1.0
zeek, zeek_cluster_core_verbose_outgoing_events_total, [handler, node, topic], [Cluster::hello, manager, zeek.cluster.nodeid.__normalized__], 1.0
zeek, zeek_cluster_core_verbose_outgoing_events_total, [handler, node, topic], [finish, manager, zeek.cluster.nodeid.__normalized__], 1.0

View file

@ -0,0 +1,9 @@
### BTest baseline data generated by btest-diff. Do not edit. Use "btest -U/-u" to update. Requires BTest >= 0.63.
### NOTE: This file has been sorted with diff-sort.
zeek, zeek_cluster_core_incoming_events_total, [node], [worker-1], 3.0
zeek, zeek_cluster_core_outgoing_events_total, [node], [worker-1], 2.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [Cluster::Backend::ZeroMQ::hello, worker-1, zeek.cluster.nodeid.__normalized__], 1.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [Cluster::hello, worker-1, zeek.cluster.nodeid.__normalized__], 1.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [finish, worker-1, zeek.cluster.nodeid.__normalized__], 1.0
zeek, zeek_cluster_core_verbose_outgoing_events_total, [handler, node, topic], [Cluster::Backend::ZeroMQ::hello, worker-1, zeek.cluster.nodeid.__normalized__], 1.0
zeek, zeek_cluster_core_verbose_outgoing_events_total, [handler, node, topic], [Cluster::hello, worker-1, zeek.cluster.nodeid.__normalized__], 1.0

View file

@ -0,0 +1,3 @@
### BTest baseline data generated by btest-diff. Do not edit. Use "btest -U/-u" to update. Requires BTest >= 0.63.
### NOTE: This file has been sorted with diff-sort.
received termination signal

View file

@ -0,0 +1,33 @@
### BTest baseline data generated by btest-diff. Do not edit. Use "btest -U/-u" to update. Requires BTest >= 0.63.
### NOTE: This file has been sorted with diff-sort.
Cluster::websocket_client_added, [/test/pings, /zeek/wstest/ws1/]
zeek, zeek_cluster_core_debug_incoming_event_sizes, [handler, node, topic], [ping, manager, /test/pings/0], [0.0, 1.0, 0.0, 3.0, 4.0, 17.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_core_debug_incoming_event_sizes, [handler, node, topic], [ping, manager, /test/pings/1], [0.0, 0.0, 1.0, 3.0, 4.0, 17.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_core_debug_incoming_event_sizes, [handler, node, topic], [ping, manager, /test/pings/2], [0.0, 0.0, 0.0, 4.0, 3.0, 18.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_core_debug_incoming_event_sizes, [handler, node, topic], [ping, manager, /test/pings/3], [0.0, 0.0, 0.0, 3.0, 4.0, 18.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_core_debug_outgoing_event_sizes, [handler, node, script_location, topic], [ping, manager, ./manager.zeek:32, /test/pings/0], [0.0, 1.0, 0.0, 3.0, 4.0, 17.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_core_debug_outgoing_event_sizes, [handler, node, script_location, topic], [ping, manager, ./manager.zeek:32, /test/pings/2], [0.0, 0.0, 0.0, 4.0, 3.0, 18.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_core_incoming_events_total, [node], [manager], 100.0
zeek, zeek_cluster_core_outgoing_events_total, [node], [manager], 50.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [ping, manager, /test/pings/0], 25.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [ping, manager, /test/pings/1], 25.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [ping, manager, /test/pings/2], 25.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [ping, manager, /test/pings/3], 25.0
zeek, zeek_cluster_core_verbose_outgoing_events_total, [handler, node, topic], [ping, manager, /test/pings/0], 25.0
zeek, zeek_cluster_core_verbose_outgoing_events_total, [handler, node, topic], [ping, manager, /test/pings/2], 25.0
zeek, zeek_cluster_websocket_debug_incoming_event_sizes, [app, handler, node, topic], [btest-python-client, ping, manager, /test/pings/0], [0.0, 1.0, 0.0, 3.0, 4.0, 17.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_websocket_debug_incoming_event_sizes, [app, handler, node, topic], [btest-python-client, ping, manager, /test/pings/2], [0.0, 0.0, 0.0, 4.0, 3.0, 18.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_websocket_debug_outgoing_event_sizes, [app, handler, node, script_location, topic], [btest-python-client, ping, manager, none, /test/pings/0], [0.0, 1.0, 0.0, 3.0, 4.0, 17.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_websocket_debug_outgoing_event_sizes, [app, handler, node, script_location, topic], [btest-python-client, ping, manager, none, /test/pings/1], [0.0, 0.0, 1.0, 3.0, 4.0, 17.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_websocket_debug_outgoing_event_sizes, [app, handler, node, script_location, topic], [btest-python-client, ping, manager, none, /test/pings/2], [0.0, 0.0, 0.0, 4.0, 3.0, 18.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_websocket_debug_outgoing_event_sizes, [app, handler, node, script_location, topic], [btest-python-client, ping, manager, none, /test/pings/3], [0.0, 0.0, 0.0, 3.0, 4.0, 18.0, 0.0, 0.0, 0.0]
zeek, zeek_cluster_websocket_incoming_events_total, [app, node], [btest-python-client, manager], 50.0
zeek, zeek_cluster_websocket_outgoing_events_total, [app, node], [btest-python-client, manager], 100.0
zeek, zeek_cluster_websocket_verbose_incoming_events_total, [app, handler, node, topic], [btest-python-client, ping, manager, /test/pings/0], 25.0
zeek, zeek_cluster_websocket_verbose_incoming_events_total, [app, handler, node, topic], [btest-python-client, ping, manager, /test/pings/2], 25.0
zeek, zeek_cluster_websocket_verbose_outgoing_events_total, [app, handler, node, topic], [btest-python-client, ping, manager, /test/pings/0], 25.0
zeek, zeek_cluster_websocket_verbose_outgoing_events_total, [app, handler, node, topic], [btest-python-client, ping, manager, /test/pings/1], 25.0
zeek, zeek_cluster_websocket_verbose_outgoing_events_total, [app, handler, node, topic], [btest-python-client, ping, manager, /test/pings/2], 25.0
zeek, zeek_cluster_websocket_verbose_outgoing_events_total, [app, handler, node, topic], [btest-python-client, ping, manager, /test/pings/3], 25.0
zeek_cluster_* histogram metrics, 12
zeek_cluster_* metrics, 16

View file

@ -0,0 +1,3 @@
### BTest baseline data generated by btest-diff. Do not edit. Use "btest -U/-u" to update. Requires BTest >= 0.63.
### NOTE: This file has been sorted with diff-sort.
received termination signal

View file

@ -0,0 +1,18 @@
### BTest baseline data generated by btest-diff. Do not edit. Use "btest -U/-u" to update. Requires BTest >= 0.63.
### NOTE: This file has been sorted with diff-sort.
Cluster::websocket_client_added, [/test/pings, /zeek/wstest/ws1/]
zeek, zeek_cluster_core_incoming_events_total, [node], [manager], 100.0
zeek, zeek_cluster_core_outgoing_events_total, [node], [manager], 50.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [ping, manager, /test/pings/0], 25.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [ping, manager, /test/pings/1], 25.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [ping, manager, /test/pings/2], 25.0
zeek, zeek_cluster_core_verbose_incoming_events_total, [handler, node, topic], [ping, manager, /test/pings/3], 25.0
zeek, zeek_cluster_core_verbose_outgoing_events_total, [handler, node, topic], [ping, manager, /test/pings], 50.0
zeek, zeek_cluster_websocket_incoming_events_total, [app, node], [unknown, manager], 50.0
zeek, zeek_cluster_websocket_outgoing_events_total, [app, node], [unknown, manager], 100.0
zeek, zeek_cluster_websocket_verbose_incoming_events_total, [app, handler, node, topic], [unknown, ping, manager, /test/pings], 50.0
zeek, zeek_cluster_websocket_verbose_outgoing_events_total, [app, handler, node, topic], [unknown, ping, manager, /test/pings/0], 25.0
zeek, zeek_cluster_websocket_verbose_outgoing_events_total, [app, handler, node, topic], [unknown, ping, manager, /test/pings/1], 25.0
zeek, zeek_cluster_websocket_verbose_outgoing_events_total, [app, handler, node, topic], [unknown, ping, manager, /test/pings/2], 25.0
zeek, zeek_cluster_websocket_verbose_outgoing_events_total, [app, handler, node, topic], [unknown, ping, manager, /test/pings/3], 25.0
zeek_cluster_* metrics, 14

View file

@ -90,14 +90,18 @@ class TestClient:
return self.__name return self.__name
def connect(name: str, url: Optional[str] = None) -> TestClient: def connect(
name: str,
url: Optional[str] = None,
additional_headers: Optional[dict[str, str]] = None,
) -> TestClient:
""" """
Connect to a WebSocket server and return a TestClient instance. Connect to a WebSocket server and return a TestClient instance.
""" """
if url is None: if url is None:
url = WS4_URL_V1 url = WS4_URL_V1
cc = websockets.sync.client.connect(url) cc = websockets.sync.client.connect(url, additional_headers=additional_headers)
return TestClient(name, cc) return TestClient(name, cc)

View file

@ -0,0 +1,68 @@
# @TEST-DOC: All parties log their cluster metrics at zeek_done() time.
#
# @TEST-REQUIRES: have-zeromq
#
# @TEST-GROUP: cluster-zeromq
#
# @TEST-PORT: XPUB_PORT
# @TEST-PORT: XSUB_PORT
# @TEST-PORT: LOG_PULL_PORT
#
# @TEST-EXEC: cp $FILES/zeromq/cluster-layout-simple.zeek cluster-layout.zeek
# @TEST-EXEC: cp $FILES/zeromq/test-bootstrap.zeek zeromq-test-bootstrap.zeek
#
# @TEST-EXEC: btest-bg-run manager "ZEEKPATH=$ZEEKPATH:.. && CLUSTER_NODE=manager zeek -b ../manager.zeek >out"
# @TEST-EXEC: btest-bg-run worker "ZEEKPATH=$ZEEKPATH:.. && CLUSTER_NODE=worker-1 zeek -b ../worker.zeek >out"
#
# @TEST-EXEC: btest-bg-wait 30
# @TEST-EXEC: TEST_DIFF_CANONIFIER=$SCRIPTS/diff-sort btest-diff ./manager/out
# @TEST-EXEC: TEST_DIFF_CANONIFIER=$SCRIPTS/diff-sort btest-diff ./worker/out
# @TEST-START-FILE common.zeek
@load base/frameworks/telemetry
@load ./zeromq-test-bootstrap
redef Cluster::Telemetry::core_metrics += {
Cluster::Telemetry::VERBOSE,
};
redef Cluster::Telemetry::websocket_metrics += {
Cluster::Telemetry::VERBOSE,
};
global finish: event(name: string);
event zeek_done()
{
local ms = Telemetry::collect_metrics("zeek", "cluster_core_*");
ms += Telemetry::collect_metrics("zeek", "cluster_websocket_*");
for ( _, m in ms )
print m$opts$prefix, m$opts$name, m$label_names, m$label_values, m$value;
}
# @TEST-END-FILE
# @TEST-START-FILE manager.zeek
@load ./common.zeek
# If a node comes up that isn't us, send it a finish event.
event Cluster::node_up(name: string, id: string) {
Cluster::publish(Cluster::nodeid_topic(id), finish, Cluster::node);
}
# If the worker vanishes, finish the test.
event Cluster::node_down(name: string, id: string) {
terminate();
}
# @TEST-END-FILE
# @TEST-START-FILE worker.zeek
@load ./common.zeek
event Cluster::node_up(name: string, id: string) {
}
event finish(name: string) &is_used {
terminate();
}
# @TEST-END-FILE

View file

@ -0,0 +1,105 @@
# @TEST-DOC: Output cluster telemetry after working with a WebSocket client. The WebSocket client sends an X-Application-Name header. Also include debug metrics as histograms in the output.
#
# @TEST-REQUIRES: have-zeromq
# @TEST-REQUIRES: python3 -c 'import websockets.sync'
#
# @TEST-GROUP: cluster-zeromq
#
# @TEST-PORT: XPUB_PORT
# @TEST-PORT: XSUB_PORT
# @TEST-PORT: LOG_PULL_PORT
# @TEST-PORT: WEBSOCKET_PORT
#
# @TEST-EXEC: cp $FILES/zeromq/cluster-layout-simple.zeek cluster-layout.zeek
# @TEST-EXEC: cp $FILES/zeromq/test-bootstrap.zeek zeromq-test-bootstrap.zeek
# @TEST-EXEC: cp $FILES/ws/wstest.py .
#
# @TEST-EXEC: zeek -b --parse-only manager.zeek
# @TEST-EXEC: python3 -m py_compile client.py
#
# @TEST-EXEC: btest-bg-run manager "ZEEKPATH=$ZEEKPATH:.. && CLUSTER_NODE=manager zeek -b ../manager.zeek"
# @TEST-EXEC: python3 client.py
#
# @TEST-EXEC: btest-bg-wait 10
# @TEST-EXEC: TEST_DIFF_CANONIFIER=$SCRIPTS/diff-sort btest-diff ./manager/.stdout
# @TEST-EXEC: TEST_DIFF_CANONIFIER=$SCRIPTS/diff-sort btest-diff ./manager/.stderr
# @TEST-START-FILE manager.zeek
@load base/frameworks/telemetry
@load ./zeromq-test-bootstrap
redef Cluster::Telemetry::core_metrics += {
Cluster::Telemetry::VERBOSE,
Cluster::Telemetry::DEBUG,
};
redef Cluster::Telemetry::websocket_metrics += {
Cluster::Telemetry::VERBOSE,
Cluster::Telemetry::DEBUG,
};
redef exit_only_after_terminate = T;
global expected_ping_count = 100;
global ping_count = 0;
global ping: event(msg: string, c: count) &is_used;
event zeek_init()
{
Cluster::subscribe("/test/pings");
Cluster::listen_websocket([$listen_addr=127.0.0.1, $listen_port=to_port(getenv("WEBSOCKET_PORT"))]);
}
event ping(msg: string, n: count) &is_used
{
if ( ping_count % 2 == 0) # Reply every other ping.
{
Cluster::publish(fmt("/test/pings/%s", ping_count % 4), ping, msg, n);
}
++ping_count;
if ( ping_count == expected_ping_count )
terminate();
}
event Cluster::websocket_client_added(info: Cluster::EndpointInfo, subscriptions: string_vec)
{
print "Cluster::websocket_client_added", subscriptions;
}
event zeek_done()
{
local ms = Telemetry::collect_metrics("zeek", "cluster_core_*");
ms += Telemetry::collect_metrics("zeek", "cluster_websocket_*");
print "zeek_cluster_* metrics", |ms|;
for ( _, m in ms )
print m$opts$prefix, m$opts$name, m$label_names, m$label_values, m$value;
local hms = Telemetry::collect_histogram_metrics("zeek", "cluster_core_*");
hms += Telemetry::collect_histogram_metrics("zeek", "cluster_websocket_*");
print "zeek_cluster_* histogram metrics", |hms|;
for ( _, hm in hms )
print hm$opts$prefix, hm$opts$name, hm$label_names, hm$label_values, hm$values;
}
# @TEST-END-FILE
# @TEST-START-FILE client.py
import wstest
def run(ws_url):
with wstest.connect("ws1", ws_url, additional_headers={"X-Application-Name": "btest-python-client"}) as tc:
tc.hello_v1(["/test/pings"])
for i in range(0, 100):
msg = f"ping {i}" + (i * 32 * "A")
tc.send_json(wstest.build_event_v1(f"/test/pings/{i % 4}", "ping", [msg, i]))
if i % 2 == 0: # Wait for a reply for every other ping
tc.recv_json()
if __name__ == "__main__":
wstest.main(run, wstest.WS4_URL_V1)
# @TEST-END-FILE

View file

@ -0,0 +1,93 @@
# @TEST-DOC: Output cluster telemetry after working with a WebSocket client.
#
# @TEST-REQUIRES: have-zeromq
# @TEST-REQUIRES: python3 -c 'import websockets.sync'
#
# @TEST-GROUP: cluster-zeromq
#
# @TEST-PORT: XPUB_PORT
# @TEST-PORT: XSUB_PORT
# @TEST-PORT: LOG_PULL_PORT
# @TEST-PORT: WEBSOCKET_PORT
#
# @TEST-EXEC: cp $FILES/zeromq/cluster-layout-simple.zeek cluster-layout.zeek
# @TEST-EXEC: cp $FILES/zeromq/test-bootstrap.zeek zeromq-test-bootstrap.zeek
# @TEST-EXEC: cp $FILES/ws/wstest.py .
#
# @TEST-EXEC: zeek -b --parse-only manager.zeek
# @TEST-EXEC: python3 -m py_compile client.py
#
# @TEST-EXEC: btest-bg-run manager "ZEEKPATH=$ZEEKPATH:.. && CLUSTER_NODE=manager zeek -b ../manager.zeek"
# @TEST-EXEC: python3 client.py
#
# @TEST-EXEC: btest-bg-wait 10
# @TEST-EXEC: TEST_DIFF_CANONIFIER=$SCRIPTS/diff-sort btest-diff ./manager/.stdout
# @TEST-EXEC: TEST_DIFF_CANONIFIER=$SCRIPTS/diff-sort btest-diff ./manager/.stderr
# @TEST-START-FILE manager.zeek
@load base/frameworks/telemetry
@load ./zeromq-test-bootstrap
redef Cluster::Telemetry::core_metrics += {
Cluster::Telemetry::VERBOSE,
};
redef Cluster::Telemetry::websocket_metrics += {
Cluster::Telemetry::VERBOSE,
};
redef exit_only_after_terminate = T;
global expected_ping_count = 100;
global ping_count = 0;
global ping: event(msg: string, c: count) &is_used;
event zeek_init()
{
Cluster::subscribe("/test/pings");
Cluster::listen_websocket([$listen_addr=127.0.0.1, $listen_port=to_port(getenv("WEBSOCKET_PORT"))]);
}
event ping(msg: string, n: count) &is_used
{
if ( ping_count % 2 == 0) # Reply every other ping.
Cluster::publish("/test/pings", ping, msg, n);
++ping_count;
if ( ping_count == expected_ping_count )
terminate();
}
event Cluster::websocket_client_added(info: Cluster::EndpointInfo, subscriptions: string_vec)
{
print "Cluster::websocket_client_added", subscriptions;
}
event zeek_done()
{
local ms = Telemetry::collect_metrics("zeek", "cluster_core_*");
ms += Telemetry::collect_metrics("zeek", "cluster_websocket_*");
print "zeek_cluster_* metrics", |ms|;
for ( _, m in ms )
print m$opts$prefix, m$opts$name, m$label_names, m$label_values, m$value;
}
# @TEST-END-FILE
# @TEST-START-FILE client.py
import wstest
def run(ws_url):
with wstest.connect("ws1", ws_url) as tc:
tc.hello_v1(["/test/pings"])
for i in range(0, 100):
tc.send_json(wstest.build_event_v1(f"/test/pings/{i % 4}", "ping", [f"ping {i}", i]))
if i % 2 == 0: # Wait for a reply for every other ping
tc.recv_json()
if __name__ == "__main__":
wstest.main(run, wstest.WS4_URL_V1)
# @TEST-END-FILE