Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[v24.1.x] tests: fixed querying new leader node in leadership transfer test #20580

Merged
merged 2 commits into from
Jun 28, 2024
Merged

[v24.1.x] tests: fixed querying new leader node in leadership transfer test #20580

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
1c236e4
tests: fixed querying new leader node in leadership transfer test
mmaslankaprv Jun 27, 2024
c7cbb68
tests: use get_node_by_id in raft availability test
mmaslankaprv Jun 27, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 20 additions & 17 deletions tests/rptest/tests/raft_availability_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,19 +187,20 @@ def test_one_node_down(self):
replication=3,
timeout_s=ELECTION_TIMEOUT * 2)

leader_node = self.redpanda.get_node(initial_leader_id)
leader_node = self.redpanda.get_node_by_id(initial_leader_id)
self.logger.info(
f"Initial leader {initial_leader_id} {leader_node.account.hostname}"
)
self.logger.info(f"id_allocator leader {allocator_info.leader}")

# Priority mechanism should reliably select next replica in list
expect_new_leader_id = replicas[1]
expect_new_leader_node = self.redpanda.get_node(expect_new_leader_id)
expect_new_leader_node = self.redpanda.get_node_by_id(
expect_new_leader_id)

observer_node_id = (set(replicas) -
{expect_new_leader_id, initial_leader_id}).pop()
observer_node = self.redpanda.get_node(observer_node_id)
observer_node = self.redpanda.get_node_by_id(observer_node_id)
self.logger.info(
f"Tracking stats on observer node {observer_node_id} {observer_node.account.hostname}"
)
Expand Down Expand Up @@ -276,9 +277,9 @@ def test_two_nodes_down(self):

self.ping_pong().ping_pong()

leader_node = self.redpanda.get_node(initial_leader_id)
leader_node = self.redpanda.get_node_by_id(initial_leader_id)
other_node_id = (set(replicas) - {initial_leader_id}).pop()
other_node = self.redpanda.get_node(other_node_id)
other_node = self.redpanda.get_node_by_id(other_node_id)

self.logger.info(
f"Stopping {initial_leader_id} ({leader_node.account.hostname}) and {other_node_id} ({other_node.account.hostname})"
Expand All @@ -290,7 +291,7 @@ def test_two_nodes_down(self):
self._expect_unavailable()

# Bring back one node (not the original leader)
self.redpanda.start_node(self.redpanda.get_node(other_node_id))
self.redpanda.start_node(self.redpanda.get_node_by_id(other_node_id))

hosts = [
n.account.hostname for n in self.redpanda.nodes
Expand Down Expand Up @@ -327,7 +328,7 @@ def test_leader_restart(self):
the original leader stopped.
"""
initial_leader_id, replicas = self._wait_for_leader()
initial_leader_node = self.redpanda.get_node(initial_leader_id)
initial_leader_node = self.redpanda.get_node_by_id(initial_leader_id)

self.logger.info(
f"Stopping initial leader {initial_leader_id} {initial_leader_node.account.hostname}"
Expand All @@ -337,7 +338,7 @@ def test_leader_restart(self):
new_leader_id, _ = self._wait_for_leader(
lambda l: l is not None and l != initial_leader_id)
self.logger.info(
f"New leader is {new_leader_id} {self.redpanda.get_node(new_leader_id).account.hostname}"
f"New leader is {new_leader_id} {self.redpanda.get_node_by_id(new_leader_id).account.hostname}"
)

self.logger.info(
Expand Down Expand Up @@ -369,7 +370,7 @@ def test_leadership_transfer(self):
continue serving requests.
"""
initial_leader_id, replicas = self._wait_for_leader()
initial_leader_node = self.redpanda.get_node(initial_leader_id)
initial_leader_node = self.redpanda.get_node_by_id(initial_leader_id)

metric_checks = {}
for n in self.redpanda.nodes:
Expand All @@ -387,10 +388,12 @@ def test_leadership_transfer(self):
leader_id=initial_leader_id)
new_leader_id, _ = self._wait_for_leader(
lambda l: l is not None and l != initial_leader_id)
new_leader_node = self.redpanda.get_node_by_id(new_leader_id)
assert new_leader_node is not None
self.logger.info(
f"New leader is {new_leader_id} {self.redpanda.get_node(new_leader_id).account.hostname}"
f"New leader is {new_leader_id} {new_leader_node.account.hostname}"
)
time.sleep(ELECTION_TIMEOUT)

for [id, metric_check] in metric_checks.items():
# the metric should be updated only on the node that was elected as a leader
if id == new_leader_id:
Expand Down Expand Up @@ -459,10 +462,10 @@ def test_leader_transfers_recovery(self, acks):
initial_leader_id = leader_node_id
for n in range(0, transfer_count):
target_idx = (initial_leader_id + n) % len(self.redpanda.nodes)
target_node_id = target_idx + 1
target_node_by_id_id = target_idx + 1

self._transfer_leadership(admin, "kafka", self.topic,
target_node_id)
target_node_by_id_id)

# Wait til we can see producer progressing, to avoid a situation where
# we do leadership transfers so quickly that we stall the producer
Expand Down Expand Up @@ -497,7 +500,7 @@ def test_follower_isolation(self):

self._expect_available()

leader_node = self.redpanda.get_node(initial_leader_id)
leader_node = self.redpanda.get_node_by_id(initial_leader_id)
self.logger.info(
f"Initial leader {initial_leader_id} {leader_node.account.hostname}"
)
Expand All @@ -523,7 +526,7 @@ def test_follower_isolation(self):
# isolate one of the followers
fi.inject_failure(
FailureSpec(FailureSpec.FAILURE_ISOLATE,
self.redpanda.get_node(follower)))
self.redpanda.get_node_by_id(follower)))

# expect messages to be produced and consumed without a timeout
connection = self.ping_pong()
Expand All @@ -545,7 +548,7 @@ def test_id_allocator_leader_isolation(self):
replication=3)
initial_leader_id = admin.get_partition_leader(
namespace='kafka_internal', topic='id_allocator', partition=0)
leader_node = self.redpanda.get_node(initial_leader_id)
leader_node = self.redpanda.get_node_by_id(initial_leader_id)
self.logger.info(
f"kafka_internal/id_allocator/0 leader: {initial_leader_id}, node: {leader_node.account.hostname}"
)
Expand All @@ -556,7 +559,7 @@ def test_id_allocator_leader_isolation(self):
# isolate id_allocator
fi.inject_failure(
FailureSpec(FailureSpec.FAILURE_ISOLATE,
self.redpanda.get_node(initial_leader_id)))
self.redpanda.get_node_by_id(initial_leader_id)))

# expect messages to be produced and consumed without a timeout
connection = self.ping_pong()
Expand Down
Loading