Open aleksbykov opened 3 months ago
I think it is a test issue.
if tablets_enabled:
new_node.start(wait_other_notice=True)
stress_thread = write_in_background(node1, duration_seconds=5)
results = stress_thread.result()
assert_cs_success(results)
logger.debug(format_cs_output(results))
if tablets enabled node trying to be rebootstrapped, but data was not removed, and because prevously bootsrap was aborted, the node with ip and host id is banned by cluster and could not be boostrapped with same host id
@bhalevy ^^^
BTW, this is also happening on dtest-release, no? https://jenkins.scylladb.com/job/scylla-6.0/job/dtest-release/10/testReport/
if tablets enabled node trying to be rebootstrapped, but data was not removed, and because prevously bootsrap was aborted, the node with ip and host id is banned by cluster and could not be boostrapped with same host id
Yup. Thanks @aleksbykov.
Looks like the test was broken in scylladb/scylla-dtest@bb71da5bd. It also has weird history after that: scylladb/scylla-dtest@49aa179b27be3178a30594da2d507f0622c605da, scylladb/scylla-dtest@93468e62fe9f8e3a6f3946e84940c09b1c3111c5
@yarongilor @bhalevy has this test ever worked in tablets mode?
if tablets enabled node trying to be rebootstrapped, but data was not removed, and because prevously bootsrap was aborted, the node with ip and host id is banned by cluster and could not be boostrapped with same host id
Yup. Thanks @aleksbykov.
Looks like the test was broken in scylladb/scylla-dtest@bb71da5bd. It also has weird history after that: scylladb/scylla-dtest@49aa179, scylladb/scylla-dtest@93468e6
@yarongilor @bhalevy has this test ever worked in tablets mode?
@kbr-scylla , yes, it passes regularly, as in https://jenkins.scylladb.com/job/scylla-master/job/tablets/job/dtest-release-with-tablets/66/testReport/bootstrap_test/TestBootstrap/.
@pehala , @bhalevy - it's unclear to me if it's a test issue or not and what's the next step here.
Scylla version 6.0.3-0.20240808.a56f7ce21ad4 with build-id 66a8b676ee9e21374a3b46538c930879e592260f
Dtests which are failed:
Looks like node which rebootstrapped is stucked after
While node 4 was found error msg:
which could be expected, because operation was terminated