Closed tchaikov closed 10 months ago
being tested at https://github.com/scylladb/scylladb/pull/16724
@tchaikov If I am not mistaken the failing test is not running through CI. Did you verify it's passing now locally ?
@tchaikov If I am not mistaken the failing test is not running through CI. Did you verify it's passing now locally ?
sorry i forgot to mention, i verified the change also using
$ reloc/build_reloc.sh --version scylla-5.5.0~dev-0.20240111.8b02583eef5c --nodeps
$ ./bin/nodetool help
usage: nodetool [(-h <host> | --host <host>)]
[(-u <username> | --username <username>)]
[(-pw <password> | --password <password>)] [(-p <port> | --port <port>)]
[(-pwf <passwordFilePath> | --password-file <passwordFilePath>)] <command>
[<args>]
Submodule update: scylladb/scylladb@3d1667c720242eab1388a992a7fd5500b2c2b426
43427dfa introduced a regression, which moved guava v18 from the feet of io.airlift v0.6, which depends on guava 12.0, see https://mvnrepository.com/artifact/io.airlift/airline/0.6. actually, it is using an API -- com.google.common.base.Objects.firstNonNull, which was removed in guava 32.1.3, See https://guava.dev/releases/19.0/api/docs/com/google/common/base/Objects.html that's why we have runtime failure like:
when running nodetool.
fortunately, io.airlift/airline switched to another method in https://github.com/airlift/airline/commit/6e57ed2512fc3660e040b226bb1716b190fe3914, which was included by io.airlift 0.8 and up. Cassandra upstream is using io.airlift 0.8. that's why it's is also free from this problem even it has updated its guava to 32.0.1-jre in 992ad25b9608e59903dea4ec8becc00efbff5340.
so, in this change, we update io.airlift to 0.9, which is the latest release of this library.
Fixes #374 Signed-off-by: Kefu Chai kefu.chai@scylladb.com