The zrepl binary is self-documenting:
zrepl help for an overview of the available subcommands or
zrepl SUBCOMMAND --help for information on available flags, etc.
||show subcommand overview|
||run the daemon, required for all zrepl functionality|
||show job activity, or with
||see ssh+stdinserver Transport|
||manually trigger replication + pruning of JOB|
||manually abort current replication + pruning of JOB|
||check if config can be parsed without errors|
perform on-disk state / ZFS property migrations
(see changelog for details)
||list and remove zrepl’s abstractions on top of ZFS, e.g. holds and step bookmarks (see overview )|
All actual work zrepl does is performed by a daemon process. The daemon supports structured logging and provides monitoring endpoints.
When installing from a package, the package maintainer should have provided an init script / systemd.service file. You should thus be able to start zrepl daemon using your init system.
Alternatively, or for running zrepl in the foreground, simply execute
Note that you won’t see much output with the default logging configuration:
Make sure to actually monitor the error level output of zrepl: some configuration errors will not make the daemon exit.
Example: if the daemon cannot create the ssh+stdinserver Transport sockets in the runtime directory, it will emit an error message but not exit because other tasks such as periodic snapshots & pruning are of equal importance.
The daemon handles SIGINT and SIGTERM for graceful shutdown. Graceful shutdown means at worst that a job will not be rescheduled for the next interval. The daemon exits as soon as all jobs have reported shut down.
Systemd Unit File¶
A systemd service definition template is available in dist/systemd. Note that some of the options only work on recent versions of systemd. Any help & improvements are very welcome, see issue #145.
Along with the main
zrepl binary, we release the
The zrepl platform tests are an integration test suite that is complementary to the pure Go unit tests.
Any test that needs to interact with ZFS is a platform test.
The platform need to run as root. For each test, we create a fresh dummy zpool backed by a file-based vdev. The file path, and a root mountpoint for the dummy zpool, must be specified on the command line:
mkdir -p /tmp/zreplplatformtest ./platformtest \ -poolname 'zreplplatformtest' \ # <- name must contain zreplplatformtest -imagepath /tmp/zreplplatformtest.img \ # <- zrepl will create the file -mountpoint /tmp/zreplplatformtest # <- must exist
platformtest will unconditionally overwrite the file at imagepath
zpool destroy $poolname.
So, don’t use a production poolname, and consider running the test in a VM.
It’ll be a lot faster as well because the underlying operations,
zfs list in particular, will be faster.
While the platformtests are running, there will be a log of log output. After all tests have run, it prints a summary with a list of tests, grouped by result type (success, failure, skipped):
PASSING TESTS: github.com/zrepl/zrepl/platformtest/tests.BatchDestroy github.com/zrepl/zrepl/platformtest/tests.CreateReplicationCursor github.com/zrepl/zrepl/platformtest/tests.GetNonexistent github.com/zrepl/zrepl/platformtest/tests.HoldsWork ... github.com/zrepl/zrepl/platformtest/tests.SendStreamNonEOFReadErrorHandling github.com/zrepl/zrepl/platformtest/tests.UndestroyableSnapshotParsing SKIPPED TESTS: github.com/zrepl/zrepl/platformtest/tests.SendArgsValidationEncryptedSendOfUnencryptedDatasetForbidden__EncryptionSupported_false FAILED TESTS: 
If there is a failure, or a skipped test that you believe should be passing, re-run the test suite, capture stderr & stdout to a text file, and create an issue on GitHub.
To run a specific test case, or a subset of tests matched by regex, use the
-run REGEX command line flag.
To stop test execution at the first failing test, and prevent cleanup of the dummy zpool, use the
To build the platformtests yourself, use
There’s also the
make test-platform target to run the platform tests with a default command line.