Coverage for bzfs_main / argparse_cli.py: 100%

1# Copyright 2024 Wolfgang Hoschek AT mac DOT com 

2# 

3# Licensed under the Apache License, Version 2.0 (the "License"); 

4# you may not use this file except in compliance with the License. 

5# You may obtain a copy of the License at 

6# 

7# http://www.apache.org/licenses/LICENSE-2.0 

8# 

9# Unless required by applicable law or agreed to in writing, software 

10# distributed under the License is distributed on an "AS IS" BASIS, 

11# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 

12# See the License for the specific language governing permissions and 

13# limitations under the License. 

14# 

15"""Documentation, definition of input data and ArgumentParser used by the 'bzfs' CLI.""" 

16 

17from __future__ import ( 

18 annotations, 

19) 

20import argparse 

21import dataclasses 

22import itertools 

23from typing import ( 

24 Final, 

25) 

26 

27from bzfs_main.argparse_actions import ( 

28 CheckPercentRange, 

29 DatasetPairsAction, 

30 DeleteDstSnapshotsExceptPlanAction, 

31 FileOrLiteralAction, 

32 IncludeSnapshotPlanAction, 

33 NewSnapshotFilterGroupAction, 

34 NonEmptyStringAction, 

35 SafeDirectoryNameAction, 

36 SafeFileNameAction, 

37 SSHConfigFileNameAction, 

38 TimeRangeAndRankRangeAction, 

39) 

40from bzfs_main.detect import ( 

41 DISABLE_PRG, 

42 DUMMY_DATASET, 

43) 

44from bzfs_main.period_anchors import ( 

45 PeriodAnchors, 

46) 

47from bzfs_main.util.check_range import ( 

48 CheckRange, 

49) 

50from bzfs_main.util.utils import ( 

51 ENV_VAR_PREFIX, 

52 PROG_NAME, 

53 format_dict, 

54) 

55 

56# constants: 

57__version__: Final[str] = "1.17.0.dev0" 

58PROG_AUTHOR: Final[str] = "Wolfgang Hoschek" 

59EXCLUDE_DATASET_REGEXES_DEFAULT: Final[str] = r"(?:.*/)?[Tt][Ee]?[Mm][Pp][-_]?[0-9]*" # skip tmp datasets by default 

60LOG_DIR_DEFAULT: Final[str] = PROG_NAME + "-logs" 

61SKIP_ON_ERROR_DEFAULT: Final[str] = "dataset" 

62CMP_CHOICES_ITEMS: Final[tuple[str, str, str]] = ("src", "dst", "all") 

63ZFS_RECV_O: Final[str] = "zfs_recv_o" 

64ZFS_RECV_X: Final[str] = "zfs_recv_x" 

65ZFS_RECV_GROUPS: Final[dict[str, str]] = {ZFS_RECV_O: "-o", ZFS_RECV_X: "-x", "zfs_set": ""} 

66ZFS_RECV_O_INCLUDE_REGEX_DEFAULT: Final[str] = "|".join( 

67 [ 

68 "aclinherit", 

69 "aclmode", 

70 "acltype", 

71 "atime", 

72 "checksum", 

73 "compression", 

74 "copies", 

75 "logbias", 

76 "primarycache", 

77 "recordsize", 

78 "redundant_metadata", 

79 "relatime", 

80 "secondarycache", 

81 "snapdir", 

82 "sync", 

83 "xattr", 

84 ] 

85) 

86 

87 

88def argument_parser() -> argparse.ArgumentParser: 

89 """Returns the CLI parser used by bzfs.""" 

90 create_src_snapshots_plan_example1: str = str({"test": {"": {"adhoc": 1}}}).replace(" ", "") 

91 create_src_snapshots_plan_example2: str = str({"prod": {"us-west-1": {"hourly": 36, "daily": 31}}}).replace(" ", "") 

92 delete_dst_snapshots_except_plan_example1: str = str( 

93 { 

94 "prod": { 

95 "onsite": { 

96 "secondly": 40, 

97 "minutely": 40, 

98 "hourly": 36, 

99 "daily": 31, 

100 "weekly": 12, 

101 "monthly": 18, 

102 "yearly": 5, 

103 } 

104 } 

105 } 

106 ).replace(" ", "") 

107 

108 # fmt: off 

109 parser: argparse.ArgumentParser = argparse.ArgumentParser( 

110 prog=PROG_NAME, 

111 allow_abbrev=False, 

112 formatter_class=argparse.RawTextHelpFormatter, 

113 description=f""" 

114*{PROG_NAME} is a reliable near real-time, parallel replication and backup command-line tool for ZFS. It replicates 

115snapshots from many local or remote source ZFS datasets (and their descendants) to local or remote destination 

116datasets, using zfs send/receive and ssh, and can operate at sub-second intervals across large fleets of hosts. 

117{PROG_NAME} incrementally replicates all ZFS snapshots since the most recent common snapshot, supporting disaster 

118recovery and high availability (DR/HA), scale-out deployments, and protection against data loss or ransomware.* 

119 

120When run for the first time, {PROG_NAME} replicates the dataset and all its snapshots from the source to the 

121destination. On subsequent runs, {PROG_NAME} transfers only the data that has changed since the previous run, 

122i.e. it incrementally replicates to the destination all intermediate snapshots that have been created on 

123the source since the last run. Source ZFS snapshots older than the most recent common snapshot found on the 

124destination are auto-skipped. 

125 

126Unless {PROG_NAME} is explicitly told to create snapshots on the source, it treats the source as read-only, 

127thus the source remains unmodified. With the --dryrun flag, {PROG_NAME} also treats the destination as read-only. 

128In normal operation, {PROG_NAME} treats the destination as append-only. Optional CLI flags are available to 

129delete destination snapshots and destination datasets as directed, for example to make the destination 

130identical to the source if the two have somehow diverged in unforeseen ways. This easily enables 

131(re)synchronizing the backup from the production state, as well as restoring the production state from 

132backup. 

133 

134In the spirit of rsync, {PROG_NAME} supports a variety of powerful include/exclude filters that can be combined to 

135select which datasets, snapshots and properties to create, replicate, delete or compare. 

136 

137Typically, a `cron` job on the source host runs `{PROG_NAME}` periodically to create new snapshots and prune outdated 

138snapshots on the source, whereas another `cron` job on the destination host runs `{PROG_NAME}` periodically to prune 

139outdated destination snapshots. Yet another `cron` job runs `{PROG_NAME}` periodically to replicate the recently created 

140snapshots from the source to the destination. The frequency of these periodic activities is typically every N milliseconds, 

141every second, minute, hour, day, week, month and/or year (or multiples thereof). 

142 

143All {PROG_NAME} functions including snapshot creation, replication, deletion, monitoring, comparison, etc. happily work 

144with any snapshots in any format, even created or managed by third party ZFS snapshot management tools, including manual 

145zfs snapshot/destroy. All functions can also be used independently. That is, if you wish you can use {PROG_NAME} just 

146for creating snapshots, or just for replicating, or just for deleting/pruning, or just for monitoring, or just for 

147comparing snapshot lists. 

148 

149The source 'pushes to' the destination whereas the destination 'pulls from' the source. {PROG_NAME} is installed 

150and executed on the 'initiator' host which can be either the host that contains the source dataset (push mode), 

151or the destination dataset (pull mode), or both datasets (local mode, no network required, no ssh required), 

152or any third-party (even non-ZFS OSX) host as long as that host is able to SSH (via standard 'ssh' OpenSSH CLI) into 

153both the source and destination host (pull-push mode). In pull-push mode the source 'zfs send's the data stream 

154to the initiator which immediately pipes the stream (without storing anything locally) to the destination 

155host that 'zfs receive's it. Pull-push mode means that {PROG_NAME} need not be installed or executed on either 

156source or destination host. Only the underlying 'zfs' CLI must be installed on both source and destination host. 

157{PROG_NAME} can run as root or non-root user, in the latter case via a) sudo or b) when granted corresponding 

158ZFS permissions by administrators via 'zfs allow' delegation mechanism. 

159 

160{PROG_NAME} is written in Python and continuously runs a wide set of unit tests and integration tests to ensure 

161coverage and compatibility with old and new versions of ZFS on Linux and FreeBSD, on all Python 

162versions ≥ 3.9 (including latest stable which is currently python-3.14). 

163 

164{PROG_NAME} is a stand-alone program with zero required dependencies, akin to a 

165stand-alone shell script or binary executable. It is designed to be able to run in restricted barebones server 

166environments. No external Python packages are required; indeed no Python package management at all is required. 

167You can just symlink the program wherever you like, for example into /usr/local/bin or similar, and simply run it like 

168any stand-alone shell script or binary executable. 

169 

170{PROG_NAME} automatically replicates the snapshots of multiple datasets in parallel for best performance. 

171Similarly, it quickly deletes (or monitors or compares) snapshots of multiple datasets in parallel. Atomic snapshots can be 

172created as frequently as every N milliseconds. 

173 

174Optionally, {PROG_NAME} applies bandwidth rate-limiting and progress monitoring (via 'pv' CLI) during 'zfs 

175send/receive' data transfers. When run across the network, {PROG_NAME} also transparently inserts lightweight 

176data compression (via 'zstd -1' CLI) and efficient data buffering (via 'mbuffer' CLI) into the pipeline 

177between network endpoints during 'zfs send/receive' network transfers. If one of these utilities is not 

178installed this is auto-detected, and the operation continues reliably without the corresponding auxiliary 

179feature. 

180 

181# Periodic Jobs with bzfs_jobrunner 

182 

183The software also ships with the [bzfs_jobrunner](README_bzfs_jobrunner.md) companion program, which is a convenience 

184wrapper around `{PROG_NAME}` that simplifies efficient periodic ZFS snapshot creation, replication, pruning, and monitoring, 

185across a fleet of N source hosts and M destination hosts, using a single shared fleet-wide 

186[jobconfig](bzfs_tests/bzfs_job_example.py) script. For example, this simplifies the deployment of an efficient 

187geo-replicated backup service where each of the M destination hosts is located in a separate geographic region and pulls 

188replicas from (the same set of) N source hosts. It also simplifies low latency replication from a primary to a secondary or 

189to M read replicas, or backup to removable drives, etc. 

190 

191# Quickstart 

192 

193* Create adhoc atomic snapshots without a schedule: 

194 

195```$ {PROG_NAME} tank1/foo/bar dummy --recursive --skip-replication --create-src-snapshots 

196--create-src-snapshots-plan "{create_src_snapshots_plan_example1}"``` 

197 

198```$ zfs list -t snapshot tank1/foo/bar 

199 

200tank1/foo/bar@test_2024-11-06_08:30:05_adhoc``` 

201 

202* Create periodic atomic snapshots on a schedule, every hour and every day, by launching this from a periodic `cron` job: 

203 

204```$ {PROG_NAME} tank1/foo/bar dummy --recursive --skip-replication --create-src-snapshots 

205--create-src-snapshots-plan "{create_src_snapshots_plan_example2}"``` 

206 

207```$ zfs list -t snapshot tank1/foo/bar 

208 

209tank1/foo/bar@prod_us-west-1_2024-11-06_08:30:05_daily 

210 

211tank1/foo/bar@prod_us-west-1_2024-11-06_08:30:05_hourly``` 

212 

213Note: A periodic snapshot is created if it is due per the schedule indicated by its suffix (e.g. `_daily` or `_hourly` 

214or `_minutely` or `_2secondly` or `_100millisecondly`), or if the --create-src-snapshots-even-if-not-due flag is specified, 

215or if the most recent scheduled snapshot is somehow missing. In the latter case {PROG_NAME} immediately creates a snapshot 

216(named with the current time, not backdated to the missed time), and then resumes the original schedule. If the suffix is 

217`_adhoc` or not a known period then a snapshot is considered non-periodic and is thus created immediately regardless of the 

218creation time of any existing snapshot. 

219 

220* Replication example in local mode (no network, no ssh), to replicate ZFS dataset tank1/foo/bar to tank2/boo/bar: 

221 

222```$ {PROG_NAME} tank1/foo/bar tank2/boo/bar``` 

223 

224```$ zfs list -t snapshot tank1/foo/bar 

225 

226tank1/foo/bar@prod_us-west-1_2024-11-06_08:30:05_daily 

227 

228tank1/foo/bar@prod_us-west-1_2024-11-06_08:30:05_hourly``` 

229 

230```$ zfs list -t snapshot tank2/boo/bar 

231 

232tank2/boo/bar@prod_us-west-1_2024-11-06_08:30:05_daily 

233 

234tank2/boo/bar@prod_us-west-1_2024-11-06_08:30:05_hourly``` 

235 

236* Same example in pull mode: 

237 

238```$ {PROG_NAME} root@host1.example.com:tank1/foo/bar tank2/boo/bar``` 

239 

240* Same example in push mode: 

241 

242```$ {PROG_NAME} tank1/foo/bar root@host2.example.com:tank2/boo/bar``` 

243 

244* Same example in pull-push mode: 

245 

246```$ {PROG_NAME} root@host1:tank1/foo/bar root@host2:tank2/boo/bar``` 

247 

248* Example in local mode (no network, no ssh) to recursively replicate ZFS dataset tank1/foo/bar and its descendant 

249datasets to tank2/boo/bar: 

250 

251```$ {PROG_NAME} tank1/foo/bar tank2/boo/bar --recursive``` 

252 

253```$ zfs list -t snapshot -r tank1/foo/bar 

254 

255tank1/foo/bar@prod_us-west-1_2024-11-06_08:30:05_daily 

256 

257tank1/foo/bar@prod_us-west-1_2024-11-06_08:30:05_hourly 

258 

259tank1/foo/bar/baz@prod_us-west-1_2024-11-06_08:40:00_daily 

260 

261tank1/foo/bar/baz@prod_us-west-1_2024-11-06_08:40:00_hourly``` 

262 

263```$ zfs list -t snapshot -r tank2/boo/bar 

264 

265tank2/boo/bar@prod_us-west-1_2024-11-06_08:30:05_daily 

266 

267tank2/boo/bar@prod_us-west-1_2024-11-06_08:30:05_hourly 

268 

269tank2/boo/bar/baz@prod_us-west-1_2024-11-06_08:40:00_daily 

270 

271tank2/boo/bar/baz@prod_us-west-1_2024-11-06_08:40:00_hourly``` 

272 

273* Example that makes destination identical to source even if the two have drastically diverged: 

274 

275```$ {PROG_NAME} tank1/foo/bar tank2/boo/bar --recursive --force --delete-dst-datasets --delete-dst-snapshots``` 

276 

277* Replicate all daily snapshots created during the last 7 days, and at the same time ensure that the latest 7 daily 

278snapshots (per dataset) are replicated regardless of creation time: 

279 

280```$ {PROG_NAME} tank1/foo/bar tank2/boo/bar --recursive --include-snapshot-regex '.*_daily' 

281--include-snapshot-times-and-ranks '7 days ago..anytime' 'latest 7'``` 

282 

283Note: The example above compares the specified times against the standard ZFS 'creation' time property of the snapshots 

284(which is a UTC Unix time in integer seconds), rather than against a timestamp that may be part of the snapshot name. 

285 

286* Delete all daily snapshots older than 7 days, but ensure that the latest 7 daily snapshots (per dataset) are retained 

287regardless of creation time: 

288 

289```$ {PROG_NAME} {DUMMY_DATASET} tank2/boo/bar --dryrun --recursive --skip-replication --delete-dst-snapshots 

290--include-snapshot-regex '.*_daily' --include-snapshot-times-and-ranks notime 'all except latest 7' 

291--include-snapshot-times-and-ranks 'anytime..7 days ago'``` 

292 

293Note: This also prints how many GB of disk space in total would be freed if the command were to be run for real without 

294the --dryrun flag. 

295 

296* Delete all daily snapshots older than 7 days, but ensure that the latest 7 daily snapshots (per dataset) are retained 

297regardless of creation time. Additionally, only delete a snapshot if no corresponding snapshot or bookmark exists in 

298the source dataset (same as above except replace the 'dummy' source with 'tank1/foo/bar'): 

299 

300```$ {PROG_NAME} tank1/foo/bar tank2/boo/bar --dryrun --recursive --skip-replication --delete-dst-snapshots 

301--include-snapshot-regex '.*_daily' --include-snapshot-times-and-ranks notime 'all except latest 7' 

302--include-snapshot-times-and-ranks '7 days ago..anytime'``` 

303 

304* Delete all daily snapshots older than 7 days, but ensure that the latest 7 daily snapshots (per dataset) are retained 

305regardless of creation time. Additionally, only delete a snapshot if no corresponding snapshot exists in the source 

306dataset (same as above except append 'no-crosscheck'): 

307 

308```$ {PROG_NAME} tank1/foo/bar tank2/boo/bar --dryrun --recursive --skip-replication --delete-dst-snapshots 

309--include-snapshot-regex '.*_daily' --include-snapshot-times-and-ranks notime 'all except latest 7' 

310--include-snapshot-times-and-ranks 'anytime..7 days ago' --delete-dst-snapshots-no-crosscheck``` 

311 

312* Delete all daily bookmarks older than 90 days, but retain the latest 200 daily bookmarks (per dataset) regardless 

313of creation time: 

314 

315```$ {PROG_NAME} {DUMMY_DATASET} tank1/foo/bar --dryrun --recursive --skip-replication --delete-dst-snapshots=bookmarks 

316--include-snapshot-regex '.*_daily' --include-snapshot-times-and-ranks notime 'all except latest 200' 

317--include-snapshot-times-and-ranks 'anytime..90 days ago'``` 

318 

319* Delete all tmp datasets within tank2/boo/bar: 

320 

321```$ {PROG_NAME} {DUMMY_DATASET} tank2/boo/bar --dryrun --recursive --skip-replication --delete-dst-datasets 

322--include-dataset-regex '(.*/)?tmp.*' --exclude-dataset-regex '!.*'``` 

323 

324* Retain all secondly snapshots that were created less than 40 seconds ago, and ensure that the latest 40 

325secondly snapshots (per dataset) are retained regardless of creation time. Same for 40 minutely snapshots, 36 hourly 

326snapshots, 31 daily snapshots, 12 weekly snapshots, 18 monthly snapshots, and 5 yearly snapshots: 

327 

328```$ {PROG_NAME} {DUMMY_DATASET} tank2/boo/bar --dryrun --recursive --skip-replication --delete-dst-snapshots 

329--delete-dst-snapshots-except 

330--include-snapshot-regex '.*_secondly' --include-snapshot-times-and-ranks '40 seconds ago..anytime' 'latest 40' 

331--new-snapshot-filter-group 

332--include-snapshot-regex '.*_minutely' --include-snapshot-times-and-ranks '40 minutes ago..anytime' 'latest 40' 

333--new-snapshot-filter-group 

334--include-snapshot-regex '.*_hourly' --include-snapshot-times-and-ranks '36 hours ago..anytime' 'latest 36' 

335--new-snapshot-filter-group 

336--include-snapshot-regex '.*_daily' --include-snapshot-times-and-ranks '31 days ago..anytime' 'latest 31' 

337--new-snapshot-filter-group 

338--include-snapshot-regex '.*_weekly' --include-snapshot-times-and-ranks '12 weeks ago..anytime' 'latest 12' 

339--new-snapshot-filter-group 

340--include-snapshot-regex '.*_monthly' --include-snapshot-times-and-ranks '18 months ago..anytime' 'latest 18' 

341--new-snapshot-filter-group 

342--include-snapshot-regex '.*_yearly' --include-snapshot-times-and-ranks '5 years ago..anytime' 'latest 5'``` 

343 

344For convenience, the lengthy command line above can be expressed in a more concise way, like so: 

345 

346```$ {PROG_NAME} {DUMMY_DATASET} tank2/boo/bar --dryrun --recursive --skip-replication --delete-dst-snapshots 

347--delete-dst-snapshots-except-plan "{delete_dst_snapshots_except_plan_example1}"``` 

348 

349* Compare source and destination dataset trees recursively, for example to check if all recently taken snapshots have 

350been successfully replicated by a periodic job. List snapshots only contained in src (tagged with 'src'), 

351only contained in dst (tagged with 'dst'), and contained in both src and dst (tagged with 'all'), restricted to hourly 

352and daily snapshots taken within the last 7 days, excluding the last 4 hours (to allow for some slack/stragglers), 

353excluding temporary datasets: 

354 

355```$ {PROG_NAME} tank1/foo/bar tank2/boo/bar --skip-replication --compare-snapshot-lists=src+dst+all --recursive 

356--include-snapshot-regex '.*_(hourly|daily)' --include-snapshot-times-and-ranks '7 days ago..4 hours ago' 

357--exclude-dataset-regex '(.*/)?tmp.*'``` 

358 

359If the resulting TSV output file contains zero lines starting with the prefix 'src' and zero lines starting with the 

360prefix 'dst' then no source snapshots are missing on the destination, and no destination snapshots are missing 

361on the source, indicating that the periodic replication and pruning jobs perform as expected. The TSV output is sorted 

362by dataset, and by ZFS creation time within each dataset - the first and last line prefixed with 'all' contains the 

363metadata of the oldest and latest common snapshot, respectively. The --compare-snapshot-lists option also directly 

364logs various summary stats, such as the metadata of the latest common snapshot, latest snapshots and oldest snapshots, 

365as well as the time diff between the latest common snapshot and latest snapshot only in src (and only in dst), as well 

366as how many src snapshots and how many GB of data are missing on dst, etc. 

367 

368* Example with further options: 

369 

370```$ {PROG_NAME} tank1/foo/bar root@host2.example.com:tank2/boo/bar --recursive 

371--exclude-snapshot-regex '.*_(secondly|minutely)' --exclude-snapshot-regex 'test_.*' 

372--include-snapshot-times-and-ranks '7 days ago..anytime' 'latest 7' --exclude-dataset /tank1/foo/bar/temporary 

373--exclude-dataset /tank1/foo/bar/baz/trash --exclude-dataset-regex '(.*/)?private' 

374--exclude-dataset-regex '(.*/)?[Tt][Ee]?[Mm][Pp][-_]?[0-9]*'``` 

375""") # noqa: S608 

376 

377 parser.add_argument( 

378 "--no-argument-file", action="store_true", 

379 # help="Disable support for reading the names of datasets and snapshots from a file.\n\n") 

380 help=argparse.SUPPRESS) 

381 parser.add_argument( 

382 "root_dataset_pairs", nargs="+", action=DatasetPairsAction, metavar="SRC_DATASET DST_DATASET", 

383 help="SRC_DATASET: " 

384 "Source ZFS dataset (and its descendants) that will be replicated. Can be a ZFS filesystem or ZFS volume. " 

385 "Format is [[user@]host:]dataset. The host name can also be an IPv4 address (or an IPv6 address where " 

386 "each ':' colon character must be replaced with a '|' pipe character for disambiguation). If the " 

387 "host name is '-', the dataset will be on the local host, and the corresponding SSH leg will be omitted. " 

388 "The same is true if the host is omitted and the dataset does not contain a ':' colon at the same time. " 

389 "Local dataset examples: `tank1/foo/bar`, `tank1`, `-:tank1/foo/bar:baz:boo` " 

390 "Remote dataset examples: `host:tank1/foo/bar`, `host.example.com:tank1/foo/bar`, " 

391 "`root@host:tank`, `root@host.example.com:tank1/foo/bar`, `user@127.0.0.1:tank1/foo/bar:baz:boo`, " 

392 "`user@||1:tank1/foo/bar:baz:boo`. " 

393 "The first component of the ZFS dataset name is the ZFS pool name, here `tank1`. " 

394 "If the option starts with a `+` prefix then dataset names are read from the UTF-8 text file given " 

395 "after the `+` prefix, with each line in the file containing a SRC_DATASET and a DST_DATASET, " 

396 "separated by a tab character. The basename must contain the substring 'bzfs_argument_file'. " 

397 "Example: `+root_dataset_names_bzfs_argument_file.txt`, " 

398 "`+/path/to/root_dataset_names_bzfs_argument_file.txt`\n\n" 

399 "DST_DATASET: " 

400 "Destination ZFS dataset for replication and deletion. Has same naming format as SRC_DATASET. During " 

401 "replication, destination datasets that do not yet exist are created as necessary, along with their " 

402 "parent and ancestors.\n\n" 

403 f"*Performance Note:* {PROG_NAME} automatically replicates multiple datasets in parallel. It replicates " 

404 "snapshots in parallel across datasets and serially within a dataset. All child datasets of a dataset " 

405 "may be processed in parallel. For consistency, processing of a dataset only starts after processing of " 

406 "all its ancestor datasets has completed. Further, when a thread is ready to start processing another " 

407 "dataset, it chooses the next dataset wrt. lexicographical sort order from the datasets that are " 

408 "currently available for start of processing. Initially, only the roots of the selected dataset subtrees " 

409 "are available for start of processing. The degree of parallelism is configurable with the --threads " 

410 "option (see below).\n\n") 

411 parser.add_argument( 

412 "--recursive", "-r", action="store_true", 

413 help="During snapshot creation, replication, deletion and comparison, also consider descendant datasets, i.e. " 

414 "datasets within the dataset tree, including children, and children of children, etc.\n\n") 

415 parser.add_argument( 

416 "--include-dataset", action=FileOrLiteralAction, nargs="+", default=[], metavar="DATASET", 

417 help="During snapshot creation, replication, deletion and comparison, select any ZFS dataset (and its descendants) " 

418 "that is contained within SRC_DATASET (DST_DATASET in case of deletion) if its dataset name is one of the " 

419 "given include dataset names but none of the exclude dataset names. If a dataset is excluded its descendants " 

420 "are automatically excluded too, and this decision is never reconsidered even for the descendants because " 

421 "exclude takes precedence over include.\n\n" 

422 "A dataset name is absolute if the specified dataset is prefixed by `/`, e.g. `/tank/baz/tmp`. " 

423 "Otherwise the dataset name is relative wrt. source and destination, e.g. `baz/tmp` if the source " 

424 "is `tank`.\n\n" 

425 "This option is automatically translated to an --include-dataset-regex (see below) and can be " 

426 "specified multiple times.\n\n" 

427 "If the option starts with a `+` prefix then dataset names are read from the newline-separated " 

428 "UTF-8 text file given after the `+` prefix, one dataset per line inside of the text file. The basename " 

429 "must contain the substring 'bzfs_argument_file'.\n\n" 

430 "Examples: `/tank/baz/tmp` (absolute), `baz/tmp` (relative), " 

431 "`+dataset_names_bzfs_argument_file.txt`, `+/path/to/dataset_names_bzfs_argument_file.txt`\n\n") 

432 parser.add_argument( 

433 "--exclude-dataset", action=FileOrLiteralAction, nargs="+", default=[], metavar="DATASET", 

434 help="Same syntax as --include-dataset (see above) except that the option is automatically translated to an " 

435 "--exclude-dataset-regex (see below).\n\n") 

436 parser.add_argument( 

437 "--include-dataset-regex", action=FileOrLiteralAction, nargs="+", default=[], metavar="REGEX", 

438 help="During snapshot creation, replication (and deletion) and comparison, select any ZFS dataset (and its " 

439 "descendants) that is contained within SRC_DATASET (DST_DATASET in case of deletion) if its relative dataset " 

440 "path (e.g. `baz/tmp`) wrt. SRC_DATASET (DST_DATASET in case of deletion) matches at least one of the given " 

441 "include regular expressions but none of the exclude regular expressions. " 

442 "If a dataset is excluded its descendants are automatically excluded too, and this decision is never " 

443 "reconsidered even for the descendants because exclude takes precedence over include.\n\n" 

444 "This option can be specified multiple times. " 

445 "A leading `!` character indicates logical negation, i.e. the regex matches if the regex with the " 

446 "leading `!` character removed does not match.\n\n" 

447 "If the option starts with a `+` prefix then regex names are read from the newline-separated " 

448 "UTF-8 text file given after the `+` prefix, one regex per line inside of the text file. The basename " 

449 "must contain the substring 'bzfs_argument_file'.\n\n" 

450 "Default: `.*` (include all datasets).\n\n" 

451 "Examples: `baz/tmp`, `(.*/)?doc[^/]*/(private|confidential).*`, `!public`, " 

452 "`+dataset_regexes_bzfs_argument_file.txt`, `+/path/to/dataset_regexes_bzfs_argument_file.txt`\n\n") 

453 parser.add_argument( 

454 "--exclude-dataset-regex", action=FileOrLiteralAction, nargs="+", default=[], metavar="REGEX", 

455 help="Same syntax as --include-dataset-regex (see above) except that the default is " 

456 f"`{EXCLUDE_DATASET_REGEXES_DEFAULT}` (exclude tmp datasets). Example: `!.*` (exclude no dataset)\n\n") 

457 parser.add_argument( 

458 "--exclude-dataset-property", default=None, action=NonEmptyStringAction, metavar="STRING", 

459 help="The name of a ZFS dataset user property (optional). If this option is specified, the effective value " 

460 "(potentially inherited) of that user property is read via 'zfs list' for each selected source dataset " 

461 "to determine whether the dataset will be included or excluded, as follows:\n\n" 

462 "a) Value is 'true' or '-' or empty string or the property is missing: Include the dataset.\n\n" 

463 "b) Value is 'false': Exclude the dataset and its descendants.\n\n" 

464 "c) Value is a comma-separated list of host names (no spaces, for example: " 

465 "'store001,store002'): Include the dataset if the host name of " 

466 f"the host executing {PROG_NAME} is contained in the list, otherwise exclude the dataset and its " 

467 "descendants.\n\n" 

468 "If a dataset is excluded its descendants are automatically excluded too, and the property values of the " 

469 "descendants are ignored because exclude takes precedence over include.\n\n" 

470 "Examples: 'syncoid:sync', 'com.example.eng.project.x:backup'\n\n" 

471 "*Note:* The use of --exclude-dataset-property is discouraged for most use cases. It is more flexible, " 

472 "more powerful, *and* more efficient to instead use a combination of --include/exclude-dataset-regex " 

473 "and/or --include/exclude-dataset to achieve the same or better outcome.\n\n") 

474 parser.add_argument( 

475 "--include-snapshot-regex", action=FileOrLiteralAction, nargs="+", default=[], metavar="REGEX", 

476 help="During replication, deletion and comparison, select any source ZFS snapshot that has a name (i.e. the part " 

477 "after the '@') that matches at least one of the given include regular expressions but none of the " 

478 "exclude regular expressions. If a snapshot is excluded this decision is never reconsidered because " 

479 "exclude takes precedence over include.\n\n" 

480 "This option can be specified multiple times. " 

481 "A leading `!` character indicates logical negation, i.e. the regex matches if the regex with the " 

482 "leading `!` character removed does not match.\n\n" 

483 "Default: `.*` (include all snapshots). " 

484 "Examples: `test_.*`, `!prod_.*`, `.*_(hourly|frequent)`, `!.*_(weekly|daily)`\n\n" 

485 "*Note:* All --include/exclude-snapshot-* CLI option groups are combined into a mini filter pipeline. " 

486 "A filter pipeline is executed in the order given on the command line, left to right. For example if " 

487 "--include-snapshot-times-and-ranks (see below) is specified on the command line before " 

488 "--include/exclude-snapshot-regex, then --include-snapshot-times-and-ranks will be applied before " 

489 "--include/exclude-snapshot-regex. The pipeline results would not always be the same if the order were " 

490 "reversed. Order matters.\n\n" 

491 "*Note:* During replication, bookmarks are always retained aka selected in order to help find common " 

492 "snapshots between source and destination.\n\n") 

493 parser.add_argument( 

494 "--exclude-snapshot-regex", action=FileOrLiteralAction, nargs="+", default=[], metavar="REGEX", 

495 help="Same syntax as --include-snapshot-regex (see above) except that the default is to exclude no " 

496 "snapshots.\n\n") 

497 parser.add_argument( 

498 "--include-snapshot-times-and-ranks", action=TimeRangeAndRankRangeAction, nargs="+", default=[], 

499 metavar=("TIMERANGE", "RANKRANGE"), 

500 help="This option takes as input parameters a time range filter and an optional rank range filter. It " 

501 "separately computes the results for each filter and selects the UNION of both results. " 

502 "To instead use a pure rank range filter (no UNION), or a pure time range filter (no UNION), simply " 

503 "use 'notime' aka '0..0' to indicate an empty time range, or omit the rank range, respectively. " 

504 "This option can be specified multiple times.\n\n" 

505 "<b>*Replication Example (UNION):* </b>\n\n" 

506 "Specify to replicate all daily snapshots created during the last 7 days, " 

507 "and at the same time ensure that the latest 7 daily snapshots (per dataset) are replicated regardless " 

508 "of creation time, like so: " 

509 "`--include-snapshot-regex '.*_daily' --include-snapshot-times-and-ranks '7 days ago..anytime' 'latest 7'`\n\n" 

510 "<b>*Deletion Example (no UNION):* </b>\n\n" 

511 "Specify to delete all daily snapshots older than 7 days, but ensure that the " 

512 "latest 7 daily snapshots (per dataset) are retained regardless of creation time, like so: " 

513 "`--include-snapshot-regex '.*_daily' --include-snapshot-times-and-ranks notime 'all except latest 7' " 

514 "--include-snapshot-times-and-ranks 'anytime..7 days ago'`" 

515 "\n\n" 

516 "This helps to safely cope with irregular scenarios where no snapshots were created or received within " 

517 "the last 7 days, or where more than 7 daily snapshots were created within the last 7 days. It can also " 

518 "help to avoid accidental pruning of the last snapshot that source and destination have in common.\n\n" 

519 "" 

520 "<b>*TIMERANGE:* </b>\n\n" 

521 "The ZFS 'creation' time of a snapshot (and bookmark) must fall into this time range in order for the " 

522 "snapshot to be included. The time range consists of a 'start' time, followed by a '..' separator, " 

523 "followed by an 'end' time. For example '2024-01-01..2024-04-01', or 'anytime..anytime' aka `*..*` aka all " 

524 "times, or 'notime' aka '0..0' aka empty time range. Only snapshots (and bookmarks) in the half-open time " 

525 "range [start, end) are included; other snapshots (and bookmarks) are excluded. If a snapshot is excluded " 

526 "this decision is never reconsidered because exclude takes precedence over include. Each of the two specified " 

527 "times can take any of the following forms:\n\n" 

528 "* a) `anytime` aka `*` wildcard; represents negative or positive infinity.\n\n" 

529 "* b) a non-negative integer representing a UTC Unix time in seconds. Example: 1728109805\n\n" 

530 "* c) an ISO 8601 datetime string with or without timezone. Examples: '2024-10-05', " 

531 "'2024-10-05T14:48:55', '2024-10-05T14:48:55+02', '2024-10-05T14:48:55-04:30'. If the datetime string " 

532 "does not contain time zone info then it is assumed to be in the local time zone. Timezone string support " 

533 "requires Python ≥ 3.11.\n\n" 

534 "* d) a duration that indicates how long ago from the current time, using the following syntax: " 

535 "a non-negative integer, followed by an optional space, followed by a duration unit that is " 

536 "*one* of 'seconds', 'secs', 'minutes', 'mins', 'hours', 'days', 'weeks', 'months', 'years', " 

537 "followed by an optional space, followed by the word 'ago'. " 

538 "Examples: '0secs ago', '40 mins ago', '36hours ago', '90days ago', '12weeksago'.\n\n" 

539 "* Note: This option compares the specified time against the standard ZFS 'creation' time property of the " 

540 "snapshot (which is a UTC Unix time in integer seconds), rather than against a timestamp that may be " 

541 "part of the snapshot name. You can list the ZFS creation time of snapshots and bookmarks as follows: " 

542 "`zfs list -t snapshot,bookmark -o name,creation -s creation -d 1 $SRC_DATASET` (optionally add " 

543 "the -p flag to display UTC Unix time in integer seconds).\n\n" 

544 "*Note:* During replication, bookmarks are always retained aka selected in order to help find common " 

545 "snapshots between source and destination.\n\n" 

546 "" 

547 "<b>*RANKRANGE:* </b>\n\n" 

548 "Specifies to include the N (or N%%) oldest snapshots or latest snapshots, and exclude all other " 

549 "snapshots (default: include no snapshots). Snapshots are sorted by creation time (actually, by the " 

550 "'createtxg' ZFS property, which serves the same purpose but is more precise). The rank position of a " 

551 "snapshot is the zero-based integer position of the snapshot within that sorted list. A rank consists of the " 

552 "optional words 'all except' (followed by an optional space), followed by the word 'oldest' or 'latest', " 

553 "followed by a non-negative integer, followed by an optional '%%' percent sign. A rank range consists of a " 

554 "lower rank, followed by a '..' separator, followed by a higher rank. " 

555 "If the optional lower rank is missing it is assumed to be 0. Examples:\n\n" 

556 "* 'oldest 10%%' aka 'oldest 0..oldest 10%%' (include the oldest 10%% of all snapshots)\n\n" 

557 "* 'latest 10%%' aka 'latest 0..latest 10%%' (include the latest 10%% of all snapshots)\n\n" 

558 "* 'all except latest 10%%' aka 'oldest 90%%' aka 'oldest 0..oldest 90%%' (include all snapshots except the " 

559 "latest 10%% of all snapshots)\n\n" 

560 "* 'oldest 90' aka 'oldest 0..oldest 90' (include the oldest 90 snapshots)\n\n" 

561 "* 'latest 90' aka 'latest 0..latest 90' (include the latest 90 snapshots)\n\n" 

562 "* 'all except oldest 90' aka 'oldest 90..oldest 100%%' (include all snapshots except the oldest 90 snapshots)" 

563 "\n\n" 

564 "* 'all except latest 90' aka 'latest 90..latest 100%%' (include all snapshots except the latest 90 snapshots)" 

565 "\n\n" 

566 "* 'latest 1' aka 'latest 0..latest 1' (include the latest snapshot)\n\n" 

567 "* 'all except latest 1' aka 'latest 1..latest 100%%' (include all snapshots except the latest snapshot)\n\n" 

568 "* 'oldest 2' aka 'oldest 0..oldest 2' (include the oldest 2 snapshots)\n\n" 

569 "* 'all except oldest 2' aka 'oldest 2..oldest 100%%' (include all snapshots except the oldest 2 snapshots)\n\n" 

570 "* 'oldest 100%%' aka 'oldest 0..oldest 100%%' (include all snapshots)\n\n" 

571 "* 'oldest 0%%' aka 'oldest 0..oldest 0%%' (include no snapshots)\n\n" 

572 "* 'oldest 0' aka 'oldest 0..oldest 0' (include no snapshots)\n\n" 

573 "*Note:* If multiple RANKRANGEs are specified within a single --include-snapshot-times-and-ranks option, each " 

574 "subsequent rank range operates on the output of the preceding rank rage.\n\n" 

575 "*Note:* Percentage calculations are not based on the number of snapshots " 

576 "contained in the dataset on disk, but rather based on the number of snapshots arriving at the filter. " 

577 "For example, if only two daily snapshots arrive at the filter because a prior filter excludes hourly " 

578 "snapshots, then 'latest 10' will only include these two daily snapshots, and 'latest 50%%' will only " 

579 "include one of these two daily snapshots.\n\n" 

580 "*Note:* During replication, bookmarks are always retained aka selected in order to help find common " 

581 "snapshots between source and destination. Bookmarks do not count towards N or N%% wrt. rank.\n\n" 

582 "*Note:* If a snapshot is excluded this decision is never reconsidered because exclude takes precedence " 

583 "over include.\n\n") 

584 

585 src_snapshot_plan_example = { 

586 "prod": { 

587 "onsite": {"secondly": 40, "minutely": 40, "hourly": 36, "daily": 31, "weekly": 12, "monthly": 18, "yearly": 5}, 

588 "us-west-1": {"secondly": 0, "minutely": 0, "hourly": 36, "daily": 31, "weekly": 12, "monthly": 18, "yearly": 5}, 

589 "eu-west-1": {"secondly": 0, "minutely": 0, "hourly": 36, "daily": 31, "weekly": 12, "monthly": 18, "yearly": 5}, 

590 }, 

591 "test": { 

592 "offsite": {"12hourly": 42, "weekly": 12}, 

593 "onsite": {"100millisecondly": 42}, 

594 }, 

595 } 

596 parser.add_argument( 

597 "--include-snapshot-plan", action=IncludeSnapshotPlanAction, default=None, metavar="DICT_STRING", 

598 help="Replication periods to be used if replicating snapshots within the selected destination datasets. " 

599 "Has the same format as --create-src-snapshots-plan and --delete-dst-snapshots-except-plan (see below). " 

600 "Snapshots that do not match a period will not be replicated. To avoid unexpected surprises, make sure to " 

601 "carefully specify ALL snapshot names and periods that shall be replicated, in combination with --dryrun.\n\n" 

602 f"Example: `{format_dict(src_snapshot_plan_example)}`. This example will, for the organization 'prod' and the " 

603 "intended logical target 'onsite', replicate secondly snapshots that were created less than 40 seconds ago, " 

604 "yet replicate the latest 40 secondly snapshots regardless of creation time. Analog for the latest 40 minutely " 

605 "snapshots, latest 36 hourly snapshots, etc. " 

606 "Note: A zero within a period (e.g. 'hourly': 0) indicates that no snapshots shall be replicated for the given " 

607 "period.\n\n" 

608 "Note: --include-snapshot-plan is a convenience option that auto-generates a series of the following other " 

609 "options: --new-snapshot-filter-group, --include-snapshot-regex, --include-snapshot-times-and-ranks\n\n") 

610 parser.add_argument( 

611 "--new-snapshot-filter-group", action=NewSnapshotFilterGroupAction, nargs=0, 

612 help="Starts a new snapshot filter group containing separate --{include|exclude}-snapshot-* filter options. The " 

613 "program separately computes the results for each filter group and selects the UNION of all results. " 

614 "This option can be specified multiple times and serves as a separator between groups. Example:\n\n" 

615 "Delete all minutely snapshots older than 40 minutes, but ensure that the latest 40 minutely snapshots (per " 

616 "dataset) are retained regardless of creation time. Additionally, delete all hourly snapshots older than 36 " 

617 "hours, but ensure that the latest 36 hourly snapshots (per dataset) are retained regardless of creation time. " 

618 "Additionally, delete all daily snapshots older than 31 days, but ensure that the latest 31 daily snapshots " 

619 "(per dataset) are retained regardless of creation time: " 

620 f"`{PROG_NAME} {DUMMY_DATASET} tank2/boo/bar --dryrun --recursive --skip-replication --delete-dst-snapshots " 

621 "--include-snapshot-regex '.*_minutely' --include-snapshot-times-and-ranks notime 'all except latest 40' " 

622 "--include-snapshot-times-and-ranks 'anytime..40 minutes ago' " 

623 "--new-snapshot-filter-group " 

624 "--include-snapshot-regex '.*_hourly' --include-snapshot-times-and-ranks notime 'all except latest 36' " 

625 "--include-snapshot-times-and-ranks 'anytime..36 hours ago' " 

626 "--new-snapshot-filter-group " 

627 "--include-snapshot-regex '.*_daily' --include-snapshot-times-and-ranks notime 'all except latest 31' " 

628 "--include-snapshot-times-and-ranks 'anytime..31 days ago'`\n\n") 

629 parser.add_argument( 

630 "--create-src-snapshots", action="store_true", 

631 help="Do nothing if the --create-src-snapshots flag is missing. Otherwise, before the replication step (see below), " 

632 "atomically create new snapshots of the source datasets selected via --{include|exclude}-dataset* policy. " 

633 "The names of the snapshots can be configured via --create-src-snapshots-* suboptions (see below). " 

634 "To create snapshots only, without any other processing such as replication, etc, consider using this flag " 

635 "together with the --skip-replication flag.\n\n" 

636 "A periodic snapshot is created if it is due per the schedule indicated by --create-src-snapshots-plan " 

637 "(for example '_daily' or '_hourly' or _'10minutely' or '_2secondly' or '_100millisecondly'), or if the " 

638 "--create-src-snapshots-even-if-not-due flag is specified, or if the most recent scheduled snapshot " 

639 f"is somehow missing. In the latter case {PROG_NAME} immediately creates a snapshot (tagged with the current " 

640 "time, not backdated to the missed time), and then resumes the original schedule.\n\n" 

641 "If the snapshot suffix is '_adhoc' or not a known period then a snapshot is considered " 

642 "non-periodic and is thus created immediately regardless of the creation time of any existing snapshot.\n\n" 

643 "The implementation attempts to fit as many datasets as possible into a single (atomic) 'zfs snapshot' command " 

644 "line, using lexicographical sort order, and using 'zfs snapshot -r' to the extent that this is compatible " 

645 "with the actual results of the schedule and the actual results of the --{include|exclude}-dataset* pruning " 

646 "policy. The snapshots of all datasets that fit " 

647 "within the same single 'zfs snapshot' CLI invocation will be taken within the same ZFS transaction group, and " 

648 "correspondingly have identical 'createtxg' ZFS property (but not necessarily identical 'creation' ZFS time " 

649 "property as ZFS actually provides no such guarantee), and thus be consistent. Dataset names that can't fit " 

650 "into a single command line are spread over multiple command line invocations, respecting the limits that the " 

651 "operating system places on the maximum length of a single command line, per `getconf ARG_MAX`.\n\n" 

652 f"Note: All {PROG_NAME} functions including snapshot creation, replication, deletion, monitoring, comparison, " 

653 "etc. happily work with any snapshots in any format, even created or managed by third party ZFS snapshot " 

654 "management tools, including manual zfs snapshot/destroy.\n\n") 

655 parser.add_argument( 

656 "--create-src-snapshots-plan", default=None, type=str, metavar="DICT_STRING", 

657 help="Creation periods that specify a schedule for when new snapshots shall be created on src within the selected " 

658 "datasets. Has the same format as --delete-dst-snapshots-except-plan.\n\n" 

659 f"Example: `{format_dict(src_snapshot_plan_example)}`. This example will, for the organization 'prod' and " 

660 "the intended logical target 'onsite', create 'secondly' snapshots every second, 'minutely' snapshots every " 

661 "minute, hourly snapshots every hour, and so on. " 

662 "It will also create snapshots for the targets 'us-west-1' and 'eu-west-1' within the 'prod' organization. " 

663 "In addition, it will create snapshots every 12 hours and every week for the 'test' organization, " 

664 "and name them as being intended for the 'offsite' replication target. Analog for snapshots that are taken " 

665 "every 100 milliseconds within the 'test' organization.\n\n" 

666 "The example creates ZFS snapshots with names like " 

667 "`prod_onsite_<timestamp>_secondly`, `prod_onsite_<timestamp>_minutely`, " 

668 "`prod_us-west-1_<timestamp>_hourly`, `prod_us-west-1_<timestamp>_daily`, " 

669 "`prod_eu-west-1_<timestamp>_hourly`, `prod_eu-west-1_<timestamp>_daily`, " 

670 "`test_offsite_<timestamp>_12hourly`, `test_offsite_<timestamp>_weekly`, and so on.\n\n" 

671 "Note: A period name that is missing indicates that no snapshots shall be created for the given period.\n\n" 

672 "The period name can contain an optional positive integer immediately preceding the time period unit, for " 

673 "example `_2secondly` or `_10minutely` or `_100millisecondly` to indicate that snapshots are taken every 2 " 

674 "seconds, or every 10 minutes, or every 100 milliseconds, respectively.\n\n") 

675 

676 def argparser_escape(text: str) -> str: 

677 return text.replace("%", "%%") 

678 

679 parser.add_argument( 

680 "--create-src-snapshots-timeformat", default="%Y-%m-%d_%H:%M:%S", metavar="STRFTIME_SPEC", 

681 help="Default is `%(default)s`. For the strftime format, see " 

682 "https://docs.python.org/3.11/library/datetime.html#strftime-strptime-behavior. " 

683 f"Examples: `{argparser_escape('%Y-%m-%d_%H:%M:%S.%f')}` (adds microsecond resolution), " 

684 f"`{argparser_escape('%Y-%m-%d_%H:%M:%S%z')}` (adds timezone offset), " 

685 f"`{argparser_escape('%Y-%m-%dT%H-%M-%S')}` (no colons).\n\n" 

686 "The name of the snapshot created on the src is `$org_$target_strftime(--create-src-snapshots-time*)_$period`. " 

687 "Example: `tank/foo@prod_us-west-1_2024-09-03_12:26:15_daily`\n\n") 

688 parser.add_argument( 

689 "--create-src-snapshots-timezone", default="", type=str, metavar="TZ_SPEC", 

690 help=f"Default is the local timezone of the system running {PROG_NAME}. When creating a new snapshot on the source, " 

691 "fetch the current time in the specified timezone, and feed that time, and the value of " 

692 "--create-src-snapshots-timeformat, into the standard strftime() function to generate the timestamp portion " 

693 "of the snapshot name. The TZ_SPEC input parameter is of the form 'UTC' or '+HHMM' or '-HHMM' for fixed UTC " 

694 "offsets, or an IANA TZ identifier for auto-adjustment to daylight savings time, or the empty string to use " 

695 "the local timezone, for example '', 'UTC', '+0000', '+0530', '-0400', 'America/Los_Angeles', 'Europe/Vienna'. " 

696 "For a list of valid IANA TZ identifiers see https://en.wikipedia.org/wiki/List_of_tz_database_time_zones#List" 

697 "\n\nTo change the timezone not only for snapshot name creation, but in all respects for the entire program, " 

698 "use the standard 'TZ' Unix environment variable, like so: `export TZ=UTC`.\n\n") 

699 parser.add_argument( 

700 "--create-src-snapshots-even-if-not-due", action="store_true", 

701 help="Take snapshots immediately regardless of the creation time of any existing snapshot, even if snapshots " 

702 "are periodic and not actually due per the schedule.\n\n") 

703 parser.add_argument( 

704 "--zfs-send-program-opts", type=str, default="--raw --compressed", metavar="STRING", 

705 help="Parameters to fine-tune 'zfs send' behaviour (optional); will be passed into 'zfs send' CLI. " 

706 "The value is split on runs of one or more whitespace characters. " 

707 "Default is '%(default)s'. To run `zfs send` without options, specify the empty " 

708 "string: `--zfs-send-program-opts=''`. " 

709 "See https://openzfs.github.io/openzfs-docs/man/master/8/zfs-send.8.html " 

710 "and https://github.com/openzfs/zfs/issues/13024\n\n") 

711 parser.add_argument( 

712 "--zfs-recv-program-opts", type=str, default="-u", metavar="STRING", 

713 help="Parameters to fine-tune 'zfs receive' behaviour (optional); will be passed into 'zfs receive' CLI. " 

714 "The value is split on runs of one or more whitespace characters. " 

715 "Default is '%(default)s'. To run `zfs receive` without options, specify the empty " 

716 "string: `--zfs-recv-program-opts=''`. " 

717 "Example: '-u -o canmount=noauto -o readonly=on -x keylocation -x keyformat -x encryption'. " 

718 "See https://openzfs.github.io/openzfs-docs/man/master/8/zfs-receive.8.html " 

719 "and https://openzfs.github.io/openzfs-docs/man/master/7/zfsprops.7.html\n\n") 

720 parser.add_argument( 

721 "--zfs-recv-program-opt", action="append", default=[], metavar="STRING", 

722 help="Parameter to fine-tune 'zfs receive' behaviour (optional); will be passed into 'zfs receive' CLI. " 

723 "The value can contain spaces and is not split. This option can be specified multiple times. Example: `" 

724 "--zfs-recv-program-opt=-o " 

725 "--zfs-recv-program-opt='org.zfsbootmenu:commandline=ro debug zswap.enabled=1'`\n\n") 

726 parser.add_argument( 

727 "--preserve-properties", nargs="+", default=[], metavar="STRING", 

728 help="On replication, preserve the current value of ZFS properties with the given names on the destination " 

729 "datasets. The destination ignores the property value it 'zfs receive's from the source if the property name " 

730 "matches one of the given blacklist values. This prevents a compromised or untrusted source from overwriting " 

731 "security-critical properties on the destination. The default is to preserve none, i.e. an empty blacklist.\n\n" 

732 "Example blacklist that protects against dangerous overwrites: " 

733 "mountpoint overlay sharenfs sharesmb exec setuid devices encryption keyformat keylocation volsize\n\n" 

734 "See https://openzfs.github.io/openzfs-docs/man/master/7/zfsprops.7.html and " 

735 "https://openzfs.github.io/openzfs-docs/man/master/8/zfs-receive.8.html#x\n\n" 

736 "Note: --preserve-properties uses the 'zfs recv -x' option and thus requires either OpenZFS ≥ 2.2.0 " 

737 "(see https://github.com/openzfs/zfs/commit/b0269cd8ced242e66afc4fa856d62be29bb5a4ff), or that " 

738 "'zfs send --props' is not used.\n\n") 

739 parser.add_argument( 

740 "--force-rollback-to-latest-snapshot", action="store_true", 

741 help="Before replication, rollback the destination dataset to its most recent destination snapshot (if there " 

742 "is one), via 'zfs rollback', just in case the destination dataset was modified since its most recent " 

743 "snapshot. This is much less invasive than the other --force* options (see below).\n\n") 

744 parser.add_argument( 

745 "--force-rollback-to-latest-common-snapshot", action="store_true", 

746 help="Before replication, delete destination ZFS snapshots that are more recent than the most recent common " 

747 "snapshot ('conflicting snapshots'), via 'zfs rollback'. Do no rollback if no common snapshot exists.\n\n") 

748 parser.add_argument( 

749 "--force", action="store_true", 

750 help="Same as --force-rollback-to-latest-common-snapshot (see above), except that additionally, if no common " 

751 "snapshot exists, then delete all destination snapshots before starting replication, and proceed " 

752 "without aborting. Without the --force* flags, the destination dataset is treated as append-only, hence " 

753 "no destination snapshot that already exists is deleted, and instead the operation is aborted with an " 

754 "error when encountering a conflicting snapshot.\n\n" 

755 "Analogy: --force-rollback-to-latest-snapshot is a tiny hammer, whereas " 

756 "--force-rollback-to-latest-common-snapshot is a medium sized hammer, --force is a large hammer, and " 

757 "--force-destroy-dependents is a very large hammer. " 

758 "Consider using the smallest hammer that can fix the problem. No hammer is ever used by default.\n\n") 

759 parser.add_argument( 

760 "--force-destroy-dependents", action="store_true", 

761 help="On destination, --force and --force-rollback-to-latest-common-snapshot and --delete-* will add the " 

762 "'-R' flag to their use of 'zfs rollback' and 'zfs destroy', causing them to delete dependents such as " 

763 "clones and bookmarks. This can be very destructive and is rarely advisable.\n\n") 

764 parser.add_argument( 

765 "--force-unmount", action="store_true", 

766 help="On destination, --force and --force-rollback-to-latest-common-snapshot will add the '-f' flag to their " 

767 "use of 'zfs rollback' and 'zfs destroy'.\n\n") 

768 parser.add_argument( 

769 "--force-once", "--f1", action="store_true", 

770 help="Use the --force option or --force-rollback-to-latest-common-snapshot option at most once to resolve a " 

771 "conflict, then abort with an error on any subsequent conflict. This helps to interactively resolve " 

772 "conflicts, one conflict at a time.\n\n") 

773 parser.add_argument( 

774 "--skip-parent", action="store_true", 

775 help="During replication and deletion, skip processing of the SRC_DATASET and DST_DATASET and only process " 

776 "their descendant datasets, i.e. children, and children of children, etc (with --recursive). No dataset " 

777 "is processed unless --recursive is also specified. " 

778 f"Analogy: `{PROG_NAME} --recursive --skip-parent src dst` is akin to Unix `cp -r src/* dst/` whereas " 

779 f" `{PROG_NAME} --recursive --skip-parent --skip-replication --delete-dst-datasets dummy dst` is akin to " 

780 "Unix `rm -r dst/*`\n\n") 

781 parser.add_argument( 

782 "--skip-missing-snapshots", choices=["fail", "dataset", "continue"], default="dataset", nargs="?", 

783 help="During replication, handle source datasets that select no snapshots (and no relevant bookmarks) " 

784 "as follows:\n\n" 

785 "a) 'fail': Abort with an error.\n\n" 

786 "b) 'dataset' (default): Skip the source dataset with a warning. Skip descendant datasets if " 

787 "--recursive and destination dataset does not exist. Otherwise skip to the next dataset.\n\n" 

788 "c) 'continue': Skip nothing. If destination snapshots exist, delete them (with --force) or abort " 

789 "with an error (without --force). If there is no such abort, continue processing with the next dataset. " 

790 "Eventually create empty destination dataset and ancestors if they do not yet exist and source dataset " 

791 "has at least one descendant that selects at least one snapshot.\n\n") 

792 parser.add_argument( 

793 "--retries", dest="max_retries", type=int, min=0, default=2, action=CheckRange, metavar="INT", 

794 help="The maximum number of times a retryable replication or deletion step shall be retried if it fails, for " 

795 "example because of network hiccups (default: %(default)s, min: %(min)s). " 

796 "Also consider this option if a periodic pruning script may simultaneously delete a dataset or " 

797 f"snapshot or bookmark while {PROG_NAME} is running and attempting to access it.\n\n") 

798 parser.add_argument( 

799 "--retry-min-sleep-secs", type=float, min=0, default=0, action=CheckRange, metavar="FLOAT", 

800 help="The minimum duration to sleep between retries (default: %(default)s).\n\n") 

801 parser.add_argument( 

802 "--retry-initial-max-sleep-secs", type=float, min=0, default=0.125, action=CheckRange, metavar="FLOAT", 

803 help="The initial maximum duration to sleep between retries (default: %(default)s).\n\n") 

804 parser.add_argument( 

805 "--retry-max-sleep-secs", type=float, min=0, default=5 * 60, action=CheckRange, metavar="FLOAT", 

806 help="The maximum duration to sleep between retries initially starts with --retry-initial-max-sleep-secs " 

807 "(see above), and doubles on each retry, up to the final maximum of --retry-max-sleep-secs " 

808 "(default: %(default)s). On each retry a random sleep time in the [--retry-min-sleep-secs, current max] range " 

809 "is picked. In a nutshell: retry-min-sleep-secs ≤ retry-initial-max-sleep-secs ≤ retry-max-sleep-secs. " 

810 "The timer resets after each operation.\n\n") 

811 parser.add_argument( 

812 "--retry-max-elapsed-secs", type=float, min=0, default=60 * 60, action=CheckRange, metavar="FLOAT", 

813 help="A single operation (e.g. 'zfs send/receive' of the current dataset, or deletion of a list of snapshots " 

814 "within the current dataset) will not be retried (or not retried anymore) once this much time has elapsed " 

815 "since the initial start of the operation, including retries (default: %(default)s). " 

816 "The timer resets after each operation completes or retries exhaust, such that subsequently failing " 

817 "operations can again be retried.\n\n") 

818 parser.add_argument( 

819 "--skip-on-error", choices=["fail", "tree", "dataset"], default=SKIP_ON_ERROR_DEFAULT, 

820 help="During replication and deletion, if an error is not retryable, or --retries has been exhausted, " 

821 "or --skip-missing-snapshots raises an error, proceed as follows:\n\n" 

822 "a) 'fail': Abort the program with an error. This mode is ideal for testing, clear " 

823 "error reporting, and situations where consistency trumps availability.\n\n" 

824 "b) 'tree': Log the error, skip the dataset tree rooted at the dataset for which the error " 

825 "occurred, and continue processing the next (sibling) dataset tree. " 

826 "Example: Assume datasets tank/user1/foo and tank/user2/bar and an error occurs while processing " 

827 "tank/user1. In this case processing skips tank/user1/foo and proceeds with tank/user2.\n\n" 

828 "c) 'dataset' (default): Same as 'tree' except if the destination dataset already exists, skip to " 

829 "the next dataset instead.\n\n" 

830 "Example: Assume datasets tank/user1/foo and tank/user2/bar and an error occurs while " 

831 "processing tank/user1. In this case processing skips tank/user1 and proceeds with tank/user1/foo " 

832 "if the destination already contains tank/user1. Otherwise processing continues with tank/user2. " 

833 "This mode is for production use cases that require timely forward progress even in the presence of " 

834 "partial failures. For example, assume the job is to backup the home directories or virtual machines " 

835 "of thousands of users across an organization. Even if replication of some of the datasets for some " 

836 "users fails due too conflicts, busy datasets, etc, the replication job will continue for the " 

837 "remaining datasets and the remaining users.\n\n") 

838 parser.add_argument( 

839 "--skip-replication", action="store_true", 

840 help="Skip replication step (see above) and proceed to the optional --delete-dst-datasets step " 

841 "immediately (see below).\n\n") 

842 parser.add_argument( 

843 "--delete-dst-datasets", action="store_true", 

844 help="Do nothing if the --delete-dst-datasets option is missing. Otherwise, after successful replication " 

845 "step, if any, delete existing destination datasets that are selected via --{include|exclude}-dataset* " 

846 "policy yet do not exist within SRC_DATASET (which can be an empty dataset, such as the hardcoded virtual " 

847 f"dataset named '{DUMMY_DATASET}'!). Do not recurse without --recursive. With --recursive, never delete " 

848 "non-selected dataset subtrees or their ancestors.\n\n" 

849 "For example, if the destination contains datasets h1,h2,h3,d1 whereas source only contains h3, " 

850 "and the include/exclude policy selects h1,h2,h3,d1, then delete datasets h1,h2,d1 on " 

851 "the destination to make it 'the same'. On the other hand, if the include/exclude policy " 

852 "only selects h1,h2,h3 then only delete datasets h1,h2 on the destination to make it 'the same'.\n\n" 

853 "Example to delete all tmp datasets within tank2/boo/bar: " 

854 f"`{PROG_NAME} {DUMMY_DATASET} tank2/boo/bar --dryrun --skip-replication --recursive " 

855 "--delete-dst-datasets --include-dataset-regex '(.*/)?tmp.*' --exclude-dataset-regex '!.*'`\n\n") 

856 parser.add_argument( 

857 "--delete-dst-snapshots", choices=["snapshots", "bookmarks"], default=None, const="snapshots", nargs="?", 

858 help="Do nothing if the --delete-dst-snapshots option is missing. Otherwise, after successful " 

859 "replication, and successful --delete-dst-datasets step, if any, delete existing destination snapshots " 

860 "whose GUID does not exist within the source dataset (which can be an empty dummy dataset!) if the " 

861 "destination snapshots are selected by the --include/exclude-snapshot-* policy, and the destination " 

862 "dataset is selected via --{include|exclude}-dataset* policy. Does not recurse without --recursive.\n\n" 

863 "For example, if the destination dataset contains snapshots h1,h2,h3,d1 (h=hourly, d=daily) whereas " 

864 "the source dataset only contains snapshot h3, and the include/exclude policy selects " 

865 "h1,h2,h3,d1, then delete snapshots h1,h2,d1 on the destination dataset to make it 'the same'. " 

866 "On the other hand, if the include/exclude policy only selects snapshots h1,h2,h3 then only " 

867 "delete snapshots h1,h2 on the destination dataset to make it 'the same'.\n\n" 

868 "*Note:* To delete snapshots regardless, consider using --delete-dst-snapshots in combination with a " 

869 f"source that is an empty dataset, such as the hardcoded virtual dataset named '{DUMMY_DATASET}', like so:" 

870 f" `{PROG_NAME} {DUMMY_DATASET} tank2/boo/bar --dryrun --skip-replication --delete-dst-snapshots " 

871 "--include-snapshot-regex '.*_daily' --recursive`\n\n" 

872 "*Note:* Use --delete-dst-snapshots=bookmarks to delete bookmarks instead of snapshots, in which " 

873 "case no snapshots are selected and the --{include|exclude}-snapshot-* filter options treat bookmarks as " 

874 "snapshots wrt. selecting.\n\n" 

875 "*Performance Note:* --delete-dst-snapshots operates on multiple datasets in parallel (and serially " 

876 f"within a dataset), using the same dataset order as {PROG_NAME} replication. " 

877 "The degree of parallelism is configurable with the --threads option (see below).\n\n") 

878 parser.add_argument( 

879 "--delete-dst-snapshots-no-crosscheck", action="store_true", 

880 help="This flag indicates that --delete-dst-snapshots=snapshots shall check the source dataset only for " 

881 "a snapshot with the same GUID, and ignore whether a bookmark with the same GUID is present in the " 

882 "source dataset. Similarly, it also indicates that --delete-dst-snapshots=bookmarks shall check the " 

883 "source dataset only for a bookmark with the same GUID, and ignore whether a snapshot with the same GUID " 

884 "is present in the source dataset.\n\n") 

885 parser.add_argument( 

886 "--delete-dst-snapshots-except", action="store_true", 

887 help="This flag indicates that the --include/exclude-snapshot-* options shall have inverted semantics for the " 

888 "--delete-dst-snapshots option, thus deleting all snapshots except for the selected snapshots (within the " 

889 "specified datasets), instead of deleting all selected snapshots (within the specified datasets). In other " 

890 "words, this flag enables to specify which snapshots to retain instead of which snapshots to delete.\n\n" 

891 "*Synchronization vs. Backup*: When a real (non-dummy) source dataset is specified in combination with " 

892 "--delete-dst-snapshots-except, then any destination snapshot retained by the rules above is actually only " 

893 "retained if it also exists in the source dataset - __all other destination snapshots are deleted__. This is " 

894 "great for synchronization use cases but should __NEVER BE USED FOR LONG-TERM ARCHIVAL__. Long-term archival " 

895 "use cases should instead specify the `dummy` source dataset as they require an independent retention policy " 

896 "that is not tied to the current contents of the source dataset.\n\n") 

897 parser.add_argument( 

898 "--delete-dst-snapshots-except-plan", action=DeleteDstSnapshotsExceptPlanAction, default=None, metavar="DICT_STRING", 

899 help="Retention periods to be used if pruning snapshots or bookmarks within the selected destination datasets via " 

900 "--delete-dst-snapshots. Has the same format as --create-src-snapshots-plan. " 

901 "Snapshots (--delete-dst-snapshots=snapshots) or bookmarks (with --delete-dst-snapshots=bookmarks) that " 

902 "do not match a period will be deleted. To avoid unexpected surprises, make sure to carefully specify ALL " 

903 "snapshot names and periods that shall be retained, in combination with --dryrun.\n\n" 

904 f"Example: `{format_dict(src_snapshot_plan_example)}`. This example will, for the organization 'prod' and " 

905 "the intended logical target 'onsite', retain secondly snapshots that were created less than 40 seconds ago, " 

906 "yet retain the latest 40 secondly snapshots regardless of creation time. Analog for the latest 40 minutely " 

907 "snapshots, latest 36 hourly snapshots, etc. " 

908 "It will also retain snapshots for the targets 'us-west-1' and 'eu-west-1' within the 'prod' organization. " 

909 "In addition, within the 'test' organization, it will retain snapshots that are created every 12 hours and " 

910 "every week as specified, and name them as being intended for the 'offsite' replication target. Analog for " 

911 "snapshots that are taken every 100 milliseconds within the 'test' organization. " 

912 "All other snapshots within the selected datasets will be deleted - you've been warned!\n\n" 

913 "The example scans the selected ZFS datasets for snapshots with names like " 

914 "`prod_onsite_<timestamp>_secondly`, `prod_onsite_<timestamp>_minutely`, " 

915 "`prod_us-west-1_<timestamp>_hourly`, `prod_us-west-1_<timestamp>_daily`, " 

916 "`prod_eu-west-1_<timestamp>_hourly`, `prod_eu-west-1_<timestamp>_daily`, " 

917 "`test_offsite_<timestamp>_12hourly`, `test_offsite_<timestamp>_weekly`, and so on, and deletes all snapshots " 

918 "that do not match a retention rule.\n\n" 

919 "Note: A zero within a period (e.g. 'hourly': 0) indicates that no snapshots shall be retained for the given " 

920 "period.\n\n" 

921 "Note: --delete-dst-snapshots-except-plan is a convenience option that auto-generates a series of the " 

922 "following other options: --delete-dst-snapshots-except, " 

923 "--new-snapshot-filter-group, --include-snapshot-regex, --include-snapshot-times-and-ranks\n\n") 

924 parser.add_argument( 

925 "--delete-empty-dst-datasets", choices=["snapshots", "snapshots+bookmarks"], default=None, 

926 const="snapshots+bookmarks", nargs="?", 

927 help="Do nothing if the --delete-empty-dst-datasets option is missing or --recursive is missing. Otherwise, " 

928 "after successful replication " 

929 "step and successful --delete-dst-datasets and successful --delete-dst-snapshots steps, if any, " 

930 "delete any selected destination dataset that has no snapshot and no bookmark if all descendants of " 

931 "that destination dataset are also selected and do not have a snapshot or bookmark either " 

932 "(again, only if the existing destination dataset is selected via --{include|exclude}-dataset* policy). " 

933 "Never delete non-selected dataset subtrees or their ancestors.\n\n" 

934 "For example, if the destination contains datasets h1,d1, and the include/exclude policy " 

935 "selects h1,d1, then check if h1,d1 can be deleted. " 

936 "On the other hand, if the include/exclude policy only selects h1 then only check if h1 can be deleted.\n\n" 

937 "*Note:* Use --delete-empty-dst-datasets=snapshots to delete snapshot-less datasets even if they still " 

938 "contain bookmarks.\n\n") 

939 monitor_snapshot_plan_example = { 

940 "prod": { 

941 "onsite": { 

942 "100millisecondly": {"latest": {"warning": "300 milliseconds", "critical": "2 seconds"}}, 

943 "secondly": {"latest": {"warning": "2 seconds", "critical": "14 seconds"}}, 

944 "minutely": {"latest": {"warning": "30 seconds", "critical": "300 seconds"}}, 

945 "hourly": {"latest": {"warning": "30 minutes", "critical": "300 minutes"}}, 

946 "daily": {"latest": {"warning": "4 hours", "critical": "8 hours"}}, 

947 "weekly": {"latest": {"warning": "2 days", "critical": "8 days"}}, 

948 "monthly": {"latest": {"warning": "2 days", "critical": "8 days"}}, 

949 "yearly": {"latest": {"warning": "5 days", "critical": "14 days"}}, 

950 "10minutely": {"latest": {"warning": "0 minutes", "critical": "0 minutes"}}, 

951 }, 

952 "": { 

953 "daily": {"latest": {"warning": "4 hours", "critical": "8 hours"}}, 

954 }, 

955 }, 

956 } 

957 parser.add_argument( 

958 "--monitor-snapshots", default="{}", type=str, metavar="DICT_STRING", 

959 help="Do nothing if the --monitor-snapshots flag is missing. Otherwise, after all other steps, " 

960 "alert the user if the ZFS 'creation' time property of the latest snapshot for any specified snapshot name " 

961 "pattern within the selected datasets is too old wrt. the specified age limit. The purpose is to check if " 

962 "snapshots are successfully taken on schedule, successfully replicated on schedule, and successfully pruned on " 

963 "schedule. Process exit code is 0, 1, 2 on OK, WARNING, CRITICAL, respectively. " 

964 f"Example DICT_STRING: `{format_dict(monitor_snapshot_plan_example)}`. " 

965 "This example alerts the user if the latest src or dst snapshot named `prod_onsite_<timestamp>_hourly` is more " 

966 "than 30 minutes late (i.e. more than 30+60=90 minutes old) [warning] or more than 300 minutes late (i.e. more " 

967 "than 300+60=360 minutes old) [critical]. " 

968 "Analog for the latest snapshot named `prod_<timestamp>_daily`, and so on.\n\n" 

969 "Note: A duration that is missing or zero (e.g. '0 minutes') indicates that no snapshots shall be checked for " 

970 "the given snapshot name pattern.\n\n") 

971 parser.add_argument( 

972 "--monitor-snapshots-dont-warn", action="store_true", 

973 help="Log a message for monitoring warnings but nonetheless exit with zero exit code.\n\n") 

974 parser.add_argument( 

975 "--monitor-snapshots-dont-crit", action="store_true", 

976 help="Log a message for monitoring criticals but nonetheless exit with zero exit code.\n\n") 

977 parser.add_argument( 

978 "--monitor-snapshots-no-latest-check", action="store_true", 

979 # help="Disable monitoring check of latest snapshot.\n\n") 

980 help=argparse.SUPPRESS) 

981 parser.add_argument( 

982 "--monitor-snapshots-no-oldest-check", action="store_true", 

983 # help="Disable monitoring check of oldest snapshot.\n\n") 

984 help=argparse.SUPPRESS) 

985 cmp_choices_dflt: str = "+".join(CMP_CHOICES_ITEMS) 

986 cmp_choices: list[str] = [] 

987 for i in range(len(CMP_CHOICES_ITEMS)): 

988 cmp_choices += ["+".join(c) for c in itertools.combinations(CMP_CHOICES_ITEMS, i + 1)] 

989 parser.add_argument( 

990 "--compare-snapshot-lists", choices=cmp_choices, default="", const=cmp_choices_dflt, nargs="?", 

991 help="Do nothing if the --compare-snapshot-lists option is missing. Otherwise, after successful replication " 

992 "step and successful --delete-dst-datasets, --delete-dst-snapshots steps and --delete-empty-dst-datasets " 

993 "steps, if any, proceed as follows:\n\n" 

994 "Compare source and destination dataset trees recursively wrt. snapshots, for example to check if all " 

995 "recently taken snapshots have been successfully replicated by a periodic job.\n\n" 

996 "Example: List snapshots only contained in source (tagged with 'src'), only contained in destination " 

997 "(tagged with 'dst'), and contained in both source and destination (tagged with 'all'), restricted to " 

998 "hourly and daily snapshots taken within the last 7 days, excluding the last 4 hours (to allow for some " 

999 "slack/stragglers), excluding temporary datasets: " 

1000 f"`{PROG_NAME} tank1/foo/bar tank2/boo/bar --skip-replication " 

1001 "--compare-snapshot-lists=src+dst+all --recursive --include-snapshot-regex '.*_(hourly|daily)' " 

1002 "--include-snapshot-times-and-ranks '7 days ago..4 hours ago' --exclude-dataset-regex 'tmp.*'`\n\n" 

1003 "This outputs a TSV file containing the following columns:\n\n" 

1004 "`location creation_iso createtxg rel_name guid root_dataset rel_dataset name creation written`\n\n" 

1005 "Example output row:\n\n" 

1006 "`src 2024-11-06_08:30:05 17435050 /foo@test_2024-11-06_08:30:05_daily 2406491805272097867 tank1/src " 

1007 "/foo tank1/src/foo@test_2024-10-06_08:30:04_daily 1730878205 24576`\n\n" 

1008 "If the TSV output file contains zero lines starting with the prefix 'src' and zero lines starting with " 

1009 "the prefix 'dst' then no source snapshots are missing on the destination, and no destination " 

1010 "snapshots are missing on the source, indicating that the periodic replication and pruning jobs perform " 

1011 "as expected. The TSV output is sorted by rel_dataset, and by ZFS creation time within each rel_dataset " 

1012 "- the first and last line prefixed with 'all' contains the metadata of the oldest and latest common " 

1013 "snapshot, respectively. Third party tools can use this info for post-processing, for example using " 

1014 "custom scripts using 'csplit' or duckdb analytics queries.\n\n" 

1015 "The --compare-snapshot-lists option also directly logs various summary stats, such as the metadata of " 

1016 "the latest common snapshot, latest snapshots and oldest snapshots, as well as the time diff between the " 

1017 "latest common snapshot and latest snapshot only in src (and only in dst), as well as how many src " 

1018 "snapshots and how many GB of data are missing on dst, etc.\n\n" 

1019 "*Note*: Consider omitting the 'all' flag to reduce noise and instead focus on missing snapshots only, " 

1020 "like so: --compare-snapshot-lists=src+dst \n\n" 

1021 "*Note*: The source can also be an empty dataset, such as the hardcoded virtual dataset named " 

1022 f"'{DUMMY_DATASET}'.\n\n" 

1023 "*Note*: --compare-snapshot-lists is typically *much* faster than standard 'zfs list -t snapshot' CLI " 

1024 "usage because the former issues requests with a higher degree of parallelism than the latter. The " 

1025 "degree is configurable with the --threads option (see below).\n\n") 

1026 parser.add_argument( 

1027 "--cache-snapshots", action="store_true",