diff --git a/PKG-INFO b/PKG-INFO index 9f072c2..49c38a0 100644 --- a/PKG-INFO +++ b/PKG-INFO @@ -1,32 +1,32 @@ Metadata-Version: 2.1 Name: swh.scheduler -Version: 0.11.0 +Version: 0.12.0 Summary: Software Heritage Scheduler Home-page: https://forge.softwareheritage.org/diffusion/DSCH/ Author: Software Heritage developers Author-email: swh-devel@inria.fr License: UNKNOWN Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest Project-URL: Funding, https://www.softwareheritage.org/donate Project-URL: Source, https://forge.softwareheritage.org/source/swh-scheduler Project-URL: Documentation, https://docs.softwareheritage.org/devel/swh-scheduler/ Description: swh-scheduler ============= Job scheduler for the Software Heritage project. Task manager for asynchronous/delayed tasks, used for both recurrent (e.g., listing a forge, loading new stuff from a Git repository) and one-off activities (e.g., loading a specific version of a source package). Platform: UNKNOWN Classifier: Programming Language :: Python :: 3 Classifier: Intended Audience :: Developers Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3) Classifier: Operating System :: OS Independent Classifier: Development Status :: 5 - Production/Stable Requires-Python: >=3.7 Description-Content-Type: text/markdown Provides-Extra: testing Provides-Extra: journal Provides-Extra: simulator diff --git a/debian/changelog b/debian/changelog index 405aec9..6dd2387 100644 --- a/debian/changelog +++ b/debian/changelog @@ -1,933 +1,937 @@ -swh-scheduler (0.11.0-1~swh1~bpo10+1) buster-swh; urgency=medium +swh-scheduler (0.12.0-1~swh1) unstable-swh; urgency=medium - * Rebuild for buster-swh + * New upstream release 0.12.0 - (tagged by Antoine R. Dumont + (@ardumont) on 2021-04-15 13:31:30 + +0200) + * Upstream changes: - v0.12.0 - Route priority tasks to + dedicated save code now queues - Fix various Sphinx warnings - -- Software Heritage autobuilder (on jenkins-debian1) Wed, 14 Apr 2021 16:21:24 +0000 + -- Software Heritage autobuilder (on jenkins-debian1) Thu, 15 Apr 2021 11:36:13 +0000 swh-scheduler (0.11.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.11.0 - (tagged by Antoine R. Dumont (@ardumont) on 2021-04-14 18:15:53 +0200) * Upstream changes: - v0.11.0 - separate-queues backend: Open endpoints to peek/grab tasks with any priority - Make origin_visit_stats_get return results from all pages - journal client: Filter out status messages without type - Simplify max_date() - journal_client: Fix date computations for (un)eventful visits - journal_client: Deal with failed status message -- Software Heritage autobuilder (on jenkins-debian1) Wed, 14 Apr 2021 16:19:31 +0000 swh-scheduler (0.10.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.10.0 - (tagged by Nicolas Dandrimont on 2021-02-03 22:53:20 +0100) * Upstream changes: - Release swh.scheduler 0.10.0 - Eagerly acknowledge celery tasks - Loads of simulator improvements - grab_next_visits: - clean up query building - account for schedule time to avoid rescheduling visits too fast - allow overriding the scheduling timestamp for the simulator -- Software Heritage autobuilder (on jenkins-debian1) Wed, 03 Feb 2021 22:10:13 +0000 swh-scheduler (0.9.2-1~swh1) unstable-swh; urgency=medium * New upstream release 0.9.2 - (tagged by Antoine Lambert on 2021-01-25 16:27:41 +0100) * Upstream changes: - version 0.9.2 -- Software Heritage autobuilder (on jenkins-debian1) Mon, 25 Jan 2021 15:31:21 +0000 swh-scheduler (0.9.1-1~swh1) unstable-swh; urgency=medium * New upstream release 0.9.1 - (tagged by Vincent SELLIER on 2021-01-21 19:20:33 +0100) * Upstream changes: - v0.9.1 - * Solve uneventful/eventful with unordered messages with snapshots - * Do not consider duplicated messages as uneventful event - * Reorganize grab_next_visits tests to better check sorting behavior -- Software Heritage autobuilder (on jenkins-debian1) Thu, 21 Jan 2021 18:28:00 +0000 swh-scheduler (0.9.0-1~swh2) unstable-swh; urgency=medium * Bump new release to unstuck packaging -- Antoine R. Dumont (@ardumont) Thu, 21 Jan 2021 13:20:14 +0000 swh-scheduler (0.9.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.9.0 - (tagged by Antoine R. Dumont (@ardumont) on 2021-01-21 11:54:47 +0100) * Upstream changes: - v0.9.0 - Populate origin_visit_stats table out of the origin_visit_status topic - Introduce a scheduler policy simulator (old task-based scheduler, ...) - Implement basic aggregated metrics on listed origins - scheduler.cli.journal: Add `swh scheduler journal-client` cli - Filter origins by visit type when scheduling the next visits - Introduce a `swh scheduler origin schedule-next` cli - Introduce a `swh scheduler origin grab-next` cli - Add an new origin visit stats model object and related backend api - Implement a basic endpoint for getting the next origins to visit - doc: Add a cli section to the doc -- Software Heritage autobuilder (on jenkins-debian1) Thu, 21 Jan 2021 11:00:29 +0000 swh-scheduler (0.8.2-1~swh2) unstable-swh; urgency=medium * Bump dependency -- Antoine R. Dumont (@ardumont) Tue, 08 Dec 2020 09:29:26 +0000 swh-scheduler (0.8.2-1~swh1) unstable-swh; urgency=medium * New upstream release 0.8.2 - (tagged by Antoine R. Dumont (@ardumont) on 2020-12-07 09:52:28 +0100) * Upstream changes: - v0.8.2 - requirement: Adapt celery requirements - Replace usage of arrow datetime objects in favor of pure datetime ones - Stop using the deprecated configuration scheme - cli.task_type: All task_type clis without a scheduler should raise -- Software Heritage autobuilder (on jenkins-debian1) Mon, 07 Dec 2020 08:55:39 +0000 swh-scheduler (0.8.1-1~swh1) unstable-swh; urgency=medium * New upstream release 0.8.1 - (tagged by Antoine R. Dumont (@ardumont) on 2020-11-24 14:13:36 +0100) * Upstream changes: - v0.8.1 - conftest: Reference swh.core.db.pytest_plugin -- Software Heritage autobuilder (on jenkins-debian1) Tue, 24 Nov 2020 13:16:08 +0000 swh-scheduler (0.8.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.8.0 - (tagged by Antoine R. Dumont (@ardumont) on 2020-11-23 13:42:05 +0100) * Upstream changes: - v0.8.0 - requirements-test.txt: Drop no longer needed pytest-postgresql requirement - scheduler.pytest_plugin: Make scheduler tests faster -- Software Heritage autobuilder (on jenkins-debian1) Mon, 23 Nov 2020 12:44:40 +0000 swh-scheduler (0.7.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.7.0 - (tagged by Antoine R. Dumont (@ardumont) on 2020-10-19 09:30:36 +0200) * Upstream changes: - v0.7.0 - scheduler: Type and unify get_scheduler factory with other factories - pytest_plugin: Explicitly name the scheduler test db differently - test_server: Simplify exception manipulations - tox.ini: pin black to the pre-commit version (19.10b0) to avoid flip-flops -- Software Heritage autobuilder (on jenkins-debian1) Mon, 19 Oct 2020 07:33:54 +0000 swh-scheduler (0.6.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.6.0 - (tagged by David Douard on 2020-09-25 12:03:33 +0200) * Upstream changes: - v0.6.0 -- Software Heritage autobuilder (on jenkins-debian1) Fri, 25 Sep 2020 10:06:32 +0000 swh-scheduler (0.5.3-1~swh1) unstable-swh; urgency=medium * New upstream release 0.5.3 - (tagged by Nicolas Dandrimont on 2020-09-24 17:49:27 +0200) * Upstream changes: - Release swh.scheduler v0.5.3 - Improve swh cli startup time - Add isort and update flake8 - Improve pytest execution time - Support recent kombu versions -- Software Heritage autobuilder (on jenkins-debian1) Thu, 24 Sep 2020 15:53:25 +0000 swh-scheduler (0.5.2-1~swh1) unstable-swh; urgency=medium * New upstream release 0.5.2 - (tagged by Antoine R. Dumont (@ardumont) on 2020-07-10 13:01:48 +0200) * Upstream changes: - v0.5.2 - Do no expose pytest-plugin through setuptools, let modules require it when needed -- Software Heritage autobuilder (on jenkins-debian1) Fri, 10 Jul 2020 11:08:30 +0000 swh-scheduler (0.5.1-1~swh1) unstable-swh; urgency=medium * New upstream release 0.5.1 - (tagged by Nicolas Dandrimont on 2020-07-09 10:18:03 +0200) * Upstream changes: - Release swh.scheduler 0.5.1 - Drop dependency on future (not needed anymore) -- Software Heritage autobuilder (on jenkins-debian1) Thu, 09 Jul 2020 09:51:38 +0000 swh-scheduler (0.5.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.5.0 - (tagged by Nicolas Dandrimont on 2020-07-09 10:16:57 +0200) * Upstream changes: - Release swh.scheduler v0.5.0 - Move celery fixtures to the pytest plugin -- Software Heritage autobuilder (on jenkins-debian1) Thu, 09 Jul 2020 08:20:42 +0000 swh-scheduler (0.4.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.4.0 - (tagged by Nicolas Dandrimont on 2020-07-06 16:47:28 +0200) * Upstream changes: - Release swh.scheduler 0.4.0 - Extract pytest fixtures to a pytest plugin -- Software Heritage autobuilder (on jenkins-debian1) Mon, 06 Jul 2020 14:52:42 +0000 swh-scheduler (0.3.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.3.0 - (tagged by Nicolas Dandrimont on 2020-07-06 12:18:28 +0200) * Upstream changes: - Release swh.scheduler 0.3.0 - Add get_listed_origins endpoint -- Software Heritage autobuilder (on jenkins-debian1) Mon, 06 Jul 2020 10:23:31 +0000 swh-scheduler (0.2.2-1~swh1) unstable-swh; urgency=medium * New upstream release 0.2.2 - (tagged by Nicolas Dandrimont on 2020-06-22 14:03:34 +0200) * Upstream changes: - Release swh.scheduler 0.2.2 - Re- introduce root endpoint for the RPC server -- Software Heritage autobuilder (on jenkins-debian1) Mon, 22 Jun 2020 12:07:05 +0000 swh-scheduler (0.2.1-1~swh1) unstable-swh; urgency=medium [ Nicolas Dandrimont ] * Force celery >= 4.3 [ Software Heritage autobuilder (on jenkins-debian1) ] * New upstream release 0.2.1 - (tagged by Nicolas Dandrimont on 2020-06-22 12:09:32 +0200) * Upstream changes: - Release swh.scheduler 0.2.1 - Bump celery requirement to 4.3+ -- Software Heritage autobuilder (on jenkins-debian1) Mon, 22 Jun 2020 10:12:50 +0000 swh-scheduler (0.2.0-1~swh1) unstable-swh; urgency=medium [ Nicolas Dandrimont ] * Switch from vcversioner to setuptools-scm * wrap-and-sort [ Software Heritage autobuilder (on jenkins-debian1) ] * New upstream release 0.2.0 - (tagged by Nicolas Dandrimont on 2020-06-22 10:33:11 +0200) * Upstream changes: - Release swh.scheduler 0.2.0 - Implement storage of lister and listed origin information - Add swh scheduler celery-monitor command - Overhaul RPC to use automatic generation -- Software Heritage autobuilder (on jenkins-debian1) Mon, 22 Jun 2020 08:36:49 +0000 swh-scheduler (0.1.1-1~swh1) unstable-swh; urgency=medium * New upstream release 0.1.1 - (tagged by Nicolas Dandrimont on 2020-06-03 11:34:19 +0200) * Upstream changes: - Release swh.scheduler v0.1.1 - Add missing dependency on future for celery 4.4.4 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 03 Jun 2020 09:39:25 +0000 swh-scheduler (0.1.0-1~swh1) unstable-swh; urgency=medium * New upstream release 0.1.0 - (tagged by Nicolas Dandrimont on 2020-05-19 11:48:34 +0200) * Upstream changes: - Release swh.scheduler v0.1.0 - Blacken source code - Disable azure http logspam - Only schedule tasks when the buffer is somewhat empty -- Software Heritage autobuilder (on jenkins-debian1) Tue, 19 May 2020 09:52:31 +0000 swh-scheduler (0.0.72-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.72 - (tagged by Nicolas Dandrimont on 2020-03-23 13:07:38 +0100) * Upstream changes: - Release swh.scheduler v0.0.72 - Update instantiation of storage in tests - ensure that create_task_type is idempotent - introduce new listener based on pika -- Software Heritage autobuilder (on jenkins-debian1) Mon, 23 Mar 2020 12:12:00 +0000 swh-scheduler (0.0.71-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.71 - (tagged by Antoine R. Dumont (@ardumont) on 2020-01-23 14:24:56 +0100) * Upstream changes: - v0.0.71 - sentry: Fix initialization init_sentry call -- Software Heritage autobuilder (on jenkins-debian1) Thu, 23 Jan 2020 13:29:33 +0000 swh-scheduler (0.0.70-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.70 - (tagged by Antoine R. Dumont (@ardumont) on 2020-01-23 13:43:35 +0100) * Upstream changes: - v0.0.70 - Use swh.core.sentry instead of calling sentry_sdk.init directly - backend_es: Fix configuration mapping -- Software Heritage autobuilder (on jenkins-debian1) Thu, 23 Jan 2020 12:47:43 +0000 swh-scheduler (0.0.69-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.69 - (tagged by Antoine R. Dumont (@ardumont) on 2019-12-17 16:00:24 +0100) * Upstream changes: - v0.0.69 - Fix scheduler's archive task cli - Make the filter task endpoint a paginated endpoint - Add coverage on the archive task cli -- Software Heritage autobuilder (on jenkins-debian1) Tue, 17 Dec 2019 15:04:48 +0000 swh-scheduler (0.0.68-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.68 - (tagged by Antoine R. Dumont (@ardumont) on 2019-12-17 15:28:13 +0100) * Upstream changes: - v0.0.68 - Fix scheduler's archive task cli - Make the filter task endpoint a paginated endpoint - Add coverage on the archive task cli -- Software Heritage autobuilder (on jenkins-debian1) Tue, 17 Dec 2019 14:33:33 +0000 swh-scheduler (0.0.67-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.67 - (tagged by Antoine R. Dumont (@ardumont) on 2019-12-17 14:33:36 +0100) * Upstream changes: - v0.0.67 - Fix scheduler's archive task cli - Make the filter task endpoint a paginated endpoint - Add coverage on the archive task cli -- Software Heritage autobuilder (on jenkins-debian1) Tue, 17 Dec 2019 13:38:03 +0000 swh-scheduler (0.0.66-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.66 - (tagged by Nicolas Dandrimont on 2019-12-17 12:04:20 +0100) * Upstream changes: - Release swh.scheduler v0.0.66 - initialize sentry on celery worker startup - improve task archival endpoints in backend api -- Software Heritage autobuilder (on jenkins-debian1) Tue, 17 Dec 2019 11:08:25 +0000 swh-scheduler (0.0.65-1~swh2) unstable-swh; urgency=medium * Add pytest-mock build-dependency. -- Nicolas Dandrimont Fri, 13 Dec 2019 11:57:41 +0100 swh-scheduler (0.0.65-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.65 - (tagged by Nicolas Dandrimont on 2019-12-13 11:45:55 +0100) * Upstream changes: - Release swh.scheduler v0.0.65 - Drop the scheduler updater - Add a statsd probe for task execution timestamps - Add listener and runner statsd probes - CLI updates - Python packaging housekeeping -- Software Heritage autobuilder (on jenkins-debian1) Fri, 13 Dec 2019 10:54:31 +0000 swh-scheduler (0.0.64-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.64 - (tagged by Antoine R. Dumont (@ardumont) on 2019-11-20 14:26:00 +0100) * Upstream changes: - v0.0.64 - req-swh*: Remove old package loader backend names -- Software Heritage autobuilder (on jenkins-debian1) Wed, 20 Nov 2019 13:29:37 +0000 swh-scheduler (0.0.63-1~swh2) unstable-swh; urgency=medium * Update build dependency -- Antoine R. Dumont (@ardumont) Tue, 19 Nov 2019 17:07:40 +0100 swh-scheduler (0.0.63-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.63 - (tagged by Antoine R. Dumont (@ardumont) on 2019-11-19 14:09:12 +0100) * Upstream changes: - v0.0.63 - swh.scheduler.cli: Add `swh scheduler task-type register` cli - Use the shared_task decorator instead of binding to a specific celery app - celery/tests: mostly revert e770eb30 to fix celery app initialization in tests -- Software Heritage autobuilder (on jenkins-debian1) Tue, 19 Nov 2019 13:14:59 +0000 swh-scheduler (0.0.62-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.62 - (tagged by Antoine R. Dumont (@ardumont) on 2019-10-18 13:39:27 +0200) * Upstream changes: - v0.0.62 - celery_backend.config: Make JournalHandler import optional - tests: rewrite tests using pytest fixtures -- Software Heritage autobuilder (on jenkins-debian1) Fri, 18 Oct 2019 11:46:26 +0000 swh-scheduler (0.0.61-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.61 - (tagged by Nicolas Dandrimont on 2019-10-07 16:33:17 +0200) * Upstream changes: - Release swh.scheduler v0.0.61 - Remove bogus dict.get(default=) statement -- Software Heritage autobuilder (on jenkins-debian1) Mon, 07 Oct 2019 14:37:37 +0000 swh-scheduler (0.0.60-1~swh2) unstable-swh; urgency=medium * Force postgresql executable to a pg_ctl that exists when running tests. -- Nicolas Dandrimont Tue, 01 Oct 2019 18:14:39 +0200 swh-scheduler (0.0.60-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.60 - (tagged by Stefano Zacchiroli on 2019-10-01 13:13:13 +0200) * Upstream changes: - v0.0.60 - * tox: anticipate mypy run to just after flake8 - * init.py: switch to documented way of extending path - * tox.ini: add mypy section - * typing: minimal changes to make a no-op mypy run pass - * fix typo in docstring and sample file name - * admin CLI: drop obsolete backward compatibility aliases - * click "required" param wants bool, not int -- Software Heritage autobuilder (on jenkins-debian1) Tue, 01 Oct 2019 11:22:43 +0000 swh-scheduler (0.0.59-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.59 - (tagged by David Douard on 2019-09-04 16:08:27 +0200) * Upstream changes: - v0.0.59 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 04 Sep 2019 14:11:48 +0000 swh-scheduler (0.0.58-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.58 - (tagged by Antoine R. Dumont (@ardumont) on 2019-09-03 10:19:34 +0200) * Upstream changes: - v0.0.58 - celery: auto add tasks declared in the swh.workers entry point in task_modules - api/client: use RPCClient instead of deprecated SWHRemoteAPI - Make schedule_origins use origin urls instead of ids in task arguments. - docs: add code of conduct document - docs: very beginning of a practical documentation on the scheduler - config: Add a pre-commit config file - data: Insert new cgit instance lister task - data: Insert load-tar task-type -- Software Heritage autobuilder (on jenkins-debian1) Tue, 03 Sep 2019 08:28:19 +0000 swh-scheduler (0.0.57-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.57 - (tagged by David Douard on 2019-06-26 14:56:32 +0200) * Upstream changes: - v0.0.57 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 26 Jun 2019 13:05:20 +0000 swh-scheduler (0.0.56-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.56 - (tagged by Nicolas Dandrimont on 2019-05-07 18:16:20 +0200) * Upstream changes: - listener: Release the db object after using it - This is the contract that get_db/put_db is supposed to conform to. -- Software Heritage autobuilder (on jenkins-debian1) Tue, 14 May 2019 12:40:09 +0000 swh-scheduler (0.0.55-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.55 - (tagged by Antoine Lambert on 2019-05-06 11:47:43 +0200) * Upstream changes: - version 0.0.55 -- Software Heritage autobuilder (on jenkins-debian1) Mon, 06 May 2019 09:54:51 +0000 swh-scheduler (0.0.54-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.54 - (tagged by Antoine R. Dumont (@ardumont) on 2019-04-11 11:33:40 +0200) * Upstream changes: - v0.0.54 - cli_utils: Use yaml.safe_load instead of yaml.load - Fix support of latest versions of swh- core and psycopg2 - sql/data: Add npm related task types -- Software Heritage autobuilder (on jenkins-debian1) Thu, 11 Apr 2019 09:40:14 +0000 swh-scheduler (0.0.53-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.53 - (tagged by Antoine Lambert on 2019-04-04 16:45:56 +0200) * Upstream changes: - version 0.0.53 -- Software Heritage autobuilder (on jenkins-debian1) Thu, 04 Apr 2019 14:55:20 +0000 swh-scheduler (0.0.52-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.52 - (tagged by Nicolas Dandrimont on 2019-04-03 10:54:06 +0200) * Upstream changes: - Release swh.scheduler v0.0.52 - Move to result_serializer = json to work around celery 4.3 bug - Fix db initialization -- Software Heritage autobuilder (on jenkins-debian1) Wed, 03 Apr 2019 08:59:00 +0000 swh-scheduler (0.0.51-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.51 - (tagged by Antoine R. Dumont (@ardumont) on 2019-03-22 12:09:22 +0100) * Upstream changes: - v0.0.51 - requirements.txt: Remove kombu dependency -- Software Heritage autobuilder (on jenkins-debian1) Fri, 22 Mar 2019 11:16:06 +0000 swh-scheduler (0.0.50-1~swh2) unstable-swh; urgency=medium * Update build- and runtime dependencies -- Nicolas Dandrimont Fri, 15 Mar 2019 18:24:11 +0100 swh-scheduler (0.0.50-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.50 - (tagged by Nicolas Dandrimont on 2019-03-15 18:07:24 +0100) * Upstream changes: - Release swh.scheduler v0.0.50 - Add an explicit log target for stdout and/or journald - Avoid useless log lines - Improve test coverage - Add support for non- string options in the CLI -- Software Heritage autobuilder (on jenkins-debian1) Fri, 15 Mar 2019 17:16:03 +0000 swh-scheduler (0.0.49-1~swh2) unstable-swh; urgency=medium * Export LC_ALL=C.UTF-8 -- Nicolas Dandrimont Thu, 14 Mar 2019 13:42:24 +0100 swh-scheduler (0.0.49-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.49 - (tagged by Nicolas Dandrimont on 2019-03-03 08:48:04 +0100) * Upstream changes: - Release swh.scheduler v0.0.49 - various fixes around celery behavior - move wsgi endpoint to a separate module - add tests for the CLI -- Software Heritage autobuilder (on jenkins-debian1) Sun, 03 Mar 2019 07:55:41 +0000 swh-scheduler (0.0.48-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.48 - (tagged by Antoine R. Dumont (@ardumont) on 2019-02-22 16:11:51 +0100) * Upstream changes: - v0.0.48 - Fix comment on main scheduler schema -- Software Heritage autobuilder (on jenkins-debian1) Fri, 22 Feb 2019 15:17:20 +0000 swh-scheduler (0.0.47-1~swh2) unstable-swh; urgency=low * Upstream release to fix build dependencies issue -- Antoine Romain Dumont (@ardumont) Thu, 21 Feb 2019 15:41:24 +0100 swh-scheduler (0.0.47-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.47 - (tagged by Valentin Lorentz on 2019-02-20 16:53:20 +0100) * Upstream changes: - Fix crash of SchedulerBackend.search_tasks when no argument is given. -- Software Heritage autobuilder (on jenkins-debian1) Thu, 21 Feb 2019 09:13:07 +0000 swh-scheduler (0.0.46-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.46 - (tagged by Antoine R. Dumont (@ardumont) on 2019-02-15 15:05:47 +0100) * Upstream changes: - v0.0.46 - scheduler.task: Remove no longer used Task class -- Software Heritage autobuilder (on jenkins-debian1) Fri, 15 Feb 2019 14:15:26 +0000 swh-scheduler (0.0.45-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.45 - (tagged by Antoine R. Dumont (@ardumont) on 2019-02-15 10:43:07 +0100) * Upstream changes: - v0.0.45 - celery_backend/config: Fix loglevel for amqp module -- Software Heritage autobuilder (on jenkins-debian1) Fri, 15 Feb 2019 09:48:25 +0000 swh-scheduler (0.0.44-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.44 - (tagged by Antoine R. Dumont (@ardumont) on 2019-02-13 16:29:05 +0100) * Upstream changes: - v0.0.44 - swh-scheduler-api: Fix configuration read too many times -- Software Heritage autobuilder (on jenkins-debian1) Wed, 13 Feb 2019 15:34:34 +0000 swh-scheduler (0.0.43-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.43 - (tagged by David Douard on 2019-02-13 15:27:27 +0100) * Upstream changes: - v0.0.43 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 13 Feb 2019 14:46:59 +0000 swh-scheduler (0.0.42-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.42 - (tagged by Antoine R. Dumont (@ardumont) on 2019-02-11 14:28:10 +0100) * Upstream changes: - v0.0.42 - Fix dependency requirements for hypothesis -- Software Heritage autobuilder (on jenkins-debian1) Mon, 11 Feb 2019 13:33:48 +0000 swh-scheduler (0.0.41-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.41 - (tagged by David Douard on 2019-02-06 15:25:56 +0100) * Upstream changes: - v0.0.41 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 06 Feb 2019 15:33:04 +0000 swh-scheduler (0.0.40-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.40 - (tagged by Antoine R. Dumont (@ardumont) on 2019-01-28 16:24:04 +0100) * Upstream changes: - v0.0.40 - swh.scheduler.tests: Mark db tests as such - Force tox environment to C.UTF-8 locale - Add debug logging in the SWHTask class -- Software Heritage autobuilder (on jenkins-debian1) Mon, 28 Jan 2019 15:30:41 +0000 swh-scheduler (0.0.39-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.39 - (tagged by David Douard on 2019-01-16 13:37:58 +0100) * Upstream changes: - v0.0.39 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 16 Jan 2019 12:42:37 +0000 swh-scheduler (0.0.38-1~swh1) unstable-swh; urgency=medium * New upstream release 0.0.38 - (tagged by David Douard on 2018-12-20 14:39:59 +0100) * Upstream changes: - v0.0.38 -- Software Heritage autobuilder (on jenkins-debian1) Wed, 09 Jan 2019 18:32:14 +0000 swh-scheduler (0.0.35-1~swh1) unstable-swh; urgency=medium * v0.0.35 * tests: Add SchedulerTestFixture * swh.scheduler.utils: Allow to add more task information * sql/40-swh-data: Update new indexer task types for local db -- Antoine R. Dumont (@ardumont) Mon, 29 Oct 2018 10:07:08 +0100 swh-scheduler (0.0.34-1~swh1) unstable-swh; urgency=medium * v0.0.34 * Finalize pytest migration -- Antoine R. Dumont (@ardumont) Thu, 25 Oct 2018 17:52:03 +0200 swh-scheduler (0.0.33-1~swh1) unstable-swh; urgency=medium * v0.0.33 -- David Douard Thu, 25 Oct 2018 16:03:16 +0200 swh-scheduler (0.0.32-1~swh1) unstable-swh; urgency=medium * v0.0.32 * tests: Add celery fixture to ease tests * tests: make tests use sql/ files from the package * tests: Starting migration towards pytest * listener: Make the listener code compatible with new celery (debian buster) * Make swh_scheduler_create_tasks_from_temp use indexes * setup: prepare for pypi upload * docs: add a simple README file -- Antoine R. Dumont (@ardumont) Mon, 22 Oct 2018 15:37:51 +0200 swh-scheduler (0.0.31-1~swh1) unstable-swh; urgency=medium * v0.0.31 * sql/swh-scheduler: Make the create_tasks call idempotent * swh.scheduler.utils: Open create_task_dict function * sql/scheduler-data: Add lister gitlab task types * sql/scheduler-data: Reference the existing production lister data * swh.scheduler.backend_es: Open sniffing options -- Antoine R. Dumont (@ardumont) Tue, 31 Jul 2018 06:55:39 +0200 swh-scheduler (0.0.30-1~swh1) unstable-swh; urgency=medium * v0.0.30 * swh-scheduler-schema.sql: Archive disabled oneshot tasks as well * swh.scheduler.cli: Add policy to pretty printing task routine * swh.scheduler.cli: Fix broken cli list-pending since api change -- Antoine R. Dumont (@ardumont) Fri, 22 Jun 2018 18:07:02 +0200 swh-scheduler (0.0.29-1~swh1) unstable-swh; urgency=medium * v0.0.29 * swh.scheduler.cli: Change archival period to rolling month - 1 week * swh.scheduler.updater.writer: Force filter resolution to list * swh.scheduler.cli: Change default archival period to current month * swh.scheduler.cli: Improve logging message * swh.scheduler.updater.backend: Adapt configuration path accordingly -- Antoine R. Dumont (@ardumont) Thu, 31 May 2018 11:42:51 +0200 swh-scheduler (0.0.28-1~swh1) unstable-swh; urgency=medium * v0.0.28 * Fix wrong runtime dependencies -- Antoine R. Dumont (@ardumont) Tue, 29 May 2018 14:12:15 +0200 swh-scheduler (0.0.27-1~swh1) unstable-swh; urgency=medium * v0.0.27 * scheduler: Deal with priority in tasks * scheduler-update: new package python3-swh.scheduler.updater * Contains tools in charge of consuming events from arbitrary sources * and update the scheduler db -- Antoine R. Dumont (@ardumont) Tue, 29 May 2018 12:27:34 +0200 swh-scheduler (0.0.26-1~swh1) unstable-swh; urgency=medium * v0.0.26 * swh.scheduler: Fix package build * swh.scheduler.tests: Test remote scheduler api as well * swh.scheduler: Add tests around removing archivable tasks * swh.scheduler: Add tests around filtering archivable tasks * swh-scheduler-schema: Fix unneeded drop instructions * swh.scheduler.cli: Improve docstring * swh.scheduler.cli: Permit to specify the backend to use in cli * swh.scheduler.api: Bootstrap scheduler's remote api * swh.scheduler: Use `get_scheduler` api to instantiate a scheduler * swh.scheduler.backend: Fix docstring -- Antoine R. Dumont (@ardumont) Thu, 26 Apr 2018 17:34:07 +0200 swh-scheduler (0.0.25-1~swh1) unstable-swh; urgency=medium * v0.0.25 * swh.scheduler.cli.archive: Index arguments.kwargs as text -- Antoine R. Dumont (@ardumont) Wed, 18 Apr 2018 12:34:43 +0200 swh-scheduler (0.0.24-1~swh1) unstable-swh; urgency=medium * v0.0.24 * data/template: Do not index the arguments field (it's in _source) * data/README: Add a small readme to explain es install step * swh.scheduler.cli: Add a bulk index flag to separate read from index -- Antoine R. Dumont (@ardumont) Fri, 13 Apr 2018 14:55:32 +0200 swh-scheduler (0.0.23-1~swh1) unstable-swh; urgency=medium * swh.scheduler.cli.archive: Delete only completely indexed tasks * Prior to this commit, it could happen that we removed tasks even * though we did not yet index associated task_run. * Related T986 -- Antoine R. Dumont (@ardumont) Tue, 10 Apr 2018 17:43:07 +0200 swh-scheduler (0.0.22-1~swh1) unstable-swh; urgency=medium * v0.0.22 * Update to a more recent python3-elasticsearch client -- Antoine R. Dumont (@ardumont) Mon, 09 Apr 2018 16:09:16 +0200 swh-scheduler (0.0.21-1~swh1) unstable-swh; urgency=medium * v0.0.21 * Adapt default configuration * Fix typo in configuration variable name -- Antoine R. Dumont (@ardumont) Fri, 30 Mar 2018 15:02:55 +0200 swh-scheduler (0.0.20-1~swh1) unstable-swh; urgency=medium * v0.0.20 * swh.scheduler.cli.archive: Open completed oneshot or disabled * recurring tasks archival endpoint * swh.core.serializer: Move to msgpack serialization format * swh.scheduler.cli: Unify pretty print output * sql/data: Add new task type for loading mercurial dump * swh.scheduler.cli: Add sample use case for the scheduling cli * swh.scheduler.cli: Open policy column to the scheduling cli * swh.scheduler.cli: Open the delimiter option as cli argument * Fix issue when updating task-type without any retry delay defined * swh-scheduler/data: Add new oneshot scheduling load-mercurial task * backend: fix default scheduling_db value for consistency * backend: doc: fix return value of create_tasks -- Antoine R. Dumont (@ardumont) Fri, 30 Mar 2018 11:44:18 +0200 swh-scheduler (0.0.19-1~swh1) unstable-swh; urgency=medium * v0.0.19 * swh.scheduler.utils: Open utility function to create oneshot task -- Antoine R. Dumont (@ardumont) Wed, 29 Nov 2017 12:51:15 +0100 swh-scheduler (0.0.18-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler v0.0.18 * Celery 4 compatibility -- Nicolas Dandrimont Wed, 08 Nov 2017 17:06:22 +0100 swh-scheduler (0.0.17-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler version 0.0.17 * Update packaging runes -- Nicolas Dandrimont Thu, 12 Oct 2017 18:49:02 +0200 swh-scheduler (0.0.16-1~swh1) unstable-swh; urgency=medium * Release swh-scheduler v0.0.16 * add some tests * implement one-shot tasks * implement retry on temporary failure -- Nicolas Dandrimont Mon, 07 Aug 2017 18:44:03 +0200 swh-scheduler (0.0.15-1~swh1) unstable-swh; urgency=medium * Release swh-scheduler v0.0.15 * Add some methods to get the length of task queues * worker: Show logs on stdout if loglevel = debug -- Nicolas Dandrimont Mon, 19 Jun 2017 19:44:56 +0200 swh-scheduler (0.0.14-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler 0.0.14 * Make the return value of tasks available in the listener -- Nicolas Dandrimont Mon, 12 Jun 2017 17:50:32 +0200 swh-scheduler (0.0.13-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler v0.0.13 * Use systemd for logging rather than PostgreSQL -- Nicolas Dandrimont Fri, 07 Apr 2017 11:57:50 +0200 swh-scheduler (0.0.12-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler v0.0.12 * Only log to database if the configuration is present -- Nicolas Dandrimont Thu, 09 Mar 2017 11:12:45 +0100 swh-scheduler (0.0.11-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler v0.0.11 * add utils.get_task -- Nicolas Dandrimont Tue, 14 Feb 2017 19:49:34 +0100 swh-scheduler (0.0.10-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler v0.0.10 * Allow disabling tasks -- Nicolas Dandrimont Thu, 20 Oct 2016 17:20:17 +0200 swh-scheduler (0.0.9-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler v0.0.9 * Revert management of one shot tasks * Add possibility of launching several worker instances -- Nicolas Dandrimont Fri, 02 Sep 2016 17:09:18 +0200 swh-scheduler (0.0.7-1~swh1) unstable-swh; urgency=medium * v0.0.7 * Add oneshot task -- Antoine R. Dumont (@ardumont) Fri, 01 Jul 2016 16:42:45 +0200 swh-scheduler (0.0.6-1~swh1) unstable-swh; urgency=medium * Release swh-scheduler v0.0.6 * More reliability and efficiency when scheduling a lot of tasks -- Nicolas Dandrimont Wed, 24 Feb 2016 18:46:57 +0100 swh-scheduler (0.0.5-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler v0.0.5 * Use copy for task mass-scheduling -- Nicolas Dandrimont Wed, 24 Feb 2016 12:13:38 +0100 swh-scheduler (0.0.4-1~swh1) unstable-swh; urgency=medium * Release swh-scheduler v0.0.4 * general cleanup of the backend * use arrow instead of dateutil * add new cli program -- Nicolas Dandrimont Tue, 23 Feb 2016 17:46:04 +0100 swh-scheduler (0.0.3-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler version 0.0.3 * Implement the timestamp arguments to the task_run functions * Make the celery event listener use a reliable queue -- Nicolas Dandrimont Mon, 22 Feb 2016 15:14:28 +0100 swh-scheduler (0.0.2-1~swh1) unstable-swh; urgency=medium * Release swh.scheduler v0.0.2 * Multiple schema changes * Initial releases for the celery job runner and the event listener -- Nicolas Dandrimont Fri, 19 Feb 2016 18:50:47 +0100 swh-scheduler (0.0.1-1~swh1) unstable-swh; urgency=medium * Initial release * Release swh.scheduler v0.0.1 * Move swh.core.scheduling and swh.core.worker to swh.scheduler -- Nicolas Dandrimont Mon, 15 Feb 2016 11:07:30 +0100 diff --git a/docs/simulator.rst b/docs/simulator.rst index 979a389..35708b6 100644 --- a/docs/simulator.rst +++ b/docs/simulator.rst @@ -1,80 +1,80 @@ .. _swh-scheduler-simulator: Software Heritage Scheduler Simulator ===================================== This component simulates the interaction between the scheduling and loading infrastructure of Software Heritage. This allows quick(er) development of new task scheduling policies without having to wait for the actual infrastructure to perform (heavy) loading tasks. Simulator components -------------------- - real instance of the scheduler database - simulated task queues: replaces RabbitMQ with simple in-memory structures - simulated workers: replaces Celery with simple while loops - simulated load tasks: replaces loaders with noops that take a certain time, and generate synthetic OriginVisitStatus objects - simulated archive -> scheduler feedback loop: OriginVisitStatus objects are pushed to a simple queue which gets processed by the scheduler journal client's process function directly (instead of going through swh.storage and swh.journal (kafka)) In short, only the scheduler database and scheduler logic is kept; every other component (RabbitMQ, Celery, Kafka, SWH loaders, SWH storage) is either replaced with an barebones in-process utility, or removed entirely. Installing the simulator ------------------------ The simulator depends on SimPy and other specific libraries. To install them, please use: .. code-block:: bash pip install 'swh.scheduler[simulator]' Running the simulator --------------------- The simulator uses a real instance of the scheduler database, which is (at least for now) persistent across runs of the simulator. You need to set that up beforehand: .. code-block:: bash # if you want to use a temporary instance of postgresql eval `pifpaf run postgresql` # Set this variable for the simulator to know which db to connect to. pifpaf # sets other variables like PGPORT, PGHOST, ... export PGDATABASE=swh-scheduler # Create/initialize the scheduler database swh db create scheduler -d $PGDATABASE swh db init scheduler -d $PGDATABASE # This generates some data in the scheduler database. You can also feed the # database with more realistic data, e.g. from a lister or from a dump of the # production database. swh scheduler -d "dbname=$PGDATABASE" simulator fill-test-data # Run the simulator itself, interacting with the scheduler database you've # just seeded. swh scheduler -d "dbname=$PGDATABASE" simulator run --scheduler origin_scheduler Origin model ------------ The origin model is how we represent the behaviors of origins: when they are created/discovered, how many commits they get and when, and when they fail to load. For now it is only a simple approximation designed to exercise simple cases: origin creation/discovery, a continuous stream of commits, and failure if they have too many commits to load at once. -For details, see :py:`swh.scheduler.simulator.origins`. +For details, see :py:mod:`swh.scheduler.simulator.origins`. To keep the simulation fast enough, each origin's state is kept in memory, so the simulator process will linearly increase in memory usage as it runs. diff --git a/swh.scheduler.egg-info/PKG-INFO b/swh.scheduler.egg-info/PKG-INFO index 9f072c2..49c38a0 100644 --- a/swh.scheduler.egg-info/PKG-INFO +++ b/swh.scheduler.egg-info/PKG-INFO @@ -1,32 +1,32 @@ Metadata-Version: 2.1 Name: swh.scheduler -Version: 0.11.0 +Version: 0.12.0 Summary: Software Heritage Scheduler Home-page: https://forge.softwareheritage.org/diffusion/DSCH/ Author: Software Heritage developers Author-email: swh-devel@inria.fr License: UNKNOWN Project-URL: Bug Reports, https://forge.softwareheritage.org/maniphest Project-URL: Funding, https://www.softwareheritage.org/donate Project-URL: Source, https://forge.softwareheritage.org/source/swh-scheduler Project-URL: Documentation, https://docs.softwareheritage.org/devel/swh-scheduler/ Description: swh-scheduler ============= Job scheduler for the Software Heritage project. Task manager for asynchronous/delayed tasks, used for both recurrent (e.g., listing a forge, loading new stuff from a Git repository) and one-off activities (e.g., loading a specific version of a source package). Platform: UNKNOWN Classifier: Programming Language :: Python :: 3 Classifier: Intended Audience :: Developers Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3) Classifier: Operating System :: OS Independent Classifier: Development Status :: 5 - Production/Stable Requires-Python: >=3.7 Description-Content-Type: text/markdown Provides-Extra: testing Provides-Extra: journal Provides-Extra: simulator diff --git a/swh.scheduler.egg-info/SOURCES.txt b/swh.scheduler.egg-info/SOURCES.txt index 4f36b88..f9d7578 100644 --- a/swh.scheduler.egg-info/SOURCES.txt +++ b/swh.scheduler.egg-info/SOURCES.txt @@ -1,130 +1,131 @@ .gitignore .pre-commit-config.yaml AUTHORS CODE_OF_CONDUCT.md CONTRIBUTORS LICENSE LICENSE.Celery MANIFEST.in Makefile README.md conftest.py mypy.ini pyproject.toml pytest.ini requirements-journal.txt requirements-simulator.txt requirements-swh.txt requirements-test.txt requirements.txt setup.cfg setup.py tox.ini data/README.md data/elastic-template.json data/update-index-settings.json docs/.gitignore docs/Makefile docs/cli.rst docs/conf.py docs/index.rst docs/simulator.rst docs/_static/.placeholder docs/_templates/.placeholder sql/.gitignore sql/Makefile sql/updates/02.sql sql/updates/03.sql sql/updates/04.sql sql/updates/05.sql sql/updates/06.sql sql/updates/07.sql sql/updates/08.sql sql/updates/09.sql sql/updates/10.sql sql/updates/11.sql sql/updates/12.sql sql/updates/13.sql sql/updates/14.sql sql/updates/15.sql sql/updates/16.sql sql/updates/17.sql sql/updates/18.sql sql/updates/19.sql sql/updates/20.sql sql/updates/23.sql sql/updates/24.sql sql/updates/25.sql sql/updates/26.sql swh/__init__.py swh.scheduler.egg-info/PKG-INFO swh.scheduler.egg-info/SOURCES.txt swh.scheduler.egg-info/dependency_links.txt swh.scheduler.egg-info/entry_points.txt swh.scheduler.egg-info/requires.txt swh.scheduler.egg-info/top_level.txt swh/scheduler/__init__.py swh/scheduler/backend.py swh/scheduler/backend_es.py swh/scheduler/cli_utils.py swh/scheduler/elasticsearch_memory.py swh/scheduler/exc.py swh/scheduler/interface.py swh/scheduler/journal_client.py swh/scheduler/model.py swh/scheduler/py.typed swh/scheduler/pytest_plugin.py swh/scheduler/task.py swh/scheduler/utils.py swh/scheduler/api/__init__.py swh/scheduler/api/client.py swh/scheduler/api/serializers.py swh/scheduler/api/server.py swh/scheduler/celery_backend/__init__.py swh/scheduler/celery_backend/config.py swh/scheduler/celery_backend/listener.py swh/scheduler/celery_backend/pika_listener.py swh/scheduler/celery_backend/runner.py swh/scheduler/cli/__init__.py swh/scheduler/cli/admin.py swh/scheduler/cli/celery_monitor.py swh/scheduler/cli/journal.py swh/scheduler/cli/origin.py swh/scheduler/cli/simulator.py swh/scheduler/cli/task.py swh/scheduler/cli/task_type.py swh/scheduler/cli/utils.py swh/scheduler/simulator/__init__.py swh/scheduler/simulator/common.py swh/scheduler/simulator/origin_scheduler.py swh/scheduler/simulator/origins.py swh/scheduler/simulator/task_scheduler.py swh/scheduler/sql/10-superuser-init.sql swh/scheduler/sql/30-schema.sql swh/scheduler/sql/40-func.sql swh/scheduler/sql/50-data.sql swh/scheduler/sql/60-indexes.sql swh/scheduler/tests/__init__.py swh/scheduler/tests/common.py swh/scheduler/tests/conftest.py swh/scheduler/tests/tasks.py swh/scheduler/tests/test_api_client.py swh/scheduler/tests/test_celery_tasks.py swh/scheduler/tests/test_cli.py swh/scheduler/tests/test_cli_celery_monitor.py swh/scheduler/tests/test_cli_journal.py swh/scheduler/tests/test_cli_origin.py swh/scheduler/tests/test_cli_task_type.py swh/scheduler/tests/test_common.py +swh/scheduler/tests/test_config.py swh/scheduler/tests/test_init.py swh/scheduler/tests/test_journal_client.py swh/scheduler/tests/test_model.py swh/scheduler/tests/test_scheduler.py swh/scheduler/tests/test_server.py swh/scheduler/tests/test_simulator.py swh/scheduler/tests/test_utils.py swh/scheduler/tests/es/__init__.py swh/scheduler/tests/es/conftest.py swh/scheduler/tests/es/test_backend_es.py swh/scheduler/tests/es/test_cli_task.py swh/scheduler/tests/es/test_elasticsearch_memory.py \ No newline at end of file diff --git a/swh/scheduler/celery_backend/runner.py b/swh/scheduler/celery_backend/runner.py index ea469ad..463a40d 100644 --- a/swh/scheduler/celery_backend/runner.py +++ b/swh/scheduler/celery_backend/runner.py @@ -1,132 +1,165 @@ -# Copyright (C) 2015-2018 The Software Heritage developers +# Copyright (C) 2015-2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information import logging +from typing import Dict, List, Tuple from kombu.utils.uuid import uuid from swh.core.statsd import statsd -from swh.scheduler import compute_nb_tasks_from, get_scheduler +from swh.scheduler import get_scheduler +from swh.scheduler.interface import SchedulerInterface from swh.scheduler.utils import utcnow logger = logging.getLogger(__name__) # Max batch size for tasks MAX_NUM_TASKS = 10000 -def run_ready_tasks(backend, app): - """Run tasks that are ready +def run_ready_tasks(backend: SchedulerInterface, app) -> List[Dict]: + """Schedule tasks ready to be scheduled. + + This lookups any tasks per task type and mass schedules those accordingly (send + messages to rabbitmq and mark as scheduled equivalent tasks in the scheduler + backend). + + If tasks (per task type) with priority exist, they will get redirected to dedicated + high priority queue (standard queue name prefixed with `save_code_now:`). Args: - backend (Scheduler): backend to read tasks to schedule + backend: scheduler backend to interact with (read/update tasks) app (App): Celery application to send tasks to Returns: A list of dictionaries:: { 'task': the scheduler's task id, 'backend_id': Celery's task id, 'scheduler': utcnow() } The result can be used to block-wait for the tasks' results:: backend_tasks = run_ready_tasks(self.scheduler, app) for task in backend_tasks: AsyncResult(id=task['backend_id']).get() """ - all_backend_tasks = [] + all_backend_tasks: List[Dict] = [] while True: task_types = {} pending_tasks = [] for task_type in backend.get_task_types(): task_type_name = task_type["type"] task_types[task_type_name] = task_type max_queue_length = task_type["max_queue_length"] if max_queue_length is None: max_queue_length = 0 backend_name = task_type["backend_name"] if max_queue_length: try: queue_length = app.get_queue_length(backend_name) except ValueError: queue_length = None if queue_length is None: # Running without RabbitMQ (probably a test env). num_tasks = MAX_NUM_TASKS else: num_tasks = min(max_queue_length - queue_length, MAX_NUM_TASKS) else: num_tasks = MAX_NUM_TASKS # only pull tasks if the buffer is at least 1/5th empty (= 80% # full), to help postgresql use properly indexed queries. if num_tasks > min(MAX_NUM_TASKS, max_queue_length) // 5: - num_tasks, num_tasks_priority = compute_nb_tasks_from(num_tasks) - + # Only grab num_tasks tasks with no priority grabbed_tasks = backend.grab_ready_tasks( - task_type_name, - num_tasks=num_tasks, - num_tasks_priority=num_tasks_priority, + task_type_name, num_tasks=num_tasks, num_tasks_priority=0 ) if grabbed_tasks: pending_tasks.extend(grabbed_tasks) logger.info( "Grabbed %s tasks %s", len(grabbed_tasks), task_type_name ) statsd.increment( "swh_scheduler_runner_scheduled_task_total", len(grabbed_tasks), tags={"task_type": task_type_name}, ) + # grab max_queue_length (or 10) potential tasks with any priority for the + # same type (limit the result to avoid too long running queries) + grabbed_priority_tasks = backend.grab_ready_priority_tasks( + task_type_name, num_tasks=max_queue_length or 10 + ) + if grabbed_priority_tasks: + pending_tasks.extend(grabbed_priority_tasks) + logger.info( + "Grabbed %s tasks %s (priority)", + len(grabbed_priority_tasks), + task_type_name, + ) + statsd.increment( + "swh_scheduler_runner_scheduled_task_total", + len(grabbed_priority_tasks), + tags={"task_type": task_type_name}, + ) + if not pending_tasks: return all_backend_tasks backend_tasks = [] - celery_tasks = [] + celery_tasks: List[Tuple[bool, str, str, List, Dict]] = [] for task in pending_tasks: args = task["arguments"]["args"] kwargs = task["arguments"]["kwargs"] backend_name = task_types[task["type"]]["backend_name"] backend_id = uuid() - celery_tasks.append((backend_name, backend_id, args, kwargs)) + celery_tasks.append( + ( + task.get("priority") is not None, + backend_name, + backend_id, + args, + kwargs, + ) + ) data = { "task": task["id"], "backend_id": backend_id, "scheduled": utcnow(), } backend_tasks.append(data) logger.debug("Sent %s celery tasks", len(backend_tasks)) backend.mass_schedule_task_runs(backend_tasks) - for backend_name, backend_id, args, kwargs in celery_tasks: - app.send_task( - backend_name, task_id=backend_id, args=args, kwargs=kwargs, - ) + for with_priority, backend_name, backend_id, args, kwargs in celery_tasks: + kw = dict(task_id=backend_id, args=args, kwargs=kwargs,) + if with_priority: + kw["queue"] = f"save_code_now:{backend_name}" + app.send_task(backend_name, **kw) all_backend_tasks.extend(backend_tasks) def main(): from .config import app as main_app for module in main_app.conf.CELERY_IMPORTS: __import__(module) main_backend = get_scheduler("local") try: run_ready_tasks(main_backend, main_app) except Exception: main_backend.rollback() raise if __name__ == "__main__": main() diff --git a/swh/scheduler/simulator/origin_scheduler.py b/swh/scheduler/simulator/origin_scheduler.py index ca16912..682e12f 100644 --- a/swh/scheduler/simulator/origin_scheduler.py +++ b/swh/scheduler/simulator/origin_scheduler.py @@ -1,68 +1,68 @@ # Copyright (C) 2021 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information """Agents using the new origin-aware scheduler.""" import logging from typing import Any, Dict, Generator, Iterator, List from simpy import Event from swh.scheduler.journal_client import process_journal_objects from .common import Environment, Queue, Task, TaskEvent logger = logging.getLogger(__name__) def scheduler_runner_process( env: Environment, task_queues: Dict[str, Queue], policy: str, min_batch_size: int ) -> Iterator[Event]: """Scheduler runner. Grabs next visits from the database according to the scheduling policy, and fills the task_queues accordingly.""" while True: for visit_type, queue in task_queues.items(): remaining = queue.slots_remaining() if remaining < min_batch_size: continue next_origins = env.scheduler.grab_next_visits( visit_type, remaining, policy=policy, timestamp=env.time ) logger.debug( "%s runner: running %s %s tasks", env.time, visit_type, len(next_origins), ) for origin in next_origins: yield queue.put(Task(visit_type=origin.visit_type, origin=origin.url)) yield env.timeout(10.0) def scheduler_journal_client_process( env: Environment, status_queue: Queue ) -> Generator[Event, TaskEvent, None]: """Scheduler journal client. Every once in a while, pulls - `OriginVisitStatus`es from the status_queue to update the scheduler - origin_visit_stats table.""" + :class:`OriginVisitStatuses ` + from the status_queue to update the scheduler origin_visit_stats table.""" BATCH_SIZE = 100 statuses: List[Dict[str, Any]] = [] while True: task_event = yield status_queue.get() statuses.append(task_event.status.to_dict()) if len(statuses) < BATCH_SIZE: continue logger.debug( "%s journal client: processing %s statuses", env.time, len(statuses) ) process_journal_objects( {"origin_visit_status": statuses}, scheduler=env.scheduler ) statuses = [] diff --git a/swh/scheduler/task.py b/swh/scheduler/task.py index 46bbe81..0db6204 100644 --- a/swh/scheduler/task.py +++ b/swh/scheduler/task.py @@ -1,81 +1,85 @@ # Copyright (C) 2015-2019 The Software Heritage developers # See the AUTHORS file at the top-level directory of this distribution # License: GNU General Public License version 3, or any later version # See top-level LICENSE file for more information from datetime import datetime from celery import current_app import celery.app.task from celery.utils.log import get_task_logger from swh.core.statsd import Statsd def ts(): return int(datetime.utcnow().timestamp()) class SWHTask(celery.app.task.Task): """a schedulable task (abstract class) Current implementation is based on Celery. See http://docs.celeryproject.org/en/latest/reference/celery.app.task.html for how to use tasks once instantiated """ _statsd = None _log = None + reject_on_worker_lost = None + """Inherited from :class:`celery.app.task.Task`, but we need to override + its docstring because it uses a custom ReST role""" + @property def statsd(self): if self._statsd: return self._statsd worker_name = current_app.conf.get("worker_name") if worker_name: self._statsd = Statsd( constant_tags={"task": self.name, "worker": worker_name,} ) return self._statsd else: statsd = Statsd( constant_tags={"task": self.name, "worker": "unknown worker",} ) return statsd def __call__(self, *args, **kwargs): self.statsd.increment("swh_task_called_count") self.statsd.gauge("swh_task_start_ts", ts()) with self.statsd.timed("swh_task_duration_seconds"): result = super().__call__(*args, **kwargs) try: status = result["status"] if status == "success": status = "eventful" if result.get("eventful") else "uneventful" except Exception: status = "eventful" if result else "uneventful" self.statsd.gauge("swh_task_end_ts", ts(), tags={"status": status}) return result def on_failure(self, exc, task_id, args, kwargs, einfo): self.statsd.increment("swh_task_failure_count") def on_success(self, retval, task_id, args, kwargs): self.statsd.increment("swh_task_success_count") # this is a swh specific event. Used to attach the retval to the # task_run self.send_event("task-result", result=retval) @property def log(self): if self._log is None: self._log = get_task_logger(self.name) return self._log def run(self, *args, **kwargs): self.log.debug("%s: args=%s, kwargs=%s", self.name, args, kwargs) ret = super().run(*args, **kwargs) self.log.debug("%s: OK => %s", self.name, ret) return ret diff --git a/swh/scheduler/tests/test_celery_tasks.py b/swh/scheduler/tests/test_celery_tasks.py index ab8be43..1a20f31 100644 --- a/swh/scheduler/tests/test_celery_tasks.py +++ b/swh/scheduler/tests/test_celery_tasks.py @@ -1,174 +1,254 @@ +# Copyright (C) 2019-2021 The Software Heritage developers +# See the AUTHORS file at the top-level directory of this distribution +# License: GNU General Public License version 3, or any later version +# See top-level LICENSE file for more information + +"""Module in charge of testing the scheduler runner module""" + from itertools import count from time import sleep from celery.result import AsyncResult, GroupResult import pytest from swh.scheduler.celery_backend.runner import run_ready_tasks from swh.scheduler.utils import create_task_dict def test_ping(swh_scheduler_celery_app, swh_scheduler_celery_worker): res = swh_scheduler_celery_app.send_task("swh.scheduler.tests.tasks.ping") assert res res.wait() assert res.successful() assert res.result == "OK" def test_ping_with_kw(swh_scheduler_celery_app, swh_scheduler_celery_worker): res = swh_scheduler_celery_app.send_task( "swh.scheduler.tests.tasks.ping", kwargs={"a": 1} ) assert res res.wait() assert res.successful() assert res.result == "OK (kw={'a': 1})" def test_multiping(swh_scheduler_celery_app, swh_scheduler_celery_worker): "Test that a task that spawns subtasks (group) works" res = swh_scheduler_celery_app.send_task( "swh.scheduler.tests.tasks.multiping", kwargs={"n": 5} ) assert res res.wait() assert res.successful() # retrieve the GroupResult for this task and wait for all the subtasks # to complete promise_id = res.result assert promise_id promise = GroupResult.restore(promise_id, app=swh_scheduler_celery_app) for i in range(5): if promise.ready(): break sleep(1) results = [x.get() for x in promise.results] assert len(results) == 5 for i in range(5): assert ("OK (kw={'i': %s})" % i) in results def test_scheduler_fixture( swh_scheduler_celery_app, swh_scheduler_celery_worker, swh_scheduler ): "Test that the scheduler fixture works properly" task_type = swh_scheduler.get_task_type("swh-test-ping") assert task_type assert task_type["backend_name"] == "swh.scheduler.tests.tasks.ping" swh_scheduler.create_tasks([create_task_dict("swh-test-ping", "oneshot")]) backend_tasks = run_ready_tasks(swh_scheduler, swh_scheduler_celery_app) assert backend_tasks for task in backend_tasks: # Make sure the task completed AsyncResult(id=task["backend_id"]).get() -def test_task_return_value( +def test_run_ready_task_standard( swh_scheduler_celery_app, swh_scheduler_celery_worker, swh_scheduler ): - task_type = swh_scheduler.get_task_type("swh-test-add") + """Ensure scheduler runner schedules tasks ready for scheduling""" + task_type_name, backend_name = "swh-test-add", "swh.scheduler.tests.tasks.add" + task_type = swh_scheduler.get_task_type(task_type_name) assert task_type - assert task_type["backend_name"] == "swh.scheduler.tests.tasks.add" + assert task_type["backend_name"] == backend_name + + task_inputs = [ + ("oneshot", (12, 30)), + ("oneshot", (20, 10)), + ("recurring", (30, 10)), + ] + + tasks = swh_scheduler.create_tasks( + create_task_dict(task_type_name, policy, *args) + for (policy, args) in task_inputs + ) + + assert len(tasks) == len(task_inputs) - swh_scheduler.create_tasks([create_task_dict("swh-test-add", "oneshot", 12, 30)]) + task_ids = set() + for task in tasks: + assert task["status"] == "next_run_not_scheduled" + assert task["priority"] is None + task_ids.add(task["id"]) backend_tasks = run_ready_tasks(swh_scheduler, swh_scheduler_celery_app) - assert len(backend_tasks) == 1 - task = backend_tasks[0] - value = AsyncResult(id=task["backend_id"]).get() - assert value == 42 + assert len(backend_tasks) == len(tasks) + + scheduled_tasks = swh_scheduler.search_tasks(task_type=task_type_name) + assert len(scheduled_tasks) == len(tasks) + for task in scheduled_tasks: + assert task["status"] == "next_run_scheduled" + assert task["id"] in task_ids + + for i, (_, args) in enumerate(task_inputs): + # take one of the task and read it from the queue backend + task = backend_tasks[i] + value = AsyncResult(id=task["backend_id"]).get() + assert value == sum(args) + + +def test_run_ready_task_with_priority( + swh_scheduler_celery_app, swh_scheduler_celery_worker, swh_scheduler +): + """Ensure scheduler runner schedules priority tasks ready for scheduling""" + task_type_name, backend_name = "swh-test-add", "swh.scheduler.tests.tasks.add" + task_type = swh_scheduler.get_task_type(task_type_name) + assert task_type + assert task_type["backend_name"] == backend_name + + task_inputs = [ + ("oneshot", (10, 22), "low"), + ("oneshot", (20, 10), "normal"), + ("recurring", (30, 10), "high"), + ] + + tasks = swh_scheduler.create_tasks( + create_task_dict(task_type_name, policy, *args, priority=priority) + for (policy, args, priority) in task_inputs + ) + + assert len(tasks) == len(task_inputs) + + task_ids = set() + for task in tasks: + assert task["status"] == "next_run_not_scheduled" + assert task["priority"] is not None + task_ids.add(task["id"]) + + backend_tasks = run_ready_tasks(swh_scheduler, swh_scheduler_celery_app) + assert len(backend_tasks) == len(tasks) + + scheduled_tasks = swh_scheduler.search_tasks(task_type=task_type_name) + assert len(scheduled_tasks) == len(tasks) + for task in scheduled_tasks: + assert task["status"] == "next_run_scheduled" + assert task["id"] in task_ids + + # TODO: Make the worker consume those messages so this can go green + # for i, (_, args, _) in enumerate(task_inputs): + # # take one of the task and read it from the queue backend + # task = backend_tasks[i] + # value = AsyncResult(id=task["backend_id"]).get() + # assert value == sum(args) def test_task_exception( swh_scheduler_celery_app, swh_scheduler_celery_worker, swh_scheduler ): task_type = swh_scheduler.get_task_type("swh-test-error") assert task_type assert task_type["backend_name"] == "swh.scheduler.tests.tasks.error" swh_scheduler.create_tasks([create_task_dict("swh-test-error", "oneshot")]) backend_tasks = run_ready_tasks(swh_scheduler, swh_scheduler_celery_app) assert len(backend_tasks) == 1 task = backend_tasks[0] result = AsyncResult(id=task["backend_id"]) with pytest.raises(NotImplementedError): result.get() def test_statsd(swh_scheduler_celery_app, swh_scheduler_celery_worker, mocker): m = mocker.patch("swh.scheduler.task.Statsd._send_to_server") mocker.patch("swh.scheduler.task.ts", side_effect=count()) mocker.patch("swh.core.statsd.monotonic", side_effect=count()) res = swh_scheduler_celery_app.send_task("swh.scheduler.tests.tasks.echo") assert res res.wait() assert res.successful() assert res.result == {} m.assert_any_call( "swh_task_called_count:1|c|" "#task:swh.scheduler.tests.tasks.echo,worker:unknown worker" ) m.assert_any_call( "swh_task_start_ts:0|g|" "#task:swh.scheduler.tests.tasks.echo,worker:unknown worker" ) m.assert_any_call( "swh_task_end_ts:1|g|" "#status:uneventful,task:swh.scheduler.tests.tasks.echo," "worker:unknown worker" ) m.assert_any_call( "swh_task_duration_seconds:1000|ms|" "#task:swh.scheduler.tests.tasks.echo,worker:unknown worker" ) m.assert_any_call( "swh_task_success_count:1|c|" "#task:swh.scheduler.tests.tasks.echo,worker:unknown worker" ) def test_statsd_with_status( swh_scheduler_celery_app, swh_scheduler_celery_worker, mocker ): m = mocker.patch("swh.scheduler.task.Statsd._send_to_server") mocker.patch("swh.scheduler.task.ts", side_effect=count()) mocker.patch("swh.core.statsd.monotonic", side_effect=count()) res = swh_scheduler_celery_app.send_task( "swh.scheduler.tests.tasks.echo", kwargs={"status": "eventful"} ) assert res res.wait() assert res.successful() assert res.result == {"status": "eventful"} m.assert_any_call( "swh_task_called_count:1|c|" "#task:swh.scheduler.tests.tasks.echo,worker:unknown worker" ) m.assert_any_call( "swh_task_start_ts:0|g|" "#task:swh.scheduler.tests.tasks.echo,worker:unknown worker" ) m.assert_any_call( "swh_task_end_ts:1|g|" "#status:eventful,task:swh.scheduler.tests.tasks.echo," "worker:unknown worker" ) m.assert_any_call( "swh_task_duration_seconds:1000|ms|" "#task:swh.scheduler.tests.tasks.echo,worker:unknown worker" ) m.assert_any_call( "swh_task_success_count:1|c|" "#task:swh.scheduler.tests.tasks.echo,worker:unknown worker" ) diff --git a/swh/scheduler/tests/test_config.py b/swh/scheduler/tests/test_config.py new file mode 100644 index 0000000..c166f62 --- /dev/null +++ b/swh/scheduler/tests/test_config.py @@ -0,0 +1,18 @@ +# Copyright (C) 2021 The Software Heritage developers +# See the AUTHORS file at the top-level directory of this distribution +# License: GNU General Public License version 3, or any later version +# See top-level LICENSE file for more information + +import pytest + +from swh.scheduler.celery_backend.config import route_for_task + + +@pytest.mark.parametrize("name", ["swh.something", "swh.anything"]) +def test_route_for_task_routing(name): + assert route_for_task(name, [], {}, {}) == {"queue": name} + + +@pytest.mark.parametrize("name", [None, "foobar"]) +def test_route_for_task_no_routing(name): + assert route_for_task(name, [], {}, {}) is None