diff options
author | Jasper Ras <jras@hostnet.nl> | 2025-01-19 21:14:51 +0100 |
---|---|---|
committer | Jasper Ras <jras@hostnet.nl> | 2025-01-19 21:14:51 +0100 |
commit | 9642cd7ae24f0ba79ce5647c709b35ae8f06a285 (patch) | |
tree | ae25c3b0db7ae4c23186b294c6d000073f085c2d /.trash/Resumable.md | |
parent | 969b96c2531fb986f6c7f21fd544391b439defd7 (diff) |
vault backup: 2025-01-19 21:14:51
Diffstat (limited to '.trash/Resumable.md')
-rw-r--r-- | .trash/Resumable.md | 13 |
1 files changed, 13 insertions, 0 deletions
diff --git a/.trash/Resumable.md b/.trash/Resumable.md new file mode 100644 index 0000000..1a36ef0 --- /dev/null +++ b/.trash/Resumable.md @@ -0,0 +1,13 @@ +--- +tags: + - taskflow +references: + - https://docs.openstack.org/taskflow/latest/user/resumption.html + - https://opendev.org/openstack/taskflow/src/branch/master//taskflow/examples/resume_from_backend.py +--- +[Example](https://opendev.org/openstack/taskflow/src/branch/master//taskflow/examples/resume_from_backend.py) + +A flow can be resumed by reconstructing the links between atoms. This is done by a factory function that has to be importable. Instead of passing a flow to the run function, we need to pass the factorkkky function to the load-from-factory function which will call the factory to create the flow and passes it to the engine. + +Atom names should be unique in a flow so that when resuming it the corresponding flow detail can be matched. + |