Initial inclusion of pre-activation triggers #129
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Ikey Doherty <[email protected]>
This ensures we only match to the bounds of directories and not beyond. Signed-off-by: Ikey Doherty <[email protected]>
Signed-off-by: Ikey Doherty <[email protected]>
While this certainly isn't our last iteration we now have enough pieces in to support `pre-activation` triggers in moss, powered by the new `triggers::Collection` management type with DAG-based dependency ordering and `fnmatch`-based path globbing. Signed-off-by: Ikey Doherty <[email protected]>
This is currently only tested (and supported) with stateful roots, not ephemeral. We identify all paths from our `vfs::Tree` and pump them via a `triggers::Collection` that we load from the usr tree within the yet-to-be-activated rootfs pre-triggers. Currently no UI integration work has been done, and we simply run all triggers via the `container` crate (clone-based ns) within their own isolated environment pre activation of `/usr`. Signed-off-by: Ikey Doherty <[email protected]>
Signed-off-by: Ikey Doherty <[email protected]>
|
For testing this PR note that |
|
yolo |
tarkah
reviewed
Jan 27, 2024
Comment on lines
+35
to
+36
| triggers = { version = "0.1.0", path = "../triggers" } | ||
| container = { version = "0.1.0", path = "../container" } |
There was a problem hiding this comment.
Nit: we have local crates grouped above
| } | ||
|
|
||
| fn merge(self, other: Self) -> Self { | ||
| todo!() |
There was a problem hiding this comment.
never used as we load all :D
Comment on lines
+121
to
+135
| // Recollect in dependency order | ||
| let built_triggers = graph | ||
| .topo() | ||
| .cloned() | ||
| .map(|i| { | ||
| self.hits | ||
| .iter() | ||
| .filter(move |(id, _, _)| i == *id) | ||
| .collect::<Vec<_>>() | ||
| }) | ||
| .flatten() | ||
| .collect::<Vec<_>>(); | ||
|
|
||
| let mut runnables: BTreeMap<format::Handler, TriggerCommand> = BTreeMap::new(); | ||
| for (_, hit, handler) in built_triggers { |
There was a problem hiding this comment.
Do we no longer care about dependency order when adding to runnables? format::Handler seems to be ordered basically off command name
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Thoughts and suggestions welcome, this lands the initial
pre-triggersas part of a new unifiedpostblitprocess that will eventually grow to absorb tasks like os-release management.We currently have no caching mechanism, so each trigger is run fresh for each new transaction. Some scaffolding is
put in place to know which paths are involved in the fnmatch-based glob, so that in future we can capture the input hashes (CA xxhash) to "save" trigger executions.
Note this PR yields a writable
/usrand/etc, however the plan is to make/etcread-only. This is currently broken and needs investigation once we have a baseline,