-
Notifications
You must be signed in to change notification settings - Fork 117
Fix boundaries for Symfony HttpKernel auto instrumentation #317
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Open
dciprian-petrisor
wants to merge
1
commit into
open-telemetry:main
Choose a base branch
from
dciprian-petrisor:fix-symfony-auto-instrumentation-boundaries
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Changes from all commits
Commits
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or 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
This file contains hidden or 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
Oops, something went wrong.
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Outdated suggestions cannot be applied.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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So if an exception occurs, this post callback will run, otherwise the
terminateone will? If so, we should document this.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That is correct.
I'm assuming this should be documented in the README.md, right?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The examples on this page seem to suggest that the
handleandterminatecan both run.So it looks like you'd expect 2 spans to be created and then with these changes only close one of them? I'm not so sure that behaviour is correct.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There appears to be a parameter to configure whether
handleshould throw here.So when
$catch = falsethe handle post-hook might have an exception andterminatewill not have been reached yet. I think this would be fine as it handles the recording of the exception and detaches the scope.However, if
$catch = falseand there is no exception, thenterminate()would be called afterhandleand theterminatepost-hook could try to close a scope which does not belong to it.Uh oh!
There was an error while loading. Please reload this page.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@ChrisLightfootWild
If
handlethrows an exception which is never caught, thenterminatewill never be reached either, according to theHttpKernelRunnerimplementation.In this situation, it is correct to close the scope on the
posthook ofhandle.If
handledoes not throw an exception (or has$catch = true), thenterminatewould run afterwards, closing the scope.The only problem I see is there is not a 100% guarantee that
terminatewill always be executed afterhandle.For example, something else might go wrong in
HttpKernelRunnerafterhandlewas called, e.g.fastcgi_finish_request()/litespeed_finish_request()might crash,flush()might crash, etc, leaving the scope still open.Unsure how much of a real issue this is, in most cases.
Normally this would be solved by hooking into
HttpKernelRunner::runwith bothpreandposthooks, as it covers bothhandleandterminate, however$requestand$responseare not available as parameters there, so I am at a loss right now.The main issue is that the
parentspan is lost in betweenhandleandterminate, due tohandledetaching it, thus producing two different traces for the same request:handleTerminateEventdone by$this->kernel->terminateafterwardsthe second should be part of the first, as it is still logically part of the same HTTP request lifecycle.
We need either a compromise or another solution entirely.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for that @dciprian-petrisor - I had another look with less bleary eyes and I finally caught up to you guys 💪
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I am having the same problem for Drupal, but opted to do an auto root span for now, which obviously duplicates the logic to do that.
That way I have a duplicated span for the Kernel::handle() + Kernel::terminate(), but that does not matter as it's still within one root trace.
The Auto Root Span uses the SDK ShutdownHandler::register so that might be also an option here.
The other thing is:
You could hook into KernelRunner::run(), but not open the scope here, but instead close it just during the post() [if ::run() was called in pre].
While that is more logic to handle, because handle() needs to check that ::run was actually called, so that it can close it, cross hook frontiers are a good way to handle this.
In general I personally will still call the Auto Root Scope hook as early as possible (e.g. DrupalKernel::__construct in my case), because else there is always an ::IO extension call, which escapes.
Until that's converted to SPI obviously there are still race conditions.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is my implementation:
It works really well.