[health checking] prep for outlier detection ejecting via health watch#33340
Merged
markdroth merged 10 commits intoJun 26, 2023
Merged
Conversation
…_ejection_via_health_watch
…r_detection_ejection_via_health_watch
…_ejection_via_health_watch
eugeneo
approved these changes
Jun 26, 2023
eugeneo
left a comment
Contributor
There was a problem hiding this comment.
Minor questions/comments.
| auto it = | ||
| health_checkers_.emplace(*health_check_service_name, nullptr).first; | ||
| auto& health_checker = it->second; | ||
| if (health_checker == nullptr) { |
Contributor
There was a problem hiding this comment.
Not sure I understand this code. Isn't this value always nullptr? Shouldn't MakeOrphanable... from line 372 replace nullptr on line 369?
Member
Author
There was a problem hiding this comment.
There are two possible outcomes from the emplace() call on line 369:
- The entry already existed in the map. In this case, no new entry is added (and the
nullptrvalue passed toemplace()is not actually stored anywhere), and the iterator will point to the preexisting entry in the map. - The entry did not already exist in the map. In this case,
emplace()will add the entry, and the iterator will point to the newly added entry, whose value isnullptr.
If health_checker is null here, that means we're in case 2, and we need to construct the new object.
The reason we don't call MakeOrphanable() before calling emplace() is that we don't want to actually allocate the object if the entry already exists in the map, because in that case, the newly allocated object would just wind up getting immediately destroyed, so we'd do unnecessary work.
| non_health_watchers_.erase(watcher); | ||
| } else { | ||
| auto it = health_checkers_.find(*health_check_service_name); | ||
| if (it == health_checkers_.end()) return; |
Contributor
There was a problem hiding this comment.
I find these several lines confusing, with multiple conditions and temp variable.
Suggested change
| if (it == health_checkers_.end()) return; | |
| if (it != health_checkers_.end() && it->second->RemoveWatcherLocked(watcher)) { | |
| health_checkers_.erase(it); | |
| } |
Member
Author
There was a problem hiding this comment.
I think the temp variable actually makes the code more readable. Without it, someone reading this code needs to go look at the comment on the RemoveWatcherLocked() method to see that it returns true when this was the last watcher in the HealthChecker.
eugeneo
approved these changes
Jun 26, 2023
markdroth
added a commit
that referenced
this pull request
Jun 28, 2023
…state (#33427) More work on the dualstack backend design: - Now that all petiole policies have been changed to delegate to pick_first, outlier detection no longer needs to eject via the subchannel's raw connectivity state; it can now eject only via the health state. See #33340. - This also removes the now-unnecessary hack to explicitly disable outlier detection in pick_first. See #33336.
mario-vimal
pushed a commit
to mario-vimal/grpc
that referenced
this pull request
Jul 13, 2023
grpc#33340) Currently, the outlier_detection policy reports ejection by intercepting a subchannel's raw connectivity state watch. In the dualstack backend design, we will instead want to report ejection via the health watch. This PR is a first step toward that goal. Specific changes in this PR: - Add `type()` method to `InternalSubchannelDataWatcherInterface`, to make it possible for LB policies to intercept data watchers. - Use that mechanism in the outlier_detection policy to report ejection both via raw connectivity state watches and via health watches. The hack to prevent outlier_detection from working with pick_first from grpc#33336 has been changed to affect only the raw connectivity watch, not the health watch. - Change health check client to fall back to reporting the raw connectivity state if client-side health checking is not enabled. This will allow pick_first to unconditionally start a health watch when it is running under a petiole policy, which will be a no-op if neither health checking nor outlier detection are configured. Once we are done changing all of the petiole policies to delegate to pick_first, we will remove the code that allows outlier_detection to work via the raw connectivity state, so it will work only via the health watch.
mario-vimal
pushed a commit
to mario-vimal/grpc
that referenced
this pull request
Jul 13, 2023
…state (grpc#33427) More work on the dualstack backend design: - Now that all petiole policies have been changed to delegate to pick_first, outlier detection no longer needs to eject via the subchannel's raw connectivity state; it can now eject only via the health state. See grpc#33340. - This also removes the now-unnecessary hack to explicitly disable outlier detection in pick_first. See grpc#33336.
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Currently, the outlier_detection policy reports ejection by intercepting a subchannel's raw connectivity state watch. In the dualstack backend design, we will instead want to report ejection via the health watch. This PR is a first step toward that goal.
Specific changes in this PR:
type()method toInternalSubchannelDataWatcherInterface, to make it possible for LB policies to intercept data watchers.Once we are done changing all of the petiole policies to delegate to pick_first, we will remove the code that allows outlier_detection to work via the raw connectivity state, so it will work only via the health watch.