A68 random subsetting #12377
A68 random subsetting #12377
Conversation
Zgoda91
force-pushed
the
A68_random_subsetting
branch
from
9178308 to
aefee00
Compare
Zgoda91
force-pushed
the
A68_random_subsetting
branch
from
aefee00 to
1870e6f
Compare
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
This reverts commit 8d349df.
|
@ejona86 - PR updated |
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
util/src/test/java/io/grpc/util/RandomSubsettingLoadBalancerTest.java
Outdated
Show resolved
Hide resolved
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancerProvider.java
Show resolved
Hide resolved
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
|
@ejona86 - PR updated. |
util/src/main/java/io/grpc/util/RandomSubsettingLoadBalancer.java
Outdated
Show resolved
Hide resolved
| Long subsetSize = JsonUtil.getNumberAsLong(rawConfig, "subsetSize"); | ||
| if (subsetSize == null) { | ||
| return ConfigOrError.fromError( | ||
| Status.INTERNAL.withDescription( |
There was a problem hiding this comment.
These should probably be UNAVAILABLE. INTERNAL isn't appropriate, because nothing in gRPC was busted; this was just wrong configuration. I suspect you probably copied this from OutlierDetectionLoadBalancerProvider, which itself is wrong. There's really no difference here than on line 57 if we get a RuntimeException. #12409 will fix that.
There was a problem hiding this comment.
Yeah, I was looking at other providers code and as well old java solution to understand which Status I am supposed to use.
Does the same apply to child policy parsing? I can see that GracefulSwitchLoadBalancer.parseLoadBalancingPolicyConfig(...) is using INTERNAL to represent parsing failures.
There was a problem hiding this comment.
Add the file to
grpc-java/xds/third_party/envoy/import.sh
Lines 26 to 30 in 349a35a
And re-run the script to download the same version as the rest of the Envoy commit we are using.
There was a problem hiding this comment.
Hmm, it seems like there is nothing like random_subsetting LB policy in envoy. The closest seems random or subset, but those aren't what we exactly need.
How should I tackle this problem? Should I update the envoy API before merging this PR then?
|
@ejona86 - pushed changes. However, I do have two questions, which I have posted as answers to your comments. |
2 participants
implementing gRFC A65 grpc/proposal/pull/423
This change contains:
Relocation ofUsage ofXxHash64library, so it can be shared betweenutilandxdsprojects. Proposed source directory is:third-party/zero-allocation-hashing.murmur3_128hashing algorithm from Guava library.RandomSubsettingLoadBalancerandRandomSubsettingLoadBalancerProviderclasses and integration into theutilproject.random_subsettingLB policy and as well new envoy proto message.