Skip to content

shorten qps worker timeouts and run remote workers under timeout#9548

Merged
apolcyn merged 1 commit intogrpc:masterfrom
apolcyn:improve_benchmark_worker_timeouts
Feb 4, 2017
Merged

shorten qps worker timeouts and run remote workers under timeout#9548
apolcyn merged 1 commit intogrpc:masterfrom
apolcyn:improve_benchmark_worker_timeouts

Conversation

@apolcyn
Copy link
Copy Markdown
Contributor

@apolcyn apolcyn commented Feb 1, 2017

to make sure remote workers killed shortly after if ssh is killed

@apolcyn apolcyn mentioned this pull request Feb 2, 2017
Closed
@apolcyn
Copy link
Copy Markdown
Contributor Author

apolcyn commented Feb 2, 2017

seeing new failures like:

/home/jenkins/workspace/gRPC_performance_pull_requests/src/node/performance/worker_service_impl.js:78
        switch (setup.payload_config.payload) {
                                    ^

TypeError: Cannot read property 'payload' of null
    at ServerDuplexStream.<anonymous> (/home/jenkins/workspace/gRPC_performance_pull_requests/src/node/performance/worker_service_impl.js:78:37)
    at emitOne (events.js:77:13)
    at ServerDuplexStream.emit (events.js:169:7)
    at readableAddChunk (_stream_readable.js:153:18)
    at ServerDuplexStream.Readable.push (_stream_readable.js:111:10)
    at readCallback (/home/jenkins/workspace/gRPC_performance_pull_requests/src/node/src/server.js:360:14)

and

/home/jenkins/workspace/gRPC_performance_pull_requests/src/ruby/qps/client.rb:86:in `initialize': undefined method `simple_params' for nil:NilClass (NoMethodError)
	from src/ruby/qps/worker.rb:79:in `new'
	from src/ruby/qps/worker.rb:79:in `block (2 levels) in run_client'
	from /home/jenkins/workspace/gRPC_performance_pull_requests/src/ruby/lib/grpc/generic/bidi_call.rb:213:in `block in read_loop'
	from /home/jenkins/workspace/gRPC_performance_pull_requests/src/ruby/lib/grpc/generic/bidi_call.rb:191:in `loop'
	from /home/jenkins/workspace/gRPC_performance_pull_requests/src/ruby/lib/grpc/generic/bidi_call.rb:191:in `read_loop'
	from src/ruby/qps/worker.rb:76:in `each'
	from src/ruby/qps/worker.rb:76:in `block in run_client'

guessing may be related to this shortened timeout

@apolcyn
Copy link
Copy Markdown
Contributor Author

apolcyn commented Feb 2, 2017

jenkins: test this please

@apolcyn apolcyn force-pushed the improve_benchmark_worker_timeouts branch from 33a7589 to 76be306 Compare February 3, 2017 18:44
@apolcyn
Copy link
Copy Markdown
Contributor Author

apolcyn commented Feb 3, 2017

got rid of update after realizing failure unrelated, fixed by #9580.

went back again to original shortened timeout

jenkins: test this please

@apolcyn
Copy link
Copy Markdown
Contributor Author

apolcyn commented Feb 3, 2017

python and performance failures which should be fixed now., re-running again.

jenkins: test this please

@apolcyn apolcyn mentioned this pull request Feb 4, 2017
Closed
@apolcyn
Copy link
Copy Markdown
Contributor Author

apolcyn commented Feb 4, 2017 •
edited
Loading

test failures:

interop PR failure was a timeout building python docker image:

-performance PR: #9587 (seen before)

@apolcyn apolcyn mentioned this pull request Feb 4, 2017
Closed
@apolcyn apolcyn merged commit cf7587e into grpc:master Feb 4, 2017
@lock lock bot locked as resolved and limited conversation to collaborators Jan 25, 2019
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Reviewers

@ctiller ctiller ctiller approved these changes

Assignees

@ctiller ctiller

Labels

cla: yes

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@apolcyn @ctiller @googlebot