Toggleable Async Propgation on the trace context#282
Conversation
realark
force-pushed
the
ark/async-refactor
branch
3 times, most recently
from
e8e9fec to
65f936f
Compare
realark
added
inst: others
realark
force-pushed
the
ark/async-refactor
branch
from
12a200f to
cfa12c9
Compare
realark
force-pushed
the
ark/async-refactor
branch
from
cfa12c9 to
03b96ca
Compare
tylerbenson
left a comment
tylerbenson
left a comment
There was a problem hiding this comment.
My main question is what is the behavior for chunked responses. I think that would be a good case to document and have a test for.
I assume our response time would not include that, right?
|
|
||
| /** Maximum number of traces kept in memory */ | ||
| static final int DEFAULT_MAX_TRACES = 1000; | ||
| static final int DEFAULT_MAX_TRACES = 7000; |
There was a problem hiding this comment.
In my testing, Play 2.6 hit ~6000 req/second running the scala-starter app. I picked 7000 to give us some breathing room.
| import io.opentracing.Span; | ||
|
|
||
| /** Simple scope implementation which does not propagate across threads. */ | ||
| public class SimpleScope implements Scope { |
There was a problem hiding this comment.
Would you rather just use ThreadLocalScopeManager thats included with OpenTracing?
There was a problem hiding this comment.
Yes, good idea. I'll do that in another PR.
|
For chunked responses, we'll capture time-to-first-byte. I'll look into getting an integration test |
2 participants
Uh oh!
There was an error while loading. Please reload this page.