-
Notifications
You must be signed in to change notification settings - Fork 304
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
HPCC-30734 Refactor the CSpan tracing classes #18060
HPCC-30734 Refactor the CSpan tracing classes #18060
Conversation
https://track.hpccsystems.com/browse/HPCC-30734 |
@rpastrana I would strongly recommend that you review this code before working on any other changes because they will almost certainly clash. |
{ | ||
out.append("\"Type\":\"Server\""); | ||
CSpan::toLog(out); | ||
|
||
if (!isEmptyString(hpccGlobalId.get())) |
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.
Moving these to serverSpans makes sense. I will mention, all these LN specific IDs have a very limited expected lifespan. One major request I've heard is to provide these values during "transition" period (from LN IDs to OTel IDs). My concern is not reporting these values for non server spans could make the mapping of LN -> Otel IDs unnecessarily difficult. If I've misunderstood the net effect of moving these up, please ignore this comment.
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.
They should still be reported exactly as before. The difference is that they're not stored in any of the other spans.
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.
It might be that the intent was that if you logged a client span it would log the global id, but it didn't before. That can be implemented as a separate PR if required (probably a sensible idea).
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.
@ghalliday I like a lot of these changes. I'm assuming CPPUNIT_TEST_SUITE(JlibTraceTest) pass successfully.
system/jlib/jtrace.cpp
Outdated
public: | ||
virtual bool getSpanContext(IProperties * ctxProps, bool otelFormatted) const override | ||
{ | ||
//MORE: It is not clear what this return value represents, and whether it would veer be checked. |
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.
minor: veer?
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.
Will fix -> ever
Signed-off-by: Gavin Halliday <[email protected]>
b55ef92
to
786c8ba
Compare
@rpastrana squashed and fixed the typo. |
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.
@ghalliday approved
Type of change:
Checklist:
Smoketest:
Testing: