You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@pinot.apache.org by "ankitsultana (via GitHub)" <gi...@apache.org> on 2023/06/28 09:55:19 UTC
[GitHub] [pinot] ankitsultana opened a new issue, #10992: [multistage] Revisit Stage Level Unique RequestId
ankitsultana opened a new issue, #10992:
URL: https://github.com/apache/pinot/issues/10992
Looks like we are making request-ids unique at stage level for the leaf stage execution.
This doesn't fit neatly with the concept of request-id because we'd want a single request-id for tracing. Can someone confirm if and what the plan is to fix this?
cc: @walterddr
https://github.com/apache/pinot/blob/04bdb9a39743569aaacfcb07e2cc7eaa08854791/pinot-query-runtime/src/main/java/org/apache/pinot/query/runtime/plan/server/ServerPlanRequestUtils.java#L126C13-L126C13
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org
For additional commands, e-mail: commits-help@pinot.apache.org
[GitHub] [pinot] walterddr commented on issue #10992: [multistage] Revisit Stage Level Unique RequestId
Posted by "walterddr (via GitHub)" <gi...@apache.org>.
walterddr commented on issue #10992:
URL: https://github.com/apache/pinot/issues/10992#issuecomment-1611731906
this is a GOOD point!
as of now, the entire v1 execution (all the way to the instanceResponseBlock) is considered as a "single-operator" of V2 (this execution is encapsulate inside the `LeafStageTransferableBlockOperator` which is the bridge between v1 and v2.
- short term suggestion: The trace info generated from the v1 side should ideally be remodified inside the bridge class
- longer term: the entire v1 operator tree should support partition/stage concept in trace context
- long-long term: model both v1 and v2 as a single operator module then there shouldn't be any issue
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org
For additional commands, e-mail: commits-help@pinot.apache.org