gerlowskija commented on pull request #1574: URL: https://github.com/apache/lucene-solr/pull/1574#issuecomment-654278567
> Shouldn't this RID be saved to MDC like a tracer ID does? **TL;DR** It _could_, but it _shouldn't_ for practical reasons. It could. Also, you could do this on all requests, not just distributed ones. And you could get it working on indexing requests too, and admin requests which store chunks of work on the overseer queue. Ideally you could do this as some new OpenTracing implementation that would serve as an alternative to the JaegerTracing implementation that's documented right now. And you could update Joel's SolrLogPostTool (which is really cool btw - just to plug it) to parse out the rid's as their own field. There's a lot of really cool and powerful improvements to make here. They're just not what I'm setting out to do right now. SOLR-8274 was created 5 years ago to add MDC-based tracing, and I'd prefer that approach stay on that JIRA and not spread to SOLR-14566. I created SOLR-14566 as a trivial fix to get users some very-performant, configured ootb tracing while others pursue a more comprehensive MDC approach in parallel. If someone ever does do SOLR-8274, they can rip out this logic here. But I want to get users (and really, the debuggers that support them behind the scenes) _something_ in the meantime. To use the perfect vs good analogy - I'd rather not tie the "good" of getting this in now to the "perfect" of SOLR-8274 that I hope gets in someday but never generated much momentum. ---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org For additional commands, e-mail: issues-h...@lucene.apache.org
