spanconfig/rangefeed: investigate microbenchmark regression #79602
Labels
A-kv
Anything in KV that doesn't belong in a more specific category.
A-zone-configs
C-bug
Code not up to spec/doc, specs & docs deemed correct. Solution expected to change code/behavior.
no-issue-activity
X-stale
Describe the problem
One of our microbenchmarks show additional memory usage and was bisected to #74555. See #74555 (comment).
To Reproduce
I ran the following and used
go tool pprof -http localhost:8082 -diff_base=mem-old.out -normalize mem-new.out
to see where the differences are coming from.Looks like it's additional rangefeed event processing. I'm not sure off-hand why; I suspect it's got something to do with ranges having to enable rangefeeds by default when no targeted config is available.
Expected resolution
Pin down exactly why we're doing more processing, and whether it's expected.
Jira issue: CRDB-14939
The text was updated successfully, but these errors were encountered: