Noah Watkins

> @dotnwat A client can't issue more than one fetch request at a time, because a client has to process the entire response so as to know what offset to...

excellent. I think then we don't need to worry about this but its something to keep in the back of our minds as either a possible optimization later or a...

seems reasonable. we could add `scram_password_file` or something. does k8s have any special formatting in the secret file or does it contain only the password?

> > we could add `scram_password_file` or something. > > makes sense > > > does k8s have any special formatting in the secret file or does it contain only...

https://github.com/vectorizedio/redpanda/blob/dev/src/v/utils/file_io.h#L27

configuration{.h,cc} for adding the file name option. then application.cc (probably something like hydrate_config) for reading up the password from the secret file.

@mmaslankaprv do you think this might be part of the compaction backlog controller?

happening more now. this definitely seems to be a new behavior. this time all I had to do was fetch a lot of data. ``` TRACE 2021-05-20 18:16:17,889 [shard 2]...

seems to be chillin in compaction ``` Thread 2 (Thread 0x7f5c9b7ff700 (LWP 2631856)): #0 absl::container_internal::raw_hash_set::iterator::skip_empty_or_deleted (this=) at rp_deps_install/include/absl/container/internal/raw_hash_set.h:688 #1 absl::container_internal::raw_hash_set::begin (this=0x6010e15c4868) at rp_deps_install/include/absl/container/internal/raw_hash_set.h:930 #2 storage::internal::compaction_key_reducer::operator() (this=0x6010e15c4840, e=...) at ../../../src/v/storage/compaction_reducers.cc:60 #3...

this is in compaction. after compaction completes redpanda exists nicely.