Tim Wojtulewicz

> When the dictionary size is set to the final size (so that increase of dictionary size and remapping of old doesn't happen), the performance is normal. This happens independent...

I made some changes to the `TableVal` implementation to pass the size of the original table down to the new table before doing the clone, but it doesn't seem to...

This is getting pretty close. One thing I noticed while doing a bit more testing with it this morning was the log entries in notice.log: ``` Failure Installing packet filter...

I'm finally happy with the state of this so I'm moving it from a draft to a regular PR. I think the only thing missing is a btest that triggers...

This PR includes changes to the zeek-testing, zeek-testing-private, and zeek-docs repos. There are branches with the same name in those repos that need to be merged when this one is...

> I think the only thing missing is a btest that triggers warnings, but I'm waiting for an OK from the person who gave me the pcap that originated this....

There's something busted about this branch on CI now. It keeps not being able to find the commit hash, even though if I re-run the job with terminal access I...

I opened https://github.com/cirruslabs/cirrus-ci-web/issues/454 to figure out why this build is failing.

> the caf.thread spends a lot of time in buffered_observable_impl::try_push() doing std::vector::erase() (which likely is related the individual publishes of intel items). Looking a bit at the code, it's erasing...

> Without any proxies or workers started, loading this file alone brings the manager from 100MB to 1.6GB memory usage (zeek version 5.1.0-dev.486). So that alone is an expansion of...