Testing Prereleases of Zeek
Before any major Zeek release, the development team makes beta versions and/or release candidates available for testing. It's extremely helpful for us to receive feedback on these versions from people used to running Zeek, and we appreciate input of any kind.
Most critical to report are crashes-—Zeek should never crash—and regressions: please let us know about anything that's unexpectedly changing from previous Zeek versions. This includes Zeek's analysis functionality (e.g., scripts not producing the same results anymore, or logs lacking information), as well as Zeek's performance, in particular noticeable increases in CPU usage or memory. In fact, if you can run and compare a current stable Zeek release in parallel to a prerelease version, with both seeing the same live traffic on your network, that's like gold to us!
If you can, please also try to test any new or changed functionality called out in the NEWS file. If some of that doesn't seem to work for you, or you just can't figure out how to make sense of it, please report that as well. Chances are that if you are running into challenges, others will as well. (Note that documentation on docs.zeek.org may still be incomplete when prerelases come out, as we tend to work on that during the time allocated for testing. Feel free to ask questions where something's not clear.)
We may also call out specific areas to watch out for with each prerelease, such as functionality that has changed internally and hence incurs a larger risk of regressions.
Please also let us know if everything works just fine for you: active confirmation of deployments going well is just as useful as bug reports.
If you are part of our Testing group, please use the corresponding channels. Otherwise, please either open tickets in the Zeek issue tracker, or report what you see in the #development channel on Slack. Don't be shy: anything is helpful when we're testing prerelease versions.