rsyslog hit the one million message per second mark

Thanks to a user report, I just found out the the recent v5 versions of rsyslog have hit and exceeded the one million message per second (mps) mark! In my lab, with commodity hardware I reach around 500,000 mps. In practice this seems to be at least doubled :)

This strengthens even further that rsyslog is the syslogd of choice for demanding high-traffic environments :)

The full report is here:

http://kb.monitorware.com/000-000-character-messages-sec-with-rsyslog-t10740.html

a new rsyslog v5 beta – and focus on v6!

I have just generated a new v5-beta. It is being processed for release right now and will be soon. From the announcement:

This release both offers a set of new features and, at the same time, turns 5.7.3 into beta state. At first, this sounds a bit contradictory, but we do this for two reasons:

a) the new features introduced are non-intrusive in regard to the existing feature set, so no “bad surprises” are expected
b) other than that, primarily bugfixing went into this release, with
only few remaining issues being open

In order to move towards a new v5-stable, we consider it useful to begin with a new v5-beta stage. Note that this time the beta phase may be rather quick, because of the argument b) above. Note that we did not have any serious bug reports (except for one open issue) since December. Once the remaining issue is solved, we plan to do a short “proof in practice” and then move on to a new stable version.

In regard to new features, this release offers imfile multi-line capability, realtime UDP reception capability and better configurability for ommysql as recently announced for 6.1.3.

With this release, I also plan to conclude moving new features into v5 and try to focus on v6. This may not be practical in all cases, but I will try to stick to this plan as much as possible. During the past few weeks I have had considerable work to do just to integrate the various new features introduced in v4 and v5 into v6. While this was not really hard to do, it requires more than a little effort and very careful handling of the changes. The primary reason is that the code base diverged quite a bit and merging isn’t so much “fun” with that. More than once I even screwed up on some minor details. I hope that with a focus on v6 (for new features), I can spare a lot of time which than can go into new features.

And please do not misunderstand me: I focus on v6 for new features. This means I can focus even more on v5 in regard to correctness (bug-freeness). As usual, I prefer to fix issue in the oldest affected (and supported) release, and traditionally the version before the most current version branch has been very attractive to users because it has a near-complete feature set and a very strong focus on correctness.