Honeycomb changelog honeycomb.io

Honeycomb's Beeline for Go has gotten an overhaul! Our Beeline now comes with support for seamless tracing out of the box - the data model no longer simply supports the concept of "one current span," it now supports a doubly-linked tree of events (now dubbed "spans") representing a trace.

To see these changes in use, check out the new function calls StartSpan, AddFieldToTrace, and the wrapped RoundTripper! (See them all together in a single example.)

For more detail on the 0.2.0 release, see the repo's CHANGELOG.

We get it--not all your data gets updated all the time. (A few columns might have been collecting dust for months!). We'd like to make it easier for you to clean those up. Starting in June, we started tracking when we'd seen the latest update -- and now we're pushing it into the interface. In the dataset schema view, check out the new "Last Data Received" feature. Use it to debug your ingest, or to hide obsolete columns.

We've just made some ✨big improvements✨ to make collaborating with your colleagues (and your future self!) easier and better. Learn more about how to share the context in your brain and get greater #observability with Boards in today's blog post!

We’ve better surfaced the ability to "right-click-to-filter" table results via a visible button :)

On the query page, we now grab frequently used columns that your team queries most often and surface them to the top of your autocomplete, giving you more context about your data.

We’ve also noticed that many datasets contain stale columns — columns that you sent us data for at some point, but that we haven’t seen in a long time. They’re still around for posterity, but now we punt stale columns to the bottom of your autocomplete.

Query on, Garth!

Boards now have a new, shinier look to them:

And are now sorted in a new way! Instead of oldest boards being shown first, we’re now sorting by date (created_at, newest first), then the author (where we have one), then alphabetically by name.

Keep an eye out for even more Boards updates, coming to a Honeycomb near you!

We’ve been working on making navigating your Honeycomb environment simpler and more intuitive, and just deployed some changes that we hope you’ll agree have made progress toward that goal.

Here's what changed:

Read more about it in the blog post and let us know what you think!

It's now easier for customers to see, in aggregate, how their datasets are sampled.

Team owners can now click the “Explore Sample Rates” button in their Usage Center, which will take you to a special version of the query sandbox where 1) calculations aren't weighted by sample rate and 2) your events' special Sample Rate column is available to operate on.

You’ve always been able to name queries, but now you can also add a longer description to help explain the query to other folks on your team.

By default, queries are now untitled — we won’t clutter things up by naming them after the time you ran them.

We’ve also made a few changes to the placement of some of the query builder UI elements. Controls that affect how data is viewed still live close to the graph, but query controls like time selection, adding to a board, sharing, and the absolute/relative time toggle now live above the graph:

We’ve also added the option to clear the query builder, if you want to start fresh and build a new query!

Read more in our blog post!

For several common technologies, Honeycomb now provides Suggested Queries when you land on the query page. Now, we support Suggested Query Boards. This feature allows you to specify a Board as the source of suggestions for a dataset.

Picking a Board as a source for Suggested Queries will look like:

And these Suggested Queries will show up like:

Read more in our blog post!