Today is an exciting day for Loki, as we have decided it’s time for Loki to graduate out of beta and into a 1.0.0 GA release!
It’s been just about a year since we announced Loki at KubeCon + CloudNativeCon North America in Seattle, and in that time over 137 contributors have made more than 1,000 contributions.
Here’s a look at where the project is today.
Stability and Durability
A few months ago, we added hosted Loki to the lineup of Grafana Cloud offerings. Loki had been stable for us internally in our ops cluster since the v0.2.0 release, storing around 40TB and half a trillion log lines each month.
We considered this a large enough volume over a long enough period of time to give us confidence that Loki can be run and scaled in a stable and reliable fashion. This is why we opened up our hosted offering to our customers’ production data. We feel our open source users should have this same confidence in running Loki in production as well.
Stability has a few definitions: From an operations standpoint, we are confident in Loki’s stability, but we should also talk about stability from an API and features point of view. Loki is still in most ways a very young project, and we plan to continue to add many features and a lot of functionality.
The 1.0.0 release also signifies our intent to be very cognizant of breaking changes and keep them to an absolute minimum. It’s difficult for me to say there will be no breaking changes with the early age of the project, but we will do our best to over-communicate changes and support backwards compatibility the best we can.
More information about the release can be found in the docs.
We are confident in running Loki in production with our logs – and yours – and we want you to be confident too. Releasing Loki at 1.0.0 makes it official!
Check out the blog for more content about Loki, including a case study describing how Paytm Insider, a popular platform for buying event tickets in India, has saved 75% of logging and monitoring costs using Loki in production.
You can also read up on key Loki features such as:
- loki-canary early detection for missing logs
- the Docker logging driver plugin and support for systemd
- adding structure to unstructured logs with the pipeline stage
- query optimization, parts one, two, and three