This is your weekly summary of Compose news for those changes which can make your life easier. In this edition, etcd 3 is updated, we close down etcd 2 deployments and we look at the new API endpoint for log file downloads.

etcd updated to 3.2.13

First up is the update to etcd 3.2, moving Compose to the latest etcd, 3.2.13. This is a normal minor version update, so there's no requirement to update as there would be if a security issue was involved. etcd 3.2.13 and the preceding 3.2.11 provided a variety of fixes, documented in the CHANGELOG, including options to use larger than 4MiB messages when talking to the server (removing previously hard coded limits).

etcd2 end of provisioning

At the other end of the etcd version train, we have finally removed etcd 2.x provisioning support. This end of provisions had originally been scheduled to occur in November, but we held off in response to allow some users to smooth their transition to etcd 3.x. We are still scheduling the end of etcd 2 support for the February.

New API endpoints for log files

We have just exposed two new endpoints to enable users to get access to log files. They are modeled on the backup endpoints we already offer, but where there's one backup for any day on Compose, with the logs there's multiple logs for any day because there's multiple capsules running the database and other elements. We have an article explaining more about log files and their endpoints and showing how, by means of a NodeJS application, you can retrieve files. If you want to get to work right away, the new endpoints 2016-07/deployments/:deploymentid/logfiles and /2016-07/deployments/:id/logfiles/:logfileid are available now.

Compose Articles

Since the last Noteworthy at Compose, we've had articles looking at PostgreSQL, Serverless computing with Compose, a review of database changes on Compose in 2017 and how to connect to Compose with MongoDB's Compass tool.

• Did you know there's a key/value store embedded in PostgreSQL? In Looking inside Books with PostgreSQL's hstore, we dived into how to apply HSTORE to handle flat but unstructured data related to book records.
• Serverless computing isn't about not having a server and all about a new, function-centric way of scaling your applications. With Going Serverless with Compose Redis, LornaJane looks at how you can plug Python and IBM Cloud Functions into Compose Redis to give your Serverless service persistence.
• Previously on Compose gave a full rundown of all the changes and enhancements that happened to Compose databases over 2017, database by database.
• And with the arrival of a community edition of MongoDB's Compass UI for MongoDB, we took a look at how to connect it to Compose's databases.

That's it for this week's Noteworthy at Compose. Onwards to next week!

Read more articles about Compose databases - use our Curated Collections Guide for articles on each database type. If you have any feedback about this or any other Compose article, drop the Compose Articles team a line at articles@compose.com. We're happy to hear from you.

attribution Kim André Fladen