Originally published at: FMN Replacement Blog - December – Fedora Community Blog
Hello and welcome to another installment of the Fedora Messaging Notification (FMN) Replacement blog! It’s been about a month since our last update, and I hope that time was as productive for you as it was for us. We’ve been like Santa’s elves working tirelessly to deliver presents for all the good Fedora users out there.
Where have we come from?
In the previous blog post and the email all the way back in August, we pointed out the reasons why FMN is being replaced. For a refresher, or if this is the first time you’re seeing this, go ahead and read that post.
What have we done so far?
In the last month, the team pushed code to the staging environment and are bug hunting like professional pest controllers. Ryan Lerch, our front-end developer extraordinaire, has done an amazing job redesigning the UI/UX. That was one of the major issues with the “old” FMN. Aurélien Bompard, our intrepid tech lead has connected the front-end to our “back-end”, which is a bit of a misnomer. The back-end is really two different things as we’ve built it. One is the database, which Nils Philippsen and I have been designing and building, and the other is the consumer.
What’s a consumer? When the user is setting a new rule, say to receive messages from Bodhi about anything that concerns that user, the consumer portion of the back-end “consumes” the messages coming over the bus, consults the rule, and sends off that message notification to the user’s email, IRC, or Matrix.
What we’re focusing on
Now that we’ve successfully deployed to staging, testing is really the main issue for us right now. This is where you can play an integral part. To ensure that we’re not hammering services, we need to make sure that this code is tested to within an inch of it’s life. Two words: BUG and HUNTING. We plan on sharing staging address soon to see if there are any issues you can help us find (and fix).
Cacheing is another major ticket for us. We want to be able to cache recent queries so that we’re not always having to query FASJSON to save time and effort.