This week: * Fixed some minor website url bugs. * Worked on identifying what parts of AMO need code fixes or supporting services for our version. * Started rewriting thundernest-ansible <https://github.com/Sancus/thundernest-ansible> documentation because it's woefully out of date. This was a slow week because I'm waiting for responses on a few AMO issues from Mozilla ops and the AMO team and I'm planning to poke them repeatedly this week and hopefully get more answers. I'm hoping to move forward with them hosting our initial version temporarily so that we don't need to worry about the data transfer until later, but in order to depend on that I need an OK from both the AMO team and Mozilla ops and I haven't received that yet. So I've proceeded working on the assumption that we may not get that. I also spent some time on some issues unrelated to AMO, but overall it was not a 40-billed-hours week. That happens, and I did quite a lot in the first few weeks this month, so I'm not worried about the overall pace. Communication with Mozilla people ebbs and flows, unfortunately. Next week: * Post completed, rewritten thundernest-ansible <https://github.com/Sancus/thundernest-ansible> docs. * Have an AMO deployment plan worked out, either on our infrastructure or Mozilla's. * Look into l10n process issues for the website, as we want to change some text but I'm not sure what we can rely on from Mozilla l10n yet. (Bug 1359297 <https://bugzilla.mozilla.org/show_bug.cgi?id=1359297>)

Andrei Hajdukewycz wrote on 29.08.2017 05:30: > This was a slow week because I'm waiting for responses on a few AMO > issues from Mozilla Hey Andrei, is there a master-plan, what kind of services we need to set up, that you work against? Could you maybe post that? If you're waiting on Mozilla, you could start with the next-most urgent server on the list. Ben

On 2017-08-29 4:39 AM, Ben Bucksch wrote: > Andrei Hajdukewycz wrote on 29.08.2017 05:30: >> This was a slow week because I'm waiting for responses on a few AMO >> issues from Mozilla > > Hey Andrei, > > is there a master-plan, what kind of services we need to set up, that > you work against? Could you maybe post that? > > If you're waiting on Mozilla, you could start with the next-most > urgent server on the list. Sorry about slow response to this. I try not to read the mailing list more than 1-2 times per week. I think this week's status report that I just posted should answer your question. When I was talking about services what I meant was that addons-server uses a bunch of stuff(external database server, redis, memcached, celery, elasticsearch, rabbitmq, and a reverse proxy cache in front of all of that) that are each run on separate deployed units(IE other virtual servers) in addition to the python code running on the primary virtual servers. This requires deployment automation for all of those types of virtual servers, and I haven't done that before so I spent some time working on it and also trying to figure out what we can cut out of that safely. I'm hoping to get access to the Mozilla ansible scripts for this to make it a lot easier... since I'm now going to have Mozilla ops support for a temporary site run on their infrastructure but with a fork under our control, it should now be easy to figure this out and I have plenty of time to do so since the temporary site they are hosting will be running for at least 3-6 months. Also, I know that sometimes the status reports are light on technical details. In general I try to document everything thoroughly once it's at least FUNCTIONALLY 'done'(so maybe 60%) and in a place where people other than me can use it, like the thunderbird-website, thundernest-ansible, and many other repos I've set up. But I don't always produce very much documentation during the development process and so I'm not always going to have an extensive step by step description of what I'm working on, cause any time I burn doing that is time billed that could have been spent just writing the code for it.