Recently it has become clear that there needs to be made significant changes to the SRE team. The current practices are unsustainable and have (in the past day or so) lead to the resignation of two volunteers, and may have contributed to the resignations of other SRE/system administrators in the past. What the team currently needs is twofold. First, we need strong, active leadership. I am confident that we can work with Reception123 and Southparkfan on this in the coming days and weeks. Second, we need to establish more policies and guidelines for SRE actions. There is a multitude of different areas to cover, and I myself will start with a "deployment policy." Somewhat included in this point, but also perhaps worth its own point is volunteer welfare. We need to improve this, and I'm not 100% certain of how we can best do that (cc @RobLa:). All I can say for now, is that if current members of SRE need to step away, or alleviate parts of their workload, to please do so. Your wellbeing comes first. All I ask is that if you need to step away, or leave some work unfinished, is that you document what you have already done, and leave notes for whoever picks up the task after you. -- Void Whispers 18:43, 14 June 2021 (UTC)
Proposed policies[edit source]
- Void Well let's be honest here: Since there's 3 Stewards now (including yourself), and there's a lot of resignations, why not appoint new Stewards and a couple of Global Sysops fill in the gaps? DarkMatterMan4500 (talk) (contribs) 19:21, 14 June 2021 (UTC)
- Stewards and Global Sysops are primarily a community function. Anyone is welcome to make a request. It is certainly worth considering though, as we need all the help we can get. I'm considering work on recruitment (and eventually donation) drives as well. But at the moment, I would like to make sure our technical team, that keeps the site running, can still function. Without them, the whole thing falls apart. -- Void Whispers 19:27, 14 June 2021 (UTC)
Here's a list of things I'd like to see, in no particular order:
- Deployments policy, working on (see link in above section). Goal is to prevent and limit SRE from deploying changes that may impact a community without properly communicating with said community.
- Accountability policy. I forget if we have one, but we should always make sure that any SRE member that deploys changes (particularly those that might break things) are available after the change to ensure nothing has broken, and properly deal with them if they do. Can also be sure to include requirements for availability during upgrade windows.
- Welfare policy. Something to make sure that everyone's workload is manageable, and that ongoing tasks can change assignments without significant trouble. Possibly also work in something about schedules and availability?
I'm just going to say basically what I said on the community noticeboard, but, as a former member of the Miraheze SRE team, I find it sad that the team has gone in the direction it has, and hope that this RFC will help make it better. I will also put it put there, that I'm willing to rejoin the team if SRE needs another team member. MacFan4000 (Talk Contribs) 13:03, 17 June 2021 (UTC)