Do you have the skills+time+interest to join the Tech Ops team?

I'm a sysadmin/developer and would love to help however I can.

It's been a few months since we discussed doing some custom development. I had more time available then, and I set myself to work on a blog and other things. The problem I found was that I'm not a DevOps guy. Whilst I understand and use Docker and the development cycle, it was very difficult for me to get anything done because of lack of documented architecture and spotty information available. It's nobody's fault, it's just how my brain works, I find it far, far less challenging to do technical work if the platform information is documented and available, instead of having to demand it on a per-case basis using a conversational process over chat.

Systems Administration is a difficult, demanding, and often unrewarding task, so it's totally understood that you guys have been doing excellent, world class work and a labour of love and us users and hanger outs should only be extremely grateful for everything you do.

That being said, I think that in order to bring people on board, there needs to be a rather detailed documentation of the architecture as it stands, especially since it's an evolving lifeform with increasing complexity. I think that documentation is a game blocker here, if you bring more people on board without an onboarding process in place, your work will be twice as hard repeating the process of teaching them every detaila and chatting for days and days every time a new tech person arrives, and they will always be lacking the level of detail required to do outstanding technical work. That method just doesn't work, and it will become impossible to maintain if the number of technical people increases. That's not a flaw in the teaching, it's that teaching person-to-person is not scalable and it's subject to time availability, reliability of the memory, and energies of the people who already have the knowledge.

I think it would be wise to at least consider the possibility of doing this documentation work first, before bringing any more people on board without a well-defined technical onboarding process. I think the lack of documentation and tech onboarding is a roadblocker here.

If you prepare a technical onboarding process furnished with detailed technical documentation, you will be able to get more programmers, sysadmins, devops, and other technical people with a predictable playground, well-defined boundaries, and the capability to take in more hands and brains to contribute more in parallel. And you won't have to ever do the (tiring, irregular, unpredictable) initial training ever again, because others will be prepared to answer questions and keep updating the tech knowledgebase so that everyone on baord is on the same page.

Much love and good vibes to you.

Thank you Horatio, I completely agree! We discussed this during the last tech call, and documentation is our first priority.

Either that, or outsource to a worker co-op.

Awesome then, I look forward to trying again as this project evolves!

I've not administered Mastodon before but I have worked as a software developer mostly focused on web technologies for 20 years (eek, I'm old). I've worked in Ruby before (the language Mastodon is implemented in) but not for a few years. I also sysadmin linux/ubuntu machines for the small center I work http://mith.umd.edu This mostly involves watching AWS instances. I don't have a lot of free time, but I would like to pitch in and help where I can. In my research I'm interested in how platform cooperatives can address the issues related to maintenance and repair of web services and architectures. So I would enjoy the opportunity to learn while participating in the work.

I was a sysadmin/devops for 12 years before moving exclusively into software development, where i've worked for 3+ years now. Have a lot of experience most of the tools listed, but I've never administered a Mastodon instance. I've not written much Ruby, but have written many languages with similar paradigms including Python and Node.js. My most recent work has been extremely Docker-involved, so I've got a lot to say there. My most limiting factor would be time availability, but would love to help out when I can.

Now that I'm retired I have more time, but I also have a lot of hobbies. I wouldn't want to be 'on-call' for outages, but wouldn't mind pitching in for less time-sensitive projects. Former I.T., spent many years futzing w/ Wordpress (including one year as front-end dev) - generally familiar with tech things. Currently running Pleroma instance on my Raspberry Pi and typing this from ArchLinux connected to my 4K TV. For reference LOL

Have used all the platforms you mentioned (not including AlpineLinux) in one way or another, so at least passing familiarity...

I'd like to be involved. I'm currently somewhat time limited but don't expect this to be forever, and I have a fairly wide variety of tech skills (development in most of the relevant languages, familiarity with POSIX environments)

I'm happy to help out where I can—I have ops and software development experience, but not a huge amount of time.

I haven't administered Mastodon specifically, but I'm a software dev with a bunch of sysadmin/devops experience on the side, including managing my own VPS for about a decade. I'm happy to help, particularly if we also start hosting other services and need people to help, but I think we should also consider compensation - not necessarily money, but membership in the hosting org for services rendered.

hi, I only joined yesterday so I'm pretty new, but what do I need to do to help out w this? as i put in the check thing, I built my own homebrew cluster in my bedroom to selfhost services for myself/my friends/my housing cooperative, and approximately 98% of my cycles are going to waste at the moment despite trying to selfhost anything and everything I can think of lol. I'd be more than happy to host things for social.coop - while I'd need to get paid so I can eat, I just finished up at university so I've already been trying my hardest to find a way to make enough money to eat/buy more hardware that doesn't involve having a boss, and being a part of a sysadmin worker coop would honestly be ideal. sign me the fuck up!

technical details of my cluster -
* i have 9 nodes, each w 16gb ram, a 1tb hdd, a 256gb ssd, and a dualcore cpu (except for the 2 that are quadcore).
* the host machines run debian/proxmox, and all the storage is networked together using ceph to create ~3tb hdd/~1tb ssd worth of shared storage for my VMs and docker containers.
* on top of that, I have a 9 vm docker swarm cluster(1vm/node) to actually run my containers in.
* I only have one ip to work with, so I reverse proxy all of my containers through nginx
* the entire cluster is fully redundant - all data in ceph is replicated three times and all vm disks and all docker volumes are accessible across all hosts, so you can cut the power to any individual node without losing any data and with only minimal downtime.
* upon the loss of a node, docker will automatically restart any killed services on other nodes, and ceph (which is honestly magic) will automatically start repairing and healing its storage cluster in the background.
* i designed my cluster this way so that it's not only fully redundant but also easily horizontally expandable with no requirement of homogeneity between nodes.
* to add more storage, you can just plug in more drives (regsrdless of size) and add them to the ceph cluster.
* more generally, basically any computer that can run debian (ie almost all of them) and has an ethernet port can be added to the cluster as needed as time goes on.

also, if I'm posting this in the wrong place, plz let me know. still getting the hang of loomio (which we could also selfhost btw!)

The Loomio devs have said staff time them far more than hosting thousand of non-paying groups, but I still like the idea of having more Loomio instance out there. I would like to see Loomio instances able to federate with each other, so I could login to Loomio.org and still participate in groups on loomio.social.coop as if they were still on the mothership instance.

Hello to everyone who's indicated they have at least some skills+time+interest to help with social.coop sysadmin!

Here's how you can help:

1. Please join the Tech WG on Loomio if you haven't already
2. Reply to the post about Tech Ops Team: Responsibilities & Initial Documentation with any clarifying questions to help flesh out the documentation
3. Help test/choose/setup a secure way to share passwords/keys
4. Figure out, amongst yourselves and/or in coordination with the rest of the membership on Loomio, how to organise the Tech Ops Team (there are already a few ongoing dicussions about this in various WG)
5. Tell me who to provide access to (SSH to the servers, and accounts on various 3rd party services), and where to store the shared passwords/keys
6. Start handling the Tech Ops Team Responsibilities A through G.
7. Profit! (i.e. make social.coop sustainable)

I would really appreciate if you can all pitch in at least in the begining, to try and reach step 6 by the end of the month, as I've now taken on a new job, and really need to free up more of my time.

As there was a lot great response from members on this thread on joining the Tech Admin Ops team @wulee is doing a great job getting things like initial documentation rolling in a proposal.. Please head over to weigh in before it closes and look out for an official Admin Ops Team formation proposal.

@horatiotrobinson @aaronwagener @robin012 @strypey @karlschultheisz @jeremyapthorp @christinahendricks @jankoscielniak @dp @michelslm @gregcerna @bilbono @fardog @dixongexpat @christianbundy @wakest @idmyn @edwardlplatt

@robertbenjamin I'm not currently a member of that group. I believe I've requested to join, though.

Sadly I'm very limited in the amount of time I can put in right now, as I'm trying to finish off abunch of stuff before my wife and I head off on a trip over July/ Aug (including Open 2018 in London). I won't be able to start putting regular time in until after we get back to China at the end of Aug.

I have lots of interest; some skills; and (at present) very little time :) I hope I'll have more time in the coming months. I think that @wulee's suggestion of breaking down the work into small issues to be picked off on gitlab will be beneficial for those who have limited time.

I have dev experience (happy picking up various languages, most prior experience in C# and PHP) and some ops experience (Linux server admin for websites/webapps.)

I'm sorry for being so late to this.

I will have time in July and August.

I'm eager to learn the relevant skills. I took university courses in system administration and programming, which I loved. I use Ubuntu. So, I think that I'm familiar enough with basic concepts to follow simple instructions and learn what I need to.

I also have a bunch of relevant skills but can't crrently commit to anything beyond trying to stay aware of what s.c and the tech WG are up to and offer knowledge if I happen to have some.