-
Notifications
You must be signed in to change notification settings - Fork 71
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Docs: Who manages which resources that Kata uses #71
Comments
/cc @ttx |
Added links to some of the resources and an issue. |
Added OBS + Facebook. |
@grahamwhaley - what about that metrics site you were using? Also anyone know what the status of waffle.io is? |
It could be useful to add a couple of extra columns:
|
The metrics site was cauldron.io, which I just set up one under my name for Kata - but, it does not auto-update and I don't update it regularly - and - the OSF were setting up afaik a hosted version of similar (all based off https://github.com/chaoss/grimoirelab I believe) - so, I think we await that to arrive from the OSF and then list it. As for more columns - we could. Or we could just assume that the key stakeholders understand the recovery/backup, and will take action if contacted. The more columns we add, the more it needs to be kept up to date ;-) |
True. But I guess it depends what our plans are here. Is this a one-off exercise, or do we plan to keep the information current? I'd hope we opt for the latter and if so it would be good to get a feel for the bus-factor / disaster recovery exposure of all the externals we consume. Which reminds me, we should add our versions database which of course specifies all external software dependencies we use: All the URL refer to github.com (surprise! ;) but we also have a ref to https://www.kernel.org/ in there which should prolly be recorded. |
My plan was to record where all the 'secrets' and people and needs for 'special access' were, so we could ensure we have multiple people having access to each item (so we don't lose a critical key or password etc.). |
Ah - your scary bold text distracted me from the essence of the table ;) Well, ok, we could just do that. Or we could collate "everything". I totally agree we need to know who holds keys, etc. But bus factor also affects the services themselves surely? One other point: regardless of scope - it might be useful to identify which region each "manager" is in since ideally we'd have full regional cover. |
Added the kata bot github user. Also, I think the table may not be triggering github 'you have been referenced' nudges, so extracting the current list of usernames implicated here as a massive CC :-) Please check the table at the top of this PR and edit or let us know if you need add/removing from it! @alicefr |
For s390x also @jschintag and @tuan-hoang1 |
Thanks @alicefr - table updated. |
Added 'kata docker hub' to the table, for which I have no owners - @mcastelino @egernst @ttx - do you know, can you find out etc. ? |
@grahamwhaley Eric, Salvador and myself have rights on the kata dockerhub. |
/cc @ClaireMassey to help with filling in some of the opens. |
Hmm, this page/Issue has been fairly static now for a while, and is probably useful.... I'll try to get around to making it an official doc on this repo, but I suspect that won't happen in the next 1.5weeks.... |
Now we have the table of Kata resources and owners on a wiki page, let's link it from the main community README to make it easier to find. Fixes: kata-containers#71 Signed-off-by: Graham Whaley <[email protected]>
(this is WIP - I'll start adding details - but if you know of anything missing or have details to add either (if you can) add them yourselves, or leave a comment and I will fold them in. If you are mentioned below, please check you are in the right slot etc. :-)
Eventually this will become a markdown doc or wiki page - probably the former)
Kata uses a number of resources, particularly to drive its CI system for instance. Most of those resources require some sort of access control (keys, logins, rights etc.) to configure, deploy or manage. We don't have a definitive list of what those resources are, or who can access them to maintain them. This can be a fairly critical component to the project, given many of those resource are key blocking parts of the CI system.
Let's collect up a list of all the resources we know of and who currently has the ability to manage them - then we can decide if we need to have a central place to lodge 'credentials', and also if we need to diversify access to avoid any bus factor effects.
The text was updated successfully, but these errors were encountered: