gov: team relationship #122
Conversation
This comment has been minimized.
This comment has been minimized.
Signed-off-by: tison <wander4096@gmail.com>
|
Hi @hi-rustin ! Thanks for your comment. I have no objection on "team" but it is no the term in previous discussion. I'd like to see more feedback. |
|
For rust-prometheus repository, with a discussion on Slack with @lucab we agree on that it is better to include rust-prometheus as a team (project) under TiKV community. Then the question is who are the collaborators of that repository so that we can formalize one file to describe it. @breeswish @lucab @mxinden do you have an idea? I'd try to figure out the collaborators list but it helps if you can say someone who you think should be a committer / maintainer of the repository. |
project/pd/project.json
Outdated
| @@ -0,0 +1,31 @@ | |||
| { | |||
| "maintainer": [ | |||
There was a problem hiding this comment.
Given that PD and TiKV can't be used alone and PD committers are also actively updating TiKV, I suggest to merge pd project and tikv project.
/cc @nolouch
There was a problem hiding this comment.
There are many PD contributors that are not updating TiKV and PD can run with unistore or TiFlash (after it supports direct write) and there are proposals not TiKV related.
There was a problem hiding this comment.
There are many PD contributors that are not updating TiKV
You can find that all PD maintainers, committers have updated TiKV except qiuyesuifeng.
PD can run with unistore or TiFlash (after it supports direct write)
Unfortunately, unistore is not expected to be used in production, and it's purpose is to provide better optimization direction for TiKV. TiFlash is using TiKV to communicate with PD.
The main goal is to have a role that can decide from the global view and choose the design that best suit the evolution of product.
There was a problem hiding this comment.
TiFlash is mainly using the Raft layer of TiKV but not the whole TiKV.
IMO, PD provides the ability to schedule data partition (called Region) under multi-raft, which is a separate layer from TIKV, a specific key-value storage implementation.
There was a problem hiding this comment.
TiFlash is mainly using the Raft layer of TiKV but not the whole TiKV.
This doesn't change the fact that PD is talking to TiKV instead of Tiflash.
IMO, PD provides the ability to schedule data partition (called Region) under multi-raft...
It's not true. PD's behavior is tightly bound to the behavior of TiKV. Of cause you can simulate a TiKV using the same protocols, but its design is not for general KV storage.
There was a problem hiding this comment.
I think the proposal does not necessarily require separation, I would like to make one RFC for one feature, I think put them in one org is accepted. such as tikv/rfcs already has 0013-schedule-limit.md it only relative PD, and 0054-joint-consensus.md relative two project.
There was a problem hiding this comment.
@nolouch I think I get your point. We can label an RFC with related projects and require at lease X (maintainer|committer) approval of each project to get the RFC merged.
There was a problem hiding this comment.
Ping @BusyJay we may think it as consensus reached to leave TiKV & PD team as is if there is no more objection tomorrow (2021-06-16). We can discuss later on the description how to accept an RFC. Roughly we tag RFC with team label and require X committers of each team give LGTM.
Signed-off-by: tison <wander4096@gmail.com>
Co-authored-by: 二手掉包工程师 <rustin.liu@gmail.com> Signed-off-by: tison <wander4096@gmail.com>
Co-authored-by: 二手掉包工程师 <rustin.liu@gmail.com> Signed-off-by: tison <wander4096@gmail.com>
Co-authored-by: 二手掉包工程师 <rustin.liu@gmail.com> Signed-off-by: tison <wander4096@gmail.com>
Co-authored-by: 二手掉包工程师 <rustin.liu@gmail.com> Signed-off-by: tison <wander4096@gmail.com>
Signed-off-by: tison <wander4096@gmail.com>
@hi-rustin I'd suggest you comment on #118 since we define the term there. If there is no objection or we reach a consensus after 72 hours, i.e., this Friday 2021-06-11, I'm gonna update all description on weekend. |
Signed-off-by: tison <wander4096@gmail.com>
There was a problem hiding this comment.
Propose to promote @zhouqiang-cl as a committer according to his work on release these projects. Also the current situation is he already has write permission for these repository to push release process. We should include it in our gov model.
Signed-off-by: tison <wander4096@gmail.com>
Signed-off-by: tison <wander4096@gmail.com>
|
[REVIEW NOTIFICATION] This pull request has been approved by:
To complete the pull request process, please ask the reviewers in the list to review by filling The full list of commands accepted by this bot can be found here. Reviewer can indicate their review by submitting an approval review. |
Signed-off-by: tison <wander4096@gmail.com>
|
/merge /hold If you think we're ready to merge it now, please feel free to |
ti-chi-bot
added
the
do-not-merge/hold
|
This pull request has been accepted and is ready to merge. Commit hash: 4ff15e6
|
ti-chi-bot
added
the
status/can-merge
|
@nolouch: Thanks for your review. The bot only counts approvals from reviewers and higher roles in list, but you're still welcome to leave your comments. In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the ti-community-infra/tichi repository. |
Signed-off-by: tison <wander4096@gmail.com>
|
Thanks to @hi-rustin 's excellent work ti-community-infra/tichi#597 and with the consensus reached at here we are going to apply the change of community governance in days. |
|
/unhold |
ti-chi-bot
removed
the
do-not-merge/hold
|
/merge |
|
@tisonkun: In response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the ti-community-infra/tichi repository. |
|
Hi audience of this thread. We are going to apply the authentication rules proposed in #123 with ti-community-infra/configs#358 . Please be aware and report on #123 if you found any bad case. |
Signed-off-by: tison wander4096@gmail.com
close #121
description
These files describe relationship in
member : team : repository. They themselves do not take effect but for textual display.Since our permissions are flatten into repository level. @hi-rustin and I tend to use GitHub team to maintain the permission and leave the bot simply respect GitHub permission. See also #123
issues
There are two repositories those do not appear in these files:
community,rust-prometheus.For
community, its permission is to all maintainers only.For
rust-prometheus, its development is with collaboration with guys pay less attention to TiKV and not yet be included in previous SIG model.I'd like to listen to @BusyJay @breeswish @lucab @mxinden on suggestions. Personally I'm thinking of transfer to prometheus community and retain current permissions.