-
Notifications
You must be signed in to change notification settings - Fork 122
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
process: update Node.js sec membership policy #558
Conversation
sam-github
commented
Jul 4, 2019
•
edited
Loading
edited
@nodejs/tsc @nodejs/security There is something I'm not clear on, is everyone who has access to the Node.js program in H1 on the Triage team? All those people have the ability to triage issues, but I'm not sure that you all will triage issues, and want to triage issues. A number of the TSC members may have access to H1 because of a responsibility to be aware of activity, but not out of an intention to be actively involved. Does it even matter? Maybe not, but if none of the "triage team" is actually triaging, things will go badly. Then again, perhaps that's best dealt with in the weekly TSC meetings, if there are open vulns that are not getting dealt with. I bother bringing it up because the old triage team was small, and with H1, its now quite large, and its not obvious that the everyone who accepted the H1 invitation realized they were implicitly joining the triage team. I'm a big fan of simplifying the complexity of the team structure. It does appear there are unavoidably, at this moment, 3 actual resources, and corresponding lists of people with access:
It is possible that we can force the latter two github lists to have the same members (manually, or automatically, don't we have some tools somewhere for updating markdown membership lists from github APIs?), in which case, we only have to document two lists: triage team and security team. What should we do here? |
and to further flood mailboxes, sorry:
The triage team policy used to be that everyone individually PRed their name here, indicating acceptance of the privacy policy. I'm not sure that happened for the TSC when they got H1 access. I'm OK with removing the process, or following the process, but we should probably do one or the other. |
Yes: https://github.com/nodejs/node-core-utils/blob/master/docs/ncu-team.md |
2e07e41
to
0ba7cda
Compare
0ba7cda
to
4bfdba9
Compare
Taking a quick look, seems OK to me. I'm not on Security WG so I'll leave it to actual members to green-check this, but 👍. |
@Trott You are among the correct people to review this, IMO. While the process wording needs updating because triage moved to H1, it is the sole responsibility of the TSC to determines the membership of the Node.js triage team, see https://github.com/nodejs/security-wg/blob/master/processes/security_team_membership_policy.md @nodejs/TSC, PTAL |
f5f47b7
to
57df1e6
Compare
... I updated the process wording, by moving its text into the same file as the membership privacy policy/membership list. So, that's all cleaned up now. |
@nodejs/tsc, PTAL - it is TSC's preexisting responsiblity to appoint sec team members, I updated the lists of members, and made minor changes to the policy falling out of the HackerOne use, which moved the location of issues out of github (so access to it is not controlled any longer by a github team). |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM with two tiny typographic changes to fix GitHub markdown rendering.
fc076f8
to
d06ca0b
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
Some background I think that is relevant:
I don't think there was any intention to change the composition of the triage team by adding the TSC members to H1. ie it still makes sense to have people PR themselves into a list that identifies who the core security triage team is. |
@mhdawson I am also unclear about who is a member of the triage team, see my questions here: #558 (comment) The triage team was a mailing list of people who received notification of new vuln reports. That mailing list no longer exists. With H1, everyone who is a member of the H1 program gets notifications, but I agree, its not clear which (if any) of them is on the triage team. I'd add an extra section specific to triage, but I don't know who the current team is. Perhaps a re-call for volunteers should be made at the next TSC meeting? |
I agree re-validating who is currently on the core triage team would make sense. @mcollina, @MylesBorins, @jasnell as well as yourself are the people who I've seen doing that work recently (apologies in advance to those that I've missed as this is just based on my recent memory of activity notifications I've gotten for H1). Maybe start with this list and then confirm those members and ask if we've missed anybody? I suspect @mcollina, @MylesBorins, @jasnell will be able to fill in the gaps. |
Everybody on the TSC has access to H1 because it’s there that the main conversation happen about vulnerabilities. Initial triaging is facilitates by H1 staff, so it’s not a burden. The current problem is a) deciding what to do for the borderline cases and b) fixing protocol-relates issues (http, http2, etc). Some of those are extremely hard, and there are not enough people engaged in the process. |
@mhdawson PTAL. Triage team is restored to its previous value. |
16d6205
to
e684eef
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
FTR, removing @mhdawson from triage, he shouldn't have been there, he only had access to H1. See #414 (comment) |
The policy was spread across two files, had mild contradictions (one doc said membership was confirmed by the TSC, the other said by it was confirmed by the current members), referenced the unused github issue tracker (HackerOne is used now), and had lots of mention of the @nodejs/security team, which does not in itself have access to ANY private issues or patches (I removed those refs). The policy is essentially unchanged, just more readable, and in one file so it shouldn't drift out of sync as easily.
cd10414
to
dfc7526
Compare