-
Notifications
You must be signed in to change notification settings - Fork 6k
Community Duty Tasks
One day each week, each contributor should do the following:
To triage an issue, search for the label status: waiting-for-triage
and follow these steps.
If it's something you feel like you cannot triage, assign it to another team member who you feel can.
Otherwise, do the following:
- Assign the ticket to yourself.
- If the ticket appears to be accidental, incomplete, or not following our code of conduct, close the issue with
status: invalid
(i.e. a 400 error). As needed, explain why it was marked as invalid in a comment. - Label the issue with the appropriate
in: xyz
label. - Correct any incorrect labels.
- If it's a duplicate, label it as a
status: duplicate
and close the issue with a comment linking to the issue. - If it's a question, label it as
type: stackoverflow
, and close with a comment inviting them to use StackOverflow
(If a feature is brand new, still consider answering the question. The reason for this is that the likelihood that it is a bug or a much-needed feature is higher and so it is worth it to allow those questions as an exception to the rule.)
If it's a type: bug
, also do the following:
- Consider the urgency of the issue:
- Does it affect many users? ⬆️
- Has it been around for a long time and we are just hearing about it now? ⬇️
- Is it part of a new feature? ⬆️
- Is there an easy workaround? ⬇️
- Reproduce it. If you don't have enough information, ask and label with
status: waiting-for-feedback
. - If not a bug, label with
status: declined
. - If a high-urgency bug, assign to the next patch release of the earliest supported version.
- If a low-urgency bug, assign to the earliest supported
.x
milestone. - Develop a workaround and post it as a comment.
- If an ideal-for-contribution bug, label it as
status: ideal-for-contribution
and invite the poster to contribute - Add any additional needed explanation in a comment.
- If at this point the bug is addressed, close it.
If it's a type: enhancement
, do the following:
- Evaluate it. If you don't have enough information, ask and label with
status: waiting-for-feedback
. - If you disagree, mark the ticket as
status: declined
. - If you agree, assign the appropriate milestone; either "General Backlog" or the next
.x
generation where it will fit. You might remind the contributor that tickets with votes usually happen before tickets that don't. - For an ideal-for-contribution feature, label it as
status: ideal-for-contribution
and invite the poster to contribute. - Add any additional needed explanation in a comment.
- If at this point the feature is addressed, close it.
At this point, also remove the status: waiting-for-triage
label.
If there is more work to be done and you want to do it, leave it assigned to yourself; otherwise, unassign.
Look for unassigned status: feedback-provided
issues and follow the same steps as Triage Issues.
For each Dependabot PR, investigate why it failed to merge and address the issue, ensuring that the upgrade gets performed. If upgrading is impossible, add those details to the ticket.
For each contributed PR:
- Follow the same rules as Triage Issues.
- If ready to merge, merge the PR. These are usually unassigned PRs where any requested changes have been approved and the issue has a concrete milestone and not just a
.x
milestone. - If there is an issue linked to the PR, close the issue, mark it as
status: duplicate
, and include some comment likeSuperceded by {the PR number}
. - If it is a simple PR, like a typo, an obvious fix, a formatting or naming convention improvement or the like, ready it for merging.
- Search for questions on StackOverflow with the tags
spring-security
,spring-session
,spring-security-oauth
,spring-authorization-server
, andspring-ldap
. You might also search for those withspring-boot
that also mention security. - As part of your answer, if there is an improvement that can be made to Spring Security that this question uncovers, open an issue or provide the improvement. An common example is the need for clearer documentation.
For each PR, do the same as Triage Dependabot PRs.
- Check the internal channel and either answer questions or ping the right person.
- Check the gitter channel and either answer questions or ping the right person.
The current schedule is: