zydronium/manage.get.gov
1
0
Fork 0
mirror of https://github.com/cisagov/manage.get.gov.git synced 2025-08-05 09:21:54 +02:00
Code Issues Projects Releases Packages Wiki Activity

moved the bug triage section to operations/readme

Browse source
This commit is contained in:
Alysia Broddrick 2023-11-08 07:18:45 -08:00
parent 9f9048409e
commit 0c045429fd
No known key found for this signature in database
GPG key ID: 03917052CD0F06B7
2 changed files with 24 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

24
docs/operations/README.md
View file

@ -174,3 +174,27 @@ it with the latest model schema. Once launched, this should never be used on
the `stable` environment, but during development, it may be useful on the
various sandbox environments. After launch, some schema changes may take the
involvement of a skilled DBA to fix problems like this.
# Bug triage
Bugs on production software need to be documented quickly and triaged to determine if fixes need to be made outside of the normal release cadence. Triage levels will be Critical, High, Medium, and Low to indicate the level of priority for fix, not neccessarily the level of severity. See below for more details
**Critical**- should only be determined by the product owner and means the fix for this critical bug needs to have a quick fix for it created ASAP. This is the only case where a bug fix can be added outside of the normal release cycle and directly onto the stable release.
**High**- Can be determined by product owner or other team member, and indicates this bug is critical enough to warrant being added into the current sprint.
**Medium**- Should be added to a sprint coming up but is not blocking users, or enough users to warrant rushing it into a sprint
**Low**- A minor bug, that could even wait until after the next big launch date to be implemented.
## Steps for Triaging
1. When a bug is found, whether by a developer/designer or from feedback from an end user, a ticket should be made immediately. The actual maker of the ticket can be a member of the product team as needed.
2. This bug ticket immediately gets a priority added Critical/High/Medium/Low, with Critical requiring the product owner's consent.
3. Anything marked as `critical` should be refined immediately and engineering should be notified in our Slack dev channel that a Critical ticket has been created (if not already notified)
4. All items not marked as `critical` by the product owner can wait until refinement to be refined and may have their prioirty level changed during that meeting.
## Steps for dealing with Critical Bugs
1. Once the critical bug ticket is refined and the bug is clear, an engineer should be assigned to work on it. (No ticket, no work)
2. At the same point, two other engineers should be assigned to review the PR once it's made. One of the reviewing engineers can be subsititued for a designer if this is a design/content/other user facing bug fix.
3. In the case where the engineering lead is is unresponsive or unavailable to assign the ticket immediately, the product team will make sure an engineer volunteers or is assigned to the ticket/PR review ASAP.
4. Once done, the developer must make a PR and should tag the assigned PR reviewers in our Slack dev channel stating that the PR is now waiting on their review. These reviewers should drop other tasks in order to review this promptly.
5. See the the section above on [Making bug fixes on stable](#making-bug-fixes-on-stable-during-production) for how to push changes to stable once the PR is approved

25
docs/product/product_strategy.md
View file

@ -22,28 +22,3 @@ CISA lacks a scalable, efficient, and secure method of managing the .gov TLD pro
| **Data:** The program maintains authoritative contacts at, metadata about, and hostname information for all registered .gov domains, and is able to track that .gov domains are actually used | - Time-to-generate internal reports decreases <br /> - Results of periodic data quality audit show improvements month-over-month |
| **User satisfaction:** Getting a .gov domain is as easy and intuitive as possible | - Completion rate of form improves <br /> - Time from domain request to approval decreases <br /> - Number of domains requiring analyst data changes decreases |
| **Program reputation and experience:** The .gov program is viewed as trustworthy and responsive | - Response time for inquiries decreases <br /> - Resolution time decreases <br /> - Rate of repeat issues for tickets decreases <br /> - Number of SLTT organizations in Community of Practice increases |
## Bug triage
Bugs on production software need to be documented quickly and triaged to determine if fixes need to be made outside of the normal release cadence. Triage levels will be Critical, High, Medium, and Low to indicate the level of priority for fix, not neccessarily the level of severity. See below for more details
**Critical**- should only be determined by the product owner and means the fix for this critical bug needs to have a quick fix for it created ASAP. This is the only case where a bug fix can be added outside of the normal release cycle and directly onto the stable release.
**High**- Can be determined by product owner or other team member, and indicates this bug is critical enough to warrant being added into the current sprint.
**Medium**- Should be added to a sprint coming up but is not blocking users, or enough users to warrant rushing it into a sprint
**Low**- A minor bug, that could even wait until after the next big launch date to be implemented.
### Steps for Triaging
1. When a bug is found, whether by a developer/designer or from feedback from an end user, a ticket should be made immediately. The actual maker of the ticket can be a member of the product team as needed.
2. This bug ticket immediately gets a priority added Critical/High/Medium/Low, with Critical requiring the product owner's consent.
3. Anything marked as `critical` should be refined immediately and engineering should be notified in our Slack dev channel that a Critical ticket has been created (if not already notified)
4. All items not marked as `critical` by the product owner can wait until refinement to be refined and may have their prioirty level changed during that meeting.
### Steps for dealing with Critical Bugs
1. Once the critical bug ticket is refined and the bug is clear, an engineer should be assigned to work on it. (No ticket, no work)
2. At the same point, two other engineers should be assigned to review the PR once it's made. One of the reviewing engineers can be subsititued for a designer if this is a design/content/other user facing bug fix.
3. In the case where the engineering lead is is unresponsive or unavailable to assign the ticket immediately, the product team will make sure an engineer volunteers or is assigned to the ticket/PR review ASAP.
4. Once done, the developer must make a PR and should tag the assigned PR reviewers in our Slack dev channel stating that the PR is now waiting on their review. These reviewers should drop other tasks in order to review this promptly.
5. See the [Making bug fixes on stable during production readme](../operations/README.md) for how to push changes to stable once the PR is approved