Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Create unit testing setup utilities that can produce a realistic Draupnir deployment #389

Open
Gnuxie opened this issue May 4, 2024 · 0 comments
Open

Create unit testing setup utilities that can produce a realistic Draupnir deployment #389

Gnuxie opened this issue May 4, 2024 · 0 comments
Labels
L5 All Users Likelihood P1 Nuisance Priority level T3 Workflow Issue User employs strategies to achieve the task that harm the experience (use minor first).

Comments

@Gnuxie
Copy link
Member

Gnuxie commented May 4, 2024

MPS already has utilities for describing a protected rooms set, we should use this to create realistic environments (for example 40 protected rooms with 50-1000 users each). This can then be used to profile and debug performance regressions in detail, in isolation or inclusive of the http layer and matrix servers. Which would have been very useful in the case of #380.
@Gnuxie Gnuxie added L5 All Users Likelihood P1 Nuisance Priority level T3 Workflow Issue User employs strategies to achieve the task that harm the experience (use minor first). labels May 4, 2024
Gnuxie added a commit to Gnuxie/matrix-protection-suite that referenced this issue May 4, 2024
@Gnuxie
Request RoomStateRevisionIssuers in order for protected rooms.
e57f266 
We believe this is the cause of a performance regression in Draupnir:
the-draupnir-project/Draupnir#380.
Unfortunately we don't have the infrastructure in our unit testing
or deployment to be able to truthfully understand the cause.

So we have created a follow up the-draupnir-project/Draupnir#389.
Our approach for now is just to guess some of the most glaring
changes that have caused the regression.

If this fails to make a significant difference, other culprits
are changes to the parsing code (ie, there was a recent version
of MPS where we moved away from accidentally using the empty
schema for most event content), including the recent upgrade
to typebox 0.32.22.
@Gnuxie Gnuxie mentioned this issue May 5, 2024
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
L5 All Users Likelihood P1 Nuisance Priority level T3 Workflow Issue User employs strategies to achieve the task that harm the experience (use minor first).
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@Gnuxie