Cumulus test runtime: fix + cleanup#12231
Conversation
serban300
force-pushed
the
cumulus-test-runtime-cleanup
branch
4 times, most recently
from
302ea4e to
ca86dfc
Compare
serban300
added
the
R0-no-crate-publish-required
serban300
force-pushed
the
cumulus-test-runtime-cleanup
branch
from
ca86dfc to
0713eb9
Compare
serban300
changed the title
serban300
force-pushed
the
cumulus-test-runtime-cleanup
branch
3 times, most recently
from
bbbb1b3 to
a78af52
Compare
serban300
changed the title
serban300
force-pushed
the
cumulus-test-runtime-cleanup
branch
from
a78af52 to
b41d778
Compare
serban300
force-pushed
the
cumulus-test-runtime-cleanup
branch
from
b41d778 to
6b729ef
Compare
iulianbarbu
left a comment
iulianbarbu
left a comment
There was a problem hiding this comment.
Very nice, thank you for this cleanup! It conflicts a bit with #11991, and the associated PR opened for it, but I think this approach is cleaner. Adding new flavours can continue these fns with more ifs.
nit: What I'd do at most is extract them in a separate constants module, that explains a bit that these are knobs that we tweak in various ways to test certain configurations of parachains, via cumulus-test-runtime.
| velocity-3 = [] | ||
| velocity-6 = [] | ||
| velocity-12 = [] |
There was a problem hiding this comment.
Please comments on what they are doing
| .with_current_project() | ||
| .enable_feature("elastic-scaling-500ms") | ||
| .import_memory() | ||
| // A runtime with low slot duration of 500ms for low-latency testing with 12 cores. |
There was a problem hiding this comment.
Comment is wrong? Because this only enables 12 blocks, but doesn't change the slot time?
|
@iulianbarbu @bkchr @pepoviola @mchristou I addressed all the comments. Can you PTAL ? |
| "async-backing-v3" => 16..21, | ||
| "async-backing-v3-rpo" => 12..21, |
There was a problem hiding this comment.
we have some fails with lower values (https://paritytech.github.io/zombienet-jobs-monitor/web/?search=zombienet-polkadot-scheduling-v2-and-v3-collator-with-v3-validators&startDate=2026-06-02)
Why we are increasing the range?
There was a problem hiding this comment.
But the only failures that matter are the ones related to this PR. So only: https://github.com/paritytech/polkadot-sdk/actions/runs/26818212970/job/79069909542 . And I adjusted the range since then from 18..21 to 16..21
The failures related to other PRs don't contain the fix mentioned here: #12231 (comment)
Might hide issues for v3 scheduling tests
Co-authored-by: Iulian Barbu <14218860+iulianbarbu@users.noreply.github.com>
75fa398
5 participants
Closes #11991
Related to #11624
This PR: