Cluster-based Tests #9472
Comments
Merged
just-mitch
added a commit
that referenced
this issue
Nov 1, 2024
Effectively runs the transfer test twice with a reorg in the middle. To do so, I allow control of the host's k8s from within our e2e test container during the jest test. This allows us to programmatically from jest: - forward ports - install helm charts (like the chaos mesh one used to kill the provers for 2 epochs) - kill pods - wait for pods - etc Note: I expected to only need to wait 2 epochs for the reorg, and then be able to run more transfers. Instead I needed to wait 3 epochs after killing the provers for 2 epochs, otherwise I get `Tx dropped by P2P node`. I filed #9613 Other changes in this PR: - use p2p prover coordination in the prover nodes by default - run the reorg test on merge to master against 16 validators _**which I expect to fail**_ - run the reorg test nightly against 16 validators _**which I expect to fail**_ (yes, this is the same as the above, but the nightly should move to AWS after #9472) - restart the transaction bot in k8s when it fails - update the transaction bot to send 1 public and 0 private transactions by default - update the `install_local_k8s` script to install metrics and chaos mesh by default
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
We should have cluster-based tests on a test k8s namespace in aws running nightly
The text was updated successfully, but these errors were encountered: