Bump a timeout in zombienet coretime smoke test #6268
Conversation
tdimitrov
added
R0-silent
| @@ -9,7 +9,7 @@ coretime-collator: is up | |||
| alice: js-script ./0004-configure-relay.js with "" return is 0 within 600 secs | |||
|
|
|||
| # Coretime chain should be producing blocks when the extrinsic is sent | |||
| alice: parachain 1005 block height is at least 10 within 120 seconds | |||
| alice: parachain 1005 block height is at least 10 within 180 seconds | |||
There was a problem hiding this comment.
is this so high because we need to wait two sessions?
There was a problem hiding this comment.
Afaik it's one session (10 blocks which is around a minute). On top of this we need some time to get to 10 blocks worst case (for my local run it was around 7 block) which is another 40 secs. Or 100 secs total.
Initially I thought 120secs should be enough to finish this but I noticed a job failed at this step so I decide to put one more minute to be on the safe side.
There was a problem hiding this comment.
hmm but with 180 seconds minus the 60 for the session change, this would basically mean 12-second blocks. isn't it a bit too much? I'm afraid of adding timeouts that are too large which would then mask other issues.
There was a problem hiding this comment.
maybe let's leave 8 seconds for each block (which is what we do in other tests AFAIK)
There was a problem hiding this comment.
Just to make sure we are on the same page. As soon as ./0004-configure-relay.js executes the coretime chain stalls. It unstalls after one complete session.
So we have:
- a few blocks to let the old session finish (let's say 5?)
- 10 blocks for a new session
- 7 blocks to reach the threshold after the change kicks in
(5+10+7)*8=176 secs for a timeout which by pure luck is around my estimate.
Maybe it's safe to decrease the threshold to 6 with a timeout of 150 secs?
There was a problem hiding this comment.
So we have:
a few blocks to let the old session finish (let's say 5?)
10 blocks for a new session
Ok so indeed two sessions need to pass. It's ok then
There was a problem hiding this comment.
Sorry, I've misunderstood you initially. I thought you mean two full sessions.
|
/cmd prdoc --bump none --audience node_dev |
|
Command "prdoc --bump none --audience node_dev" has failed ❌! See logs here |
|
/cmd prdoc --bump no_change --audience node_dev |
Signed-off-by: Oliver Tale-Yazdi <[email protected]>
polkadot/zombienet_tests/smoke/0004-coretime-smoke-test.zndsl still timeouts on CI from time to time. Bumping the timeout a bit more.
Related to #6226