Open pepoviola opened 1 week ago
@pepoviola
I see this test internally is using substrate_beefy_best_block
- why is it a beefy block and not just a best block?
Test is just using best block:
https://github.com/paritytech/polkadot-sdk/blob/6c2b46f9620fd147e63b151320cc145551fd2c18/substrate/zombienet/0001-basic-warp-sync/test-warp-sync.zndsl#L17-L19
And it seems that this number is not being synced fast enough:
2024-10-07T16:20:55.002Z zombie::network-node using comparator isAtLeast for 10797, 56687
2024-10-07T16:20:55.002Z zombie::network-node [alice] Current value: 10797 for metric substrate_beefy_best_block, keep trying...
2024-10-07T16:20:56.002Z zombie::network-node [alice] Fetching metrics - q: 174 time: Mon Oct 07 2024 16:20:56 GMT+0000 (Coordinated Universal Time)
2024-10-07T16:20:56.026Z zombie::network-node returning for: substrate_beefy_best_block from ns: _raw
2024-10-07T16:20:56.026Z zombie::network-node returning: 10997
2024-10-07T16:20:56.026Z zombie::network-node using comparator isAtLeast for 10997, 56687
2024-10-07T16:20:56.026Z zombie::network-node [alice] Current value: 10997 for metric substrate_beefy_best_block, keep trying...
2024-10-07T16:20:57.027Z zombie::network-node [alice] Fetching metrics - q: 175 time: Mon Oct 07 2024 16:20:57 GMT+0000 (Coordinated Universal Time)
2024-10-07T16:20:57.048Z zombie::network-node returning for: substrate_beefy_best_block from ns: _raw
2024-10-07T16:20:57.048Z zombie::network-node returning: 11197
2024-10-07T16:20:57.048Z zombie::network-node using comparator isAtLeast for 11197, 56687
2024-10-07T16:20:57.048Z zombie::network-node [alice] Current value: 11197 for metric substrate_beefy_best_block, keep trying...
Error:
[alice] Timeout(180), "getting desired metric value 56687 within 180 secs".
From a quick glance it looks like beefy sync issue?
I see this test internally is using substrate_beefy_best_block - why is it a beefy block and not just a best block? Test is just using best block:
Thanks for pointing this @michalkucharczyk, I think this is a bug. I will fix it and re-check this test.
Thanks for the help!
Hi @michalkucharczyk, I double check and the lines you mention are from other test (0001). This issue is for 0002 and is set to check that metric
Thanks!
Oh, sorry.
So it looks like 180s is not enough for beefy to sync the best block. @serban300 what are your thoughts on this? Do you think it can be some regression?
I don't think it's a regression, but I would have to look more closely. I'll look on it these days.
Examples (from yesterday 10 of 78 fails)
https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7528201 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7527201 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7526808 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7526157 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7523749 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7519971 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7518972 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7518884 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7518515 https://gitlab.parity.io/parity/mirrors/polkadot-sdk/-/jobs/7517649
cc: @paritytech/sdk-node