You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Nov 28, 2023. It is now read-only.
Adding the x/nft simulation messages makes the Simapp tests very slow (46m) which is >2x than the avg sim time. This was experienced during the last commits of cosmos/cosmos-sdk#4838. The solution was to remove the nft simulations from the simapp until we solve this issue. This could probably related to either #6 or the custom (Un)MarshalJSON used by the nft module used to return an Object-like response instead of an Array.
The most affected simulation seems to be test_sim_nondeterminism, the other sim tests finish earlier but wait for it to finish. Weirdly enough, test_sim_nondeterminism doesn't run any invariants that could explain the bad performance.
I reduced the NFT operations weight and the number of accounts that hold NFTs (from 50% to 15%) but didn't improve the performance much.
yes it's still relevant as the simulations don't perform well with the current state of the NFT module. Moving this to the https://github.com/cosmos/modules repo
Summary of Bug
Adding the
x/nft
simulation messages makes the Simapp tests very slow (46m) which is >2x than the avg sim time. This was experienced during the last commits of cosmos/cosmos-sdk#4838. The solution was to remove the nft simulations from the simapp until we solve this issue. This could probably related to either #6 or the custom(Un)MarshalJSON
used by the nft module used to return an Object-like response instead of an Array.The most affected simulation seems to be
test_sim_nondeterminism
, the other sim tests finish earlier but wait for it to finish. Weirdly enough,test_sim_nondeterminism
doesn't run any invariants that could explain the bad performance.I reduced the NFT operations weight and the number of accounts that hold NFTs (from 50% to 15%) but didn't improve the performance much.
Version
a90002f
Steps to Reproduce
(see 68a1a9c)
cc: @rigelrozanski @alexanderbez @okwme
For Admin Use
The text was updated successfully, but these errors were encountered: