A few words on Babylon

Babylon successfully activated last Friday and most of you are aware that the rivers of Babylon where not as calm as they could have been.
We’ll publish a blogpost that retrospectively highlights what went right, what didn’t, and what lessons we’re drawing from our point of view.

Here is a recap of what you might have missed:

-The formula for the block reward doesn’t exactly match with the one described in the Emmy+ posts. The bonus to include endorsements is too coarse leading to a slight decrease in inflation. This has no impact on the security of the chain, and it can easily be fixed with the next proposal.

-The network slowed down right after the migration due to a regression in mainnet (Because of our overly complex release process). This will be solved ASAP by having a single Tezos node which will be able to run all networks.

-A tiny automated airdrop of 1µꜩ was performed to prevent managers of migrated KT1s from having to pay origination fees twice. We apologize to authors of indexers that were disrupted for insufficiently advertizing the change.

In relation to this, the KT1 migration led to some managers temporarily being prevented from accessing their tokens because they couldn’t cover the transaction fee. All the affected accounts have been credited with enough to pay for a few transactions.

-The command line wallet Tezos client has a glitch where some amounts have to be provided in µꜩ instead of ꜩ amounts. This is being fixed.

Reminder: as expected, the gas costs are now updated (in particular for KT1 accounts that have been migrated to smart contracts). More info in our blogpost: https://blog.nomadic-labs.com/babylon-update-instructions-for-delegation-wallet-developers.html

In the meantime, stay tuned for our upcoming blogpost!


What is that second item about?: “network slowed down …”. Is that where Athens operations still in the mempool were causing errors?

1 Like

Yes that’s correct. There used to be a bug in the mempool. The nodes would receive malformed operations which caused them to crash. Although the bug was fixed fairly quickly, it was re-introduced in a mainnet release by mistake.

As a result, nodes that upgraded to Babylon stopped working because they saw lingering Athens operations that they did not understand. This led to them not being able to bake.
Shortly after the Babylon upgrade a patch was released to fix the issue. After about one hour, Athens operations became outdated and got flushed from the mempool.

1 Like