Spike: are there serious issues with L2->L1 message ordering in sandbox?

[dbanks12]

If this is an issue then we need to do:

• Test to replicate: private kernel outputs L2->L1 messages in wrong order #1625
• Test to replicate: public kernel outputs L2->L1 messages in wrong order #1626
• And then come up with a workaround or print nice errors

[dbanks12]

@LHerskind said this is not a problem:

"The L2 -> L1 ordering should not matter. As they are not executed on L2 in the same tx, but deferred to L1, it should not care at all about the ordering of it. L1 won't know or care about the order, it will only know the messages. Ordering of that is handled with the designated caller."

[LHerskind]

Link to docs:

https://docs.aztec.network/aztec/how-it-works/l1-l2-messaging#designated-caller