Repository navigation
docs: tw-1122 - #3799
docs: tw-1122#3799pete-vielhaber wants to merge 12 commits into
Conversation
# Conflicts: # content/partials/_reference-sequencer-endpoint-behavior-partial.mdx
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
leeederek
left a comment
There was a problem hiding this comment.
Thank you Pete for putting this together! I appreciate it. There's alot of nuanced details that will require someone like @mglowack to help us confirm.
I also want to make a note that we will want to update all this when the change is live on arbitrum sepolia and then again when it is live on mainnet arbitrum one.
| @@ -0,0 +1,171 @@ | |||
| --- | |||
| title: 'Migrate your transaction submission code' | |||
There was a problem hiding this comment.
Generally, this sounds fine but won't it be a temporary page when we have yet to enable it formally?
There was a problem hiding this comment.
Could you clarify what you mean by temporary page, @leeederek ?
| | `timestampMin` / `timestampMax` | L2 timestamp window. | | ||
|
|
||
| - `eth_sendRawTransactionSync` submits a raw signed transaction and blocks until the node sees its receipt (or a timeout fires), returning the full receipt instead of just the tx hash. | ||
| - `eth_sendRawTransactionSync` submits a raw signed transaction and returns its receipt, or an error if a timeout fires first. Today the node returns the full receipt after the block is complete. A change rolling out to Arbitrum Sepolia first, then Arbitrum One, returns the receipt as soon as the sequencer executes the transaction. That receipt's `blockHash` is zero, because the block isn't complete yet. As a result, the method doesn't comply with [EIP-7966](https://eips.ethereum.org/EIPS/eip-7966), the Ethereum Improvement Proposal that defines it. For how it compares with `eth_sendRawTransaction`, see [Transaction submission behavior](/arbitrum-essentials/arbitrum-vs-ethereum/transaction-submission). |
There was a problem hiding this comment.
Let's make sure we include that the chain must be on the required version (3.13) and also that node operators need to set this flag --execution.rpc.tx-sync-forward=true
|
|
||
| </Callout> | ||
|
|
||
| Arbitrum chains give you two ways to submit a signed transaction. They differ in when they return and what they return: |
There was a problem hiding this comment.
Just as a note: there is a 3rd way to submit txns but we are removing it in Jan 2027 with a public deprecation notice posted separately. The language here is fine as is but wanted to flag this in case you wanted to say something more about it
|
|
||
| When a submission hits a queue limit, `eth_sendRawTransaction` behaves as follows: | ||
|
|
||
| - **Queue full:** it returns an error right away. |
There was a problem hiding this comment.
does the txn get dropped in this case too?
…transaction-submission.mdx Co-authored-by: Derek Lee <103802618+leeederek@users.noreply.github.com>
…transaction-submission.mdx Co-authored-by: Derek Lee <103802618+leeederek@users.noreply.github.com>
…ion-submission.mdx Co-authored-by: Derek Lee <103802618+leeederek@users.noreply.github.com>
…ion-submission.mdx Co-authored-by: Derek Lee <103802618+leeederek@users.noreply.github.com>
…transaction-submission.mdx Co-authored-by: Derek Lee <103802618+leeederek@users.noreply.github.com>
…transaction-submission.mdx Co-authored-by: Derek Lee <103802618+leeederek@users.noreply.github.com>
Apply the rest of Derek's review on PR #3799 and fix the formatting that broke the Gates and Build jobs. - Use Nitro 3.13.x and --execution.rpc.tx-sync-forward=true in the guide, the behavior page, rpc-methods and the published notice, so every page names the same version. - Mark eth_sendRawTransactionConditional out of scope and link its deprecation notice. - Add the 12-second default queue timeout to the deadline step, and state that a queue-full error means the sequencer never admitted the transaction. - Rename the "Stop treating..." heading and update its anchor link.
|
Thanks Derek. All suggestions are applied, and every page now says Nitro 3.13.x. Four open items:
|
The wider cells from the review edits left two tables unaligned, which failed the Formatting gate.
No description provided.