Child Pays For Parent (CPFP) is what you're thinking about. But if the fee for the parent transaction is too too low, it will not even be allowed into most mempools, so the child transaction with high fee will be stuck, unable to be mined.
Shit , I messed up here as well , I have a a lightning close that I accidentally put through at 1sat/vb , and then I used cpfp to bump up the fee . It's not been confirmed for like a week now , but I assumed that it was because of all of this stupid high fees stuff going on . How would you know if it is stuck ? Also , what do you mean "most mempools" , I thought that there is only one mempool?
reply
Each node has its own mempool with its own memory limit (300MB, 1GB, etc). Transactions also take some time to propagate between the nodes, so even same size mempools may have slightly different content.
Maybe check https://mempool.space (possibly over Tor) to see if the original tx-id is still in their pool. But what matters more is whether the original tx is still in the mempools of mining pool operators.
@wiz of mempool.space announced a Transaction Accelerator project but I don't know if it has come online yet.
Bitcoin Core devs are working on "package relay" to help the underpaying txs into the mempools together with their high-fee child, but work is not yet complete.
reply
I gotcha so how about transactions that were in the mempool but are no longer there? Should I submit the transaction when and if fees go down?
reply
Mempools will unfortunately not accept transactions under a certain fee-limit while they are still full. Waiting until fees go down seems to be the only option, or if something like the Transaction Accelerator comes online (see my other post in the thread). The Accelerator effectively would need to communicate with one or several mining pools to sneak the tx in.