Closed bombnp closed 3 weeks ago
Hello @bombnp, thank you
Unfortunately we do not support testnet3 since it has more than 10 blocks reorg constantly.
For testnet4, there is an experimental branch of OPI that you can use and since it only has around 50k blocks, it syncs very fast.
Do you mean you do not provide backups for testnet3, or OPI does not support testnet3 at all? The code seems to allow setting network as testnet.
It did support testnet3 at first, before the difficulty bug is used widely. However currently it gets stuck a lot since the ord version that OPI relies on does not support more than 10 blocks reorgs.
I see. If I decide to run OPI on testnet3 anyways, and problems occur, are those problems recoverable? Or will it be a pain to resolve (from your experience)? Asking for advice.
I'm testing out some brc20 apps and testnet3 is the only option for me right now.
If a reorg occurs, index.redb will not be able to rollback so you have to start syncing from the beginning. However you can change the rollback height limit parameter in ord and rebuild it
You mean SAVEPOINT_INTERVAL
and CHAIN_TIP_DISTANCE
variable in this file? Do I need to change anything else?
https://github.com/bestinslot-xyz/OPI/blob/main/ord/src/index/reorg.rs#L23
yes, you should also edit https://github.com/bestinslot-xyz/OPI/blob/5b96207e1e84cc23594c41a71d47c1d475288000/modules/brc20_index/brc20_index.py#L680 this line and increase the limit in the query, otherwise brc20_index will stop after a reorg.
Awesome, thank you so much, @samedcildir !
Hi @samedcildir , thank you for your work.
I see that there is a
restore.py
file for restoring backups (both redb and postgres), but it seems to only work with mainnet? Are there backups for testnet as well, or do I need to start from genesis?