[indexer] Get ready for nearcore 1.23.0

[khorolets]

nearcore is planned to be updated to the version of 1.23.0:

• testnet on Nov 29, 2021
• mainnet on Dec 13, 2021

Protocol upgrade is expected, also this update includes database migration which takes ~20 min on the regular node and ~5 hour on the archival one

Quick info

---

Commit hash 397a7a45b8efd3e4604374a4efabd567c03c0a19 1.23.0-rc.1
Expected issues: Indexer won't be working and available during the data migration (if you have more than one node, consider updating them one by one)
Reference: have a look at how we made it on NEAR Indexer for Explorer

Feel free to leave your questions in the comments below.

[ghost]

thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: Custom { kind: InvalidData, error: "Not all bytes read" }', /home/admin/.cargo/git/checkouts/nearcore-5bf7818cf2261fd0/397a7a4/core/store/src/migrations.rs:788:88

can't upgrade

[khorolets]

I've updated the commit hash recently, can you ensure you're using the latest one 397a7a4 ?

[ghost]

thats the hash that doesn't work. i was already aware the other hash was deleted near/near-indexer-for-explorer@0c5e9cc

[khorolets]

Can you provide more logs, what was the indexer doing before this message appeared?

[ghost]

2021-11-30T15:53:06.973913Z  INFO indexer: Load config from /home/admin/.near/testnet2/...     
2021-11-30T15:55:01.591235Z  INFO near: Opening store database at "/home/admin/.near/testnet2/data"     
2021-11-30T15:55:01.604904Z  INFO near: Migrate DB from version 29 to 30     
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: Custom { kind: InvalidData, error: "Not all bytes read" }', /home/admin/.cargo/git/checkouts/nearcore-5bf7818cf2261fd0/397a7a4/core/store/src/migrations.rs:788:88
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

[khorolets]

@frol @bowenwang1996 seems like data became corrupted during the migration

[bowenwang1996]

Protocol upgrade is expected, also this update includes database migration which takes ~20 min on the regular node and ~1 hour on the archival one

I think it definitely takes more than one hour for an archival nodes @khorolets. Where did you get the 1 hour information from?

[khorolets]

I've seen it somewhere in chats, but I've corrected to 5 hours which we've experienced on testnet

[Ljzn]

Hi, is this a MUST upgrade? We're running an old version RPC node. If the old node can keep working we don't want to bother our users.

[ghost]

yes or your indexer will stop working