This repository was archived by the owner on Nov 15, 2023. It is now read-only.
This repository was archived by the owner on Nov 15, 2023. It is now read-only.
Rococo validator using paritydb stuck on startup with "Missing table, starting reindex" #13179
Description
I have several Rococo validator which remains stuck after startup. The only way to fix the issue is to restore the database from a backup.
Version: 0.9.37
Relevant flags:
--chain=rococo
--validator
--database=paritydb
--pruning=1000
--enable-offchain-indexing=true
2023-01-19 14:08:16.389 INFO main sc_cli::runner: Parity Polkadot
2023-01-19 14:08:16.389 INFO main sc_cli::runner: ✌️ version 0.9.37-645723987cf
2023-01-19 14:08:16.389 INFO main sc_cli::runner: ❤️ by Parity Technologies <[email protected]>, 2017-2023
2023-01-19 14:08:16.389 INFO main sc_cli::runner: 📋 Chain specification: Rococo
2023-01-19 14:08:16.389 INFO main sc_cli::runner: 🏷 Node name: rococo-validator-f-node-9
2023-01-19 14:08:16.389 INFO main sc_cli::runner: 👤 Role: AUTHORITY
2023-01-19 14:08:16.389 INFO main sc_cli::runner: 💾 Database: ParityDb at /chain-data/chains/rococo_v2_2/paritydb/full
2023-01-19 14:08:16.389 INFO main sc_cli::runner: ⛓ Native runtime: rococo-9370 (parity-rococo-v2.0-0.tx18.au0)
2023-01-19 14:08:27.060 WARN main parity-db: Missing table 09-19, starting reindex
2023-01-19 14:08:27.060 INFO main parity-db: Started reindex for 09-18
I also noticed that just before the node became stuck like this it experienced huge cpu/memory usage which caused several OOM kills. Maybe the database was corrupted by OOM kills ?
