Corrupted Chain State

Corrupted chain state is the operational failure mode where a node's local database (block files, UTXO set, indexes) becomes internally inconsistent. The cryptographic chain itself is fine; the local copy is broken.

Common causes:

• Power loss or hard shutdown during a database write. LevelDB and Bitcoin Core handle most of these gracefully but not all.
• Disk corruption (failing SSD, bad sectors, RAID rebuild bug).
• Filesystem-level issues, especially over networked storage (NFS, sshfs - never run a node off these).
• Bitcoin Core software bugs during major version upgrades, occasionally.
• Out-of-memory kills mid-write.

Symptoms: the node refuses to start, reports "corrupted block database," fails to validate new blocks, or shows wildly wrong balances.

The fix ladder, easiest first:

• -reindex-chainstate. Rebuilds the UTXO set from the existing on-disk blocks. Fast if blocks are intact; cleans most corruption in the chainstate database.
• -reindex. Rebuilds both the block index and the chainstate by re-reading and re-validating every block file. Takes longer (hours), but doesn't require re-downloading.
• Delete chainstate/ and reindex. If -reindex-chainstate fails, manually remove the chainstate directory before reindexing.
• Delete and resync from scratch. Delete blocks/ and chainstate/, start fresh. Last resort; takes a full IBD time (12+ hours on good hardware).

Prevention is mostly operational: run on reliable hardware (consumer SSD with power-loss protection, ECC RAM if you're paranoid), use an uninterruptible power supply, don't run nodes on networked filesystems, and back up the wallet (not the chain state - the chain state is reproducible from the network).