Corrupt block found #26983

Expected behavior

I had an issue doing an IBD using this command: bitcoind -server -txindex=1 -par=15 -dbcache=16384

Actual behavior

It goes for a bit, but then when it gets to this block, it fails.

2023-01-28T00:10:20Z UpdateTip: new best=000000000000000661ca32dcfffcfbb4a290c86ccdef4e5444a6bbc45d367ade height=269433 version=0x00000002 log2_work=73.842515 tx=26980926 date='2013-11-13T15:58:39Z' progress=0.033783 cache=1038.3MiB(7715007txo)
2023-01-28T00:10:20Z *** Corrupt block found indicating potential hardware failure; shutting down
2023-01-28T00:10:20Z Error: A fatal internal error occurred, see debug.log for details
Error: A fatal internal error occurred, see debug.log for details
2023-01-28T00:10:20Z ERROR: ConnectTip: ConnectBlock 0000000000000000f2fae0af54fde656fb542eb80728d941ef4f92ba08c5da16 failed, bad-txnmrklroot, hashMerkleRoot mismatch
2023-01-28T00:10:20Z tor: Thread interrupt
2023-01-28T00:10:20Z Shutdown: In progress...
2023-01-28T00:10:20Z torcontrol thread exit
2023-01-28T00:10:20Z Loaded 0 blocks from external file in 360619ms
2023-01-28T00:10:20Z Shutdown requested. Exit ThreadImport
2023-01-28T00:10:20Z loadblk thread exit
2023-01-28T00:10:20Z addcon thread exit
2023-01-28T00:10:21Z net thread exit
2023-01-28T00:10:21Z txindex thread exit
2023-01-28T00:10:21Z msghand thread exit
2023-01-28T00:10:25Z opencon thread exit
2023-01-28T00:10:25Z DumpAnchors: Flush 2 outbound block-relay-only peer addresses to anchors.dat started
2023-01-28T00:10:25Z DumpAnchors: Flush 2 outbound block-relay-only peer addresses to anchors.dat completed (0.00s)
2023-01-28T00:10:25Z scheduler thread exit
2023-01-28T00:10:33Z Shutdown: done

After trying to start bitcoind again, I get this error:

2023-01-28T00:11:47Z *** Corrupt block found indicating potential hardware failure; shutting down
2023-01-28T00:11:47Z Error: A fatal internal error occurred, see debug.log for details
Error: A fatal internal error occurred, see debug.log for details
2023-01-28T00:11:47Z ERROR: ConnectTip: ConnectBlock 0000000000000000f2fae0af54fde656fb542eb80728d941ef4f92ba08c5da16 failed, bad-txnmrklroot, hashMerkleRoot mismatch

To reproduce

I can reliably reproduce this, it's just on IBD on mainnet.

System information

Bitcoin Core version v24.0.1 (release build)

Garuda Linux, Ryzen 5950, 128GB DDR4, Sabrent 4TB PCIe 4 NVMe

Bitcoin Core makes heavy use of CPU, RAM and disk IO. Hardware defects might only become visible when running Bitcoin Core. You might want to check your hardware for defects.

• memtest86 to check your RAM
• to check the CPU behaviour under load, use linpack or Prime95
• to test your storage device use smartctl or CrystalDiskInfo

Source: https://bitcoin.stackexchange.com/a/12206

I would look into the hardware you are using and the stability. Tested with your settings and I reached the chain tip without issue.

Note maybe rm -rf the contents of the blocks and chainstate dirs and start from scratch.

Not sure what to do here. Let us know if you have any other questions.