erigontech
repeated prune error - prune history storage got invalid value length: 0 < 8
System information
Erigon version: v3.2.2
Erigon Command (with flags/config):
erigon --chain hoodi --datadir=xxx --prune.mode=full --nat=none --downloader.verify \
--externalcl \
--torrent.port=xxx --port=xxx --p2p.allowed-ports=xxx \
--http --http.addr=0.0.0.0 --http.port=xxx --http.api=engine,erigon,eth,net,web3 \
--authrpc.port=xxx --authrpc.jwtsecret xxx
Chain/Network: hoodi
CL: nimbus 25.9.2
Expected behaviour
No prune errors
Actual behaviour
repeatedly getting
[WARN] [11-11|09:57:14.876] [snapshots] PruneSmallBatches failed err="prune history at step 23 [0, 37500000): loadIntoTable : fn error: prune history storage got invalid value length: 0 < 8"
[EROR] [11-11|09:57:14.876] runPostForkchoiceInBackground error="[4/6 Execution] prune history at step 23 [0, 37500000): loadIntoTable : fn error: prune history storage got invalid value length: 0 < 8"
user reported at https://discord.com/channels/687972960811745322/1437441328349712445
full logs can be found at https://github.com/erigontech/security/issues/63
This is still broken on erigon version 3.3.0-a0c55b44. Actually, no, it reproduced with that version on the existing database, but seems to have worked without the error with a fresh resync.
Also it may mean hardware error. To check hardware: use tools like https://www.memtest86.com to test RAM and tools like https://www.smartmontools.org to test Disk
If hardware is ok, can run: “make db-tools” and “mdbx_chk -vvv datadir/chaindata”
I didn't check but I wouldn't be surprised if it was a hardware issue, so I would say deprioritise. Anyway I got my node working again with a resync on the new release.