pull down to refresh

Bitcoin node suspected hardware/disk issue
1067 sats \ 7 comments \ @unschooled 29 Jul bitcoin
I suspect im dealing with a hardware issue. here are the logs from my bitcoin core node, which is getting a database corruption fatal error, causing it to enter a cycle of repeated restarts. A clean reboot usually resolves this, but it happens somewhat periodically.
a similar error caused my fulcrum server also to go down a little while ago.
the reason i think it might be hardware related is i am running my node on an old office computer. to my knowledge, its specs check out (16gb ram, 2tb HD, Intel quad-core processor), but i thought id see what the community here has to say.
any thoughts or ideas?
2025-07-29T11:43:54Z UpdateTip: new best=00000000000000000001f4cd0fe5496a8e5b24b7c5cf5bd81df133b5c9441290 height=907653 version=0x2e000000 log2_work=95.740849 tx=1219799617 date='2025-07-29T07:06:35Z' progress=0.999908 cache=39.6MiB(299442txo) 2025-07-29T11:43:55Z New outbound-full-relay v1 peer connected: version: 70016, blocks=907673, peer=7 2025-07-29T11:43:56Z LevelDB read failure: Corruption: block checksum mismatch: /home/satoshi/.bitcoin/chainstate/321640.ldb 2025-07-29T11:43:56Z Fatal LevelDB error: Corruption: block checksum mismatch: /home/satoshi/.bitcoin/chainstate/321640.ldb 2025-07-29T11:43:56Z You can use -debug=leveldb to get more complete diagnostic messages 2025-07-29T11:43:56Z [error] Error reading from database, shutting down. 2025-07-29T11:43:56Z [error] Error reading from database: Fatal LevelDB error: Corruption: block checksum mismatch: /home/satoshi/.bitcoin/chainstate/321640.ldb
write
preview
related posts
160 sats \ 4 replies \ @optimism 29 Jul
This is often a disk (or disk-controller) issue. What disk do you use?
reply
100 sats \ 3 replies \ @unschooled OP 29 Jul
Lexar NS100 if memory serves. i can confirm this later.
reply
0 sats \ 2 replies \ @unschooled OP 29 Jul
Yep, this is the one. I ran a few stress tests and it seemed to be ok. would you recommend another brand?
reply
312 sats \ 1 reply \ @Filiprogrammer 12h
A year ago I cheaped out when buying this exact 2TB SSD. At the time of buying I unfortunatelly did not pay attention to the fact that it does not have DRAM cache. This causes it to slow down to a crawl during high write pressure. Also it only took me a few months for the first I/O errors to start appearing with this SSD. Next time you buy an SSD for a Bitcoin node you should make sure that it has DRAM cache. When it comes to SSD brands, Crucial and Samsung are usually good.
Coming back to your problem, you might want to run a SMART test on your SSD and see if it reports any errors. Also check your logs for any file system errors. It could theoretically also be a RAM issue. So might make sense to run a memtest.
reply
120 sats \ 0 replies \ @optimism 11h
I second Crucial. Samsung - I'm currently testing a 990 EVO Plus, but that's M2, since 3 months. So far so good. This only cost me ~$50 per TB, so it wasn't crazy expensive either.
reply
110 sats \ 0 replies \ @gregtonoski 29 Jul
I would suggest reporting the issue at the "https://github.com/bitcoin/bitcoin/issues".
reply
0 sats \ 0 replies \ @LightningG 13h
From what I understand from the log it tells us that LevelDB read failure and block checksum mismatch mean that data read from disk is not matching expected integrity checks. Did you try running a SMART test? If SMART shows issues or fsck errors, swap the HDD for a new SSD
reply