pull down to refresh

node directions
229 sats \ 10 comments \ @billytheked 19 Aug bitcoin
as per #1059299
ive been having some issues with my node--
its running on an old office computer with an upgraded 2tb HD (see above)
i ran a few memtests and it seems to fail at anything over 4gb-- Claude says remove the memory cards and test again to further diagnose --
seems invasive to me -- and time consuming --
for a non-techy frugal gunslinger suches myself --what do stackers recommend my next course of action be??
from this standpoint it looks like I can try and upgrade the ram and/or hd OR just bite the bullet an buy an umbrel/start9 etc.
what do y'allz think
write
preview
related posts
11 sats \ 5 replies \ @Filiprogrammer 19 Aug
i ran a few memtests and it seems to fail at anything over 4gb
A failing memtest is a strong indication that something is wrong with your memory. At least one of your memory sticks seems to be faulty. If you have multiple RAM sticks in your system, you could try removing one and rerun the memtest and see if it passes. If it does pass, then the stick that you removed must have been the faulty one. You could then either proceed to run your system without the faulty RAM stick and just deal with the fact that you have less memory, or you could buy a new memory stick as a replacement.
reply
10 sats \ 3 replies \ @billytheked OP 19 Aug
ineresting -- thought youed like to know -- started removing memory cards and re-running memtester and found them all working -- changed slots etc and still good -- maybe dust/seating/corrosion issues that got worked out by sliding them out and re-inserting them -- also ran some SMART tests as per your other suggestion and got all good results -- anyway -- reindexing chainstate now and i will find out within the next few days i guess if everything is in order again -- wish me luck
reply
11 sats \ 2 replies \ @Filiprogrammer 19 Aug
Huh, that's interesting. Let's hope it really was just some dust, corrosion or improper seating. Good to hear that it works now! Hope it stays that way, I wish you luck!
reply
10 sats \ 1 reply \ @billytheked OP 21 Aug
Tried a reindex chainstate and got checksum mismatch errors -- went for full reindex and got this:
[error] A fatal internal error occurred, see debug.log for details: Corrupt block found indicating potential hardware failure. 2025-08-21T01:37:02Z [error] ConnectTip: ConnectBlock 000000000000005d0566652b2549637192d8830bcf524349f074c3eaa7d0b4da failed, bad-txnmrklroot, hashMerkleRoot mismatch
now im at a loss
reply
0 sats \ 0 replies \ @Filiprogrammer 21 Aug
Seems like your block 251415 is corrupted and possibly others aswell. This might be some leftover file system corruption from before. Did you run a file system check (fsck)? Even if that does not return any errors, some of the bitcoin block files have gotten corrupted. The safest way to completly get rid of the corruption would be to reformat the drive and redownload the blockchain. (as annoying as that sounds)
Also when formatting the drive, make sure to use a journaling file system or ZFS. This helps detect file system corruption and also repair it in some cases.
reply
0 sats \ 0 replies \ @billytheked OP 19 Aug
If you have multiple RAM sticks in your system
i think there's three. i can try this.
reply
11 sats \ 1 reply \ @DarthCoin 19 Aug
Claude says
2tb HD on old office PC
If is an old HDD you should forget using it. Replace it with faster SSD disk.
reply
0 sats \ 0 replies \ @billytheked OP 19 Aug
it is sdd -- see other post
reply
11 sats \ 1 reply \ @OT 19 Aug
You could try getting 8G of RAM and a 2TB SSD.
reply
0 sats \ 0 replies \ @billytheked OP 19 Aug
currently i have 16gb ram and 2tb hd
at least one ram card seems to be failing hence the diagnostics being suggested by Claude
other stackers have pointed out my hd might not be up to snuff
reply