Never saw these....
If you run a Hive-Engine node, you have probably seen some version of this:
the node restarts and throws "block not found" errors
you see "duplicate transaction" errors on startup
the node needs to be restarted multiple times before it "goes on"
Under what conditions?
I lose power frequently, more than I like, but also when I do kernel updates and reboot, almost any time I restart the service I have issues.
Would be nice to narrow these. As I restart my node at least once a month and I have never seen anything like these.
Probably a lack of resources thing? I am testing with a 15 year old machine and I have never had problems too..
I have a big machine too, my main server... so surprised about these...
Share more details internally... this does not sound at all like something I was expecting...
Could be similar to the issues I had with io on the ssd, still don't have an nvme.
But still the replica set transaction wasn't implemented right. Right now it:
writes block
starts processing transactions
if something happens here - block is already written
whereas it should be:
block write pending completion of transactions or rollback.
Don't recall this ever; very strange... What Linux kernel are you using?
Please tag me on the commit/issue where you think this is happening.
hummm I don't have anything to test kernel 7 yet...
But this should not be a target for fixes... as this is very recent and will have a lot of "BUGS", even if RUST based! Might need new adapted libraries...
"forky" should be still 6 kernel... so you might have some updates mixed. Unless they released a new stream and I haven't yet found out about it...
Note: Forky here is not related to me...
It's been an ongoing issue for as long as I've tried to run a node, earliest kernel i can remember off the top of my head was before debian 13 was released, so 6.12.xx(ish), I just got tired of restoring snapshots and looking into why it freaks out on restarts.
Hummm, need to try my Debian machines then... never seen this.
Suspect something else is happening... thanks for this. Will get back to you!
add vm.swappiness=0 into your /etc/sysctl.conf 😎
Swap usage is not needed unless the application needs full memory.
