-
jwinterm
is there a way to compile or run monerod without huge pages/huge page support?
-
jwinterm
-
moneromooo
It's not fatal, it goes on to allocate "normally".
-
gingeropolous
jwinterm, there should be... thats sorta the point of the "lite" mode for randomx. im guessing it tries to run it big and then defaults to lite mode
-
moneromooo
If it's the spam that's the problem, I think yo need to renmove the first alloc call.
-
jwinterm
I think it's the spam that's the problem
-
gingeropolous
those INFO messages
-
jwinterm
.shrug
-
moneromooo
Line 225 in rx-slow-hash.c
-
gingeropolous
ima make tshirts. INFO WARN ERROR
-
moneromooo
cache = randomx_alloc_cache(flags | RANDOMX_FLAG_LARGE_PAGES);
-
moneromooo
Skipping that should do it.
-
asymptotically
wasn't there an environment variable to control it?
-
gingeropolous
i thought there was too, but i just scanned the help and didn't find a flag
-
gingeropolous
oh you mean export=something do magic
-
jwinterm
yea I looked at monerod help and didn't see option
-
jwinterm
set variable before compile?
-
gingeropolous
.dunno
-
moneromooo
I remember talk of one, but git grep getenv does not pick up anything in randomx. Maybe only in new versions.
-
gingeropolous
i mean, perhaps a user would want to thrash their CPU instead of use a lot of memory
-
moneromooo
/window 27
-
gingeropolous
nowindowforyou
-
hyc
use MONERO_RANDOMX_UMASK env var
-
hyc
you'll need to check randomx.h for what the flag values are
-
hyc
jwinterm: ^^
-
hyc
-
Feuermagier
I ran into an error while importing the chain from file
-
Feuermagier
block height: 1959719
-
Feuermagier
block height: 1964659
-
Feuermagier
block height: 1968799
-
Feuermagier
Done scanning bootstrap file
-
Feuermagier
Full header length: 1028 bytes
-
Feuermagier
Scanned for blocks: 59216999345 bytes
-
Feuermagier
Total: 59217000373 bytes
-
Feuermagier
Number of blocks: 1968800
-
Feuermagier
2020-03-10 22:07:56.796 I bootstrap file last block number: 1968799 (zero-based height) total blocks: 1968800
-
Feuermagier
Preparing to read blocks...
-
Feuermagier
2020-03-10 22:07:56.842 I bootstrap file recognized
-
Feuermagier
2020-03-10 22:07:56.842 I bootstrap::file_info size: 4
-
Feuermagier
2020-03-10 22:07:56.842 I bootstrap file v0.1
-
Feuermagier
2020-03-10 22:07:56.842 I bootstrap magic size: 4
-
selsta
please use paste.debian.net
-
Feuermagier
2020-03-10 22:07:56.842 I bootstrap header size: 1024
-
Feuermagier
2020-03-10 22:07:56.842 I start block: 1696170 stop block: 1968799
-
Feuermagier
2020-03-10 22:07:56.842 I Reading blockchain from bootstrap file...
-
Feuermagier
2020-03-10 22:09:33.873 F ERROR: unexpected end of file: bytes read before error: 0 of chunk_size 30702
-
Feuermagier
Bus error
-
selsta
the file is corrupted
-
selsta
I would sync using monerod, it is way easier
-
Feuermagier
-
Feuermagier
Yes, I'm trying to do that now.
-
Feuermagier
However
-
selsta
Did you continue the download somehow?
-
Feuermagier
-
selsta
Network shares are not supported.
-
Feuermagier
Now it complains about a read only database
-
selsta
I would recommend to use an SSD + pruning
-
selsta
only takes 25GB and fast
-
Feuermagier
What is pruning?
-
selsta
Reduces the file size from 70GB to 25GB
-
Feuermagier
How?
-
selsta
You only store 1/7th of some less important data.
-
selsta
Your privacy stays the same.
-
selsta
And other nodes can sync from your node.
-
selsta
`./monerod --prune-blockchain`
-
Feuermagier
How do i fix the database-read-only error?
-
selsta
but you have use a HDD or SDD
-
selsta
you are using a NAS according to the logs
-
selsta
oh, you are using a raspberry
-
selsta
syncing will be slow :)
-
Feuermagier
The RaspberryPi *hosts* the NAS
-
Feuermagier
I'm using the same HDD for the chain
-
selsta
It looked like a mounted NAS.
-
selsta
Same HDD should work.
-
Feuermagier
It is a plugged in HDD
-
Feuermagier
The error occured after I tried to import the chain from file
-
Feuermagier
I had 70% of the chain downloaded already
-
Feuermagier
and then started the import of the .raw file
-
selsta
you can’t import 70%
-
Feuermagier
the 55gb
-
selsta
ok, move `/media/NAS/Monero/lmdb` to a different folder and start the daemon again
-
Feuermagier
Won't that reset my progress?
-
selsta
Just for testing.
-
Feuermagier
ok, rebooting the Pi first, sec
-
selsta
rename lmdb to lmdb2 and try
-
Feuermagier
ok. will do
-
Feuermagier
I restarted and am running the daemon again now
-
Feuermagier
reboot alone might have fixed it
-
Feuermagier
seems to be doing stuff for now at least
-
selsta
ok, I’m not sure if it’s going to work
-
Feuermagier
It seems to have ignored my attempt to import from file completely and is now continuing it's download: 2020-03-10 22:58:22.802 I Synced 1696210/2051663 (82%, 355453 left)
-
selsta
looks good
-
selsta
I think it will take a few more days.
-
Feuermagier
yep
-
Feuermagier
probably 1-2 weeks more
-
selsta
v0.15.0.5 will be out soon, you should update then even if the sync is still in progress
-
selsta
it includes new checkpoints
-
Feuermagier
what are checkpoints?
-
selsta
I can’t explain it in technical terms but the sync will be faster until the last checkpoint.
-
selsta
v0.15.0.1 is 3 months old, so the last checkpoint is also 3 months old
-
selsta
the last 3 months will be super slow to sync
-
Feuermagier
do you have some additional reading on checkpoints and pruning?
-
selsta
-
Feuermagier
thx. seems interesting. but as I can afford the additional space and it helps the network more i think i'll keep the full chain
-
selsta
the difference for the network is minimal
-
selsta
I have nothing on checkpoints but usually it is nothing to worry about as a user
-
Feuermagier
is it like a checksum for all blocks in a specific period?
-
moneromooo
The binary includes a database of known hashes for blocks 0 to N. If a block you get over the network matches the expected one, it skips some validation tests.
-
moneromooo
Well, now it's more like the hash of a set of hashes of blocks, since it saves space, but same idea.
-
Feuermagier
makes sense
-
moneromooo
You can disavle that with --fast-sync 0 IIRC.
-
selsta
checkpoints are only for rollbacks?
-
selsta
what is the correct name for the database?
-
Feuermagier
syncing past 75% definitely seems to go a *lot* slower than 0-50%
-
moneromooo
I don't understand selsta.
-
selsta
There are checkpoints in checkpoints.cpp and there is the checkpoints.dat with expected_block_hashes_hash[].
-
selsta
Are they related?
-
moneromooo
In some way I guess, since they're both about checking block hashes.
-
moneromooo
It's a bit unfortunate checkpoints.dat is called that but hey.
-
selsta
That is what always confused me :D
-
Feuermagier
Is there a technical explanation somewhere I can read up on these things?
-
selsta
not aware of any