-
newyearday
has monero some timestamp check that decide independently about block version from other parameters
-
moneromooo
There's a v1_till variable or so, somewhere right near the fork table.
-
newyearday
i changed this to 0
-
moneromooo
Can't think of anything else.
-
newyearday
is crazy i have change even original version on hardfork.h
-
newyearday
everything!
-
newyearday
i did original_version = 12;
-
newyearday
when i debug i c that version is 1
-
moneromooo
IIRC your problem is the serialization bails midway.
-
moneromooo
The traces should show you the line where it stops.
-
newyearday
why when i insert version 1 daemon start
-
newyearday
i insert this mginfo trace but nothing
-
newyearday
MGINFO("trace");
-
newyearday
every second line
-
moneromooo
I'm fairly sure the log you posted had a few traces in it.
-
newyearday
that common_trace ?
-
newyearday
-
newyearday
look line 38,39,40
-
newyearday
on gdb the first line give the error is ss << tx_blob;
-
moneromooo
I think I told you to put this in the serialization code in cryptonote_basic.h. That's what apparently bailing.
-
moneromooo
Both in transaction and transaction_prefox.
-
newyearday
i insert it even there and again the output was the same "trace,trace,trace"
-
newyearday
on daemon.log
-
moneromooo
OK. Then I guess it's bailing earlier even. Is your blob empty ?
-
newyearday
no if i insert version 1 works
-
newyearday
when i insert 12
-
newyearday
is error
-
moneromooo
You're really really sure you put those logs in the tx serialization code, right ?
-
newyearday
yes!!
-
moneromooo
Dunno then. Very odd.
-
newyearday
i inserted after BEGIN_SERIALIZE() on basic.h
-
newyearday
and the same "trace,trace,trace"
-
moneromooo
Guess you can break in gdb in the parse_tx_from_blob function then step.
-
newyearday
the starnge is that on gdb i c version 1 after inserting 12 on hardfork.h,genesis start
-
newyearday
after step what
-
moneromooo
After step you step again till you find what fails.
-
moneromooo
You might have to step a lot. n for next (run calls in one go), s to dive down into the call.
-
newyearday
so step not next
-
newyearday
ok i do it right now
-
moneromooo
Both. n if you don't care about the call (ie, smart pointer stuff, logs, etc). s for stuff like serialialization code.
-
newyearday
is better here the break ss << tx_blob; or CHECK_AND_ASSERT_MES(r, false, "Failed to parse transaction from blob");
-
moneromooo
You want to first break before the call that sets r.
-
newyearday
binary_archive.h:101 first line after step
-
newyearday
now again step
-
newyearday
or next
-
moneromooo
Dude. I'm not telestepping for you, come on.
-
newyearday
same file line 67
-
newyearday
102,103,104 and after on postypes.h i have "value has been optimized out"
-
moneromooo
You can build with -O0 to fix most of these.
-
newyearday
u mean to insert more bytes
-
newyearday
on genesis
-
newyearday
after 105 on binary_archive.h im back on format_utils
-
newyearday
on the line that error start
-
newyearday
what do u mean with "build with -O0"
-
newyearday
moneromooo please can u explain what is need
-
moneromooo
-O0 is a gcc option to not optimize.
-
newyearday
so u mean to change compiling type and i'll be ok?
-
newyearday
or i'll have more info on debugging
-
moneromooo
It'll fix most of the "value has been optimized out".
-
moneromooo
Though you should be able to do most of the stepping without -O0.
-
newyearday
ok
-
newyearday
but where to insert in which cmake file
-
moneromooo
Replace current -O option, revert when done debugging.
-
newyearday
so i remove on cmakelists the "-O0" with "-O"
-
newyearday
ok
-
AlyRocks
hey
-
mfoolb
niocbrrrrrr: updates?
-
pureos
hi
-
kinghat[m]
-
yanmaani
Wonder how many millionaires there are who just so happened to mine Monero during its first two weeks?
-
yanmaani
In fairness, you can be compelled to divulge your audit key to the IRS.
-
kinghat[m]
what if they were lost in a 🚤
-
yanmaani
boating accident!
-
kinghat[m]
-
bigslim[m]
Smooth been around these parts in the last week?