-
iDunk
-
selsta
pls no
-
iDunk
I get 3f2ea03affdc75ccccbee148d03af9ad92da4af012c91a84a447863aa48f5b36 here.
-
selsta
ok got the same message
-
selsta
db55aa2ac8c7b8d8f9e471c09b0d3b7214360e24e94558f120124a1b3d84b336 osslsigncode-1.7.1.tar.gz
-
selsta
-
iDunk
It's a different SHA256 hash on every run here.
-
elFado
why is tor gateway +q here?
-
iDunk
It's not a tar.gz archive, it's a web page.
-
selsta
I'm currently testing with the 2.0 version
-
selsta
elFado: spam, though should get removed again
-
elFado
somethingUniqueR: kastelo was a scam afaik
-
selsta
why do we even use osslsigncode, we don't sign our .exe
-
selsta
guess we can just delete that from gitian-build.py
-
selsta
iDunk: I'm building now with line 97-100 removed
-
selsta
would it be fine if everyone does that for this release or should we tag v0.17.2.1?
-
selsta
removing line 97-100 does not change the resulting hashes so should be fine without retagging
-
bugs_
built 0.17.2.0 on fedora with cmake -DMANUAL_SUBMODULES=1; make
-
bugs_
regular make fails with Submodule 'external/miniupnp' is not up-to-date.
-
bugs_
FYI
-
iDunk
git submodule sync
-
iDunk
git submodule update
-
bugs_
ok those two seem to do the track after git checkout release-v0.17
-
bugs_
trick
-
bugs_
github lists boost dependency as 1.58 - has that changed? now i see:
-
bugs_
monero/tests/unit_tests/node_server.cpp:355:21: error: max_listen_connections is not a member of boost::asio::socket_base
-
bugs_
i have boost 1.60
-
bugs_
(i didn't see that error when i did cmake -DMANUAL_SUBMODULES=1
-
selsta
that is in unit tests
-
selsta
which afaik do not get build if you execute cmake manually
-
bugs_
ah ok
-
selsta
guess that is something we should fix
-
selsta
so that minimum dependency is 1.58 again
-
bugs_
i also tried to build with a private openssl 1.1.1k library and got a lot
-
bugs_
of boost related ssl issues
-
bugs_
tomorrow i'll try with a newer boost
-
bugs_
thanks
-
iDunk
-
selsta
-
selsta
two don't match so far :/
-
selsta
did you do apt-get update and apt-get upgrade? we had issues with a gcc update once
-
iDunk
That host is always up to date.
-
selsta
android hashes match again
-
msaqvhsxtikq
FUK has bееn intеnsеly hostilе to Monеro for ovеr 4 yеars. Making minеrs for Monеro, making a pool for Monеro, making codе that Monеro copiеd, not using bugs to stеal millions from еxchangеs. Wow. Such grеat hostility.
-
mj-xmr
selsta, may you point me to some info how to make such a reproducible build? I already was able to create a couple of "depends"
-
mj-xmr
or iDunk
-
mj-xmr
./contrib/gitian/gitian-build.py ?
-
diso
buenas tardes a todos
-
mj-xmr
hola
-
iDunk
mj-xmr: there's a README.md with instructions in contrib/gitian
-
mj-xmr
Thanks
-
mj-xmr
just a heads up re: boost. I see on Actions, that "Perfect-daemon" is building a fix for the Boost 1.58
-
selsta
hmm it is already PRed I think
-
sethsimmons
Should I reproduce 0.17.2.0 or wait?
-
sethsimmons
Want to be sure before I kick it off.
-
selsta
sethsimmons: I guess a third result would be helpful
-
sethsimmons
I don't have my env anymore so will take some time
-
sethsimmons
But I can
-
selsta
2 of my hashes don't match
-
selsta
and I don't know how to proceed lol
-
mj-xmr
I will prepare it for the next time.
-
selsta
sethsimmons: you have to change the gitian-build.py like this:
monero-project/monero #7652/files
-
selsta
before building
-
sethsimmons
Ok will take a stab at that shortly
-
sethsimmons
Running now with those lines removed
-
selsta
mostly interested in linux hashes
-
sech1
Running now too
-
sech1
I had to reinstall everything from scratch, it's been a while since 0.17.1.9 :D
-
scoobybejesus
-
selsta
yours match iDunk
-
» selsta waits for more hashes
-
gingeropolous
so whats the deal? is the thing all comboobled or can i just follow the same instructions as before?
-
-
sethsimmons
Linux hashes ^
-
selsta
sethsimmons: yours match mine again
-
selsta
scoobybejesus: iDunk: did you have the old cache?
-
sethsimmons
Yeah don't match scoobys
-
sethsimmons
weird
-
sethsimmons
Mine was completely from scratch, had never built on this host before
-
selsta
both sethsimmons and me did it from scratch
-
scoobybejesus
i did not do from scratch
-
selsta
seems like "something" changed that causes inconsistencies if not done from scratch
-
iDunk
I had the old cache.
-
sethsimmons
-
mj-xmr
Silly question: do I need to follow this step: git clone git⊙gc:monero-project/gitian.sigs.git
-
mj-xmr
?
-
sethsimmons
Only to push sigs in a PR
-
sethsimmons
Not to generate the hashes and compare FWIW
-
mj-xmr
thx
-
sethsimmons
but pushing sigs via PR is important for those who can (I need to)
-
Inge-
We are not at the point where I can deploy a docker container, git clone, build and get a hash?
-
sethsimmons
Thats what this does
-
sethsimmons
It's a script to automate that for you
-
Inge-
ooooh
-
sethsimmons
<sethsimmons "I did this but edited the gitian"> The steps here create the user needed, install dependencies, run the build script, and outputs all hashes
-
mj-xmr
I'm building on one slow machine, but I will plug a faster one soon.
-
sethsimmons
Just edit user and version to match
-
Inge-
hm so still have to run gitian-build.py outside docker. I was kind of hoping it would be even more containerized so everything happened inside docker :)
-
skkjwzowfjbp
For a community that prides itself on manipulating people, you can't manage a single guy that spends most of his time in his underpants :D
-
mj-xmr
OK, building on my best horse as well.
-
sech1
I'm still building somehow
-
sech1
Previous builds took less time
-
sethsimmons
<Inge- "hm so still have to run gitian-b"> All of the building does, but still have to setup the build env for Docker
-
sethsimmons
But yes, would be nice to have a pre-built container image that can be run to build the latest tag
-
-
-
sech1
-
sech1
I don't remember how to build MacOS binaries
-
selsta
-
selsta
sech1: ^
-
sech1
My linux hashes didn't match with scoobybejesus - did you edit gitian-build.py as was recommended?
-
scoobybejesus
i commented out ilnes 97-100, yeah
-
sech1
me too
-
selsta
it seems to be an issue with the cache, maybe some compiler got updated or so
-
sech1
But I screwed up and restarted the build several times before it ran smoothly
-
selsta
-
sech1
selsta your hashes match with mine. I'll start the build again to build MacOS binary
-
sethsimmons
-
sethsimmons
Docker image pushed if anyone wants to test/get ahead of the game running it.
-
sethsimmons
Still building reproducible hashes.
-
selsta
I guess downloading some docker image to do the process for you also isn't ideal if you can't verify what the docker image does :D
-
selsta
(not familiar with docker so I can be wrong here)
-
Inge-
heh
-
Inge-
well ideally there is a github page with a Dockerfile that is used in the creation of said image
-
Inge-
so you can inspect the Dockerfile, and use it to build your own image
-
sech1
Running v0.17.2.0 on my server already
-
Inge-
are there release notes or something to see the main changes? or only commits at this point?
-
selsta
-
sethsimmons
```
-
-
selsta
Inge-: mostly stability fixes
-
Inge-
and some longer-term hardening against node attacks I presume
-
selsta
not yet in this release
-
Inge-
k
-
sech1
I see quite a few P2P fixes from anon CCS
-
sech1
in v0.17.2.0
-
selsta
mostly seg fault / deadlock / race condition fixes
-
sech1
Oh, and my little bugfix PR (7615) is there too
-
selsta
GUI has more interesting changes this time
-
selsta
.merge+ 7652 7653 7654 7655
-
xmr-pr
Added
-
selsta
have been running the anon CCS fixes for a while on multiple nodes and everything seems quite stable :)
-
selsta
though v0.17.1.9 also had 70 days update for me
-
Inge-
selsta: with or without banlist?
-
sech1
do you mean 70 days uptime?
-
selsta
yes
-
selsta
typo
-
selsta
or thought about the right word and wrote the wrong one lol
-
sech1
yes, I've been running 0.17.1.9 non-stop pretty much since it was released
-
selsta
Inge-: most with, 1 without
-
sethsimmons
-
selsta
-
sethsimmons
"- Only print blocked nodes once with --enable-dns-blocklist (#7295)"
-
sethsimmons
Thank God, my poor logs XD
-
gingeropolous
im gettin this: /home/xmrchain/monero/tests/unit_tests/node_server.cpp: In lambda function:
-
gingeropolous
acceptor.listen(boost::asio::socket_base::max_listen_connections, ec);
-
gingeropolous
when trying to compile 17.2
-
selsta
-
selsta
or update boost
-
selsta
but that's just test, you can ignore it
-
selsta
daemon should built fine
-
selsta
build*
-
gingeropolous
k. danke
-
mj-xmr
Hashes late into the show:
paste.debian.net/1192805
-
sech1
selsta "7 people who worked ..." and the list has 9 people
-
selsta
ty
-
gingeropolous
xmrchain is on 17.2
-
sech1
hashes from me, this time with MacOS binaries:
paste.debian.net/hidden/6d92502e
-
selsta
sech1: please post to gitian.sigs repo
-
selsta
if you have gpg installed :D
-
sech1
I don't...
-
CSDUMMI[m]
Does anyone here know of a secure implementation of Ring Signatures. Specifically traceable Ring Signatures?
-
moneromooo
Monero's implementation is meant to be secure. At least for Monero's usage.
-
kinghat[m]
i know you guys are busy atm but maybe someone has help:
monero-project/monero-gui #2806#issuecomment-816019049
-
kinghat[m]
> Can an @monero-project maintainer comment on why there is a requirement for a git clone in order to run make release as noted in the README.md file??
-
moneromooo
That's how it determines the version string. I don't think there's another reason.
-
mj-xmr
kinghat[m], I'd propose to clone from a local repository.
-
mj-xmr
So clone from outside their build system, pass the cloned local copy to the BS, and from within clone the passed one.
-
mj-xmr
You're not the issue starter, are you?
-
kinghat[m]
no sorry, am just following along as I think monero should be packaged in flatpak for Linux.
-
moneromooo
I think best would be to add a NO_GIT_VERSION cmake option.
-
moneromooo
Set that and it uses... "unknown" as a version string.
-
mj-xmr
I replied him. No problem.
-
Inge-
thanks selsta
-
kinghat[m]
thanks for the replies all