<monero.arbo:matrix.org> is docker still using bionic for builds? is that working for everyone?

<ofrnxmr:xmr.mx> Worked for me, and yes, bionic

the docker host doesn't even matter when you use dockrun, right?

<ofrnxmr:xmr.mx> Dockrun uses am 18.04 image

<ofrnxmr:xmr.mx> (i use slightly modified dockrun, since repo vers doesnt work on deb 12)

.merge+ 10506 10507 10508 10509 9899

Added

monero-i686-linux-gnu-v0.18.5.0.tar.bz2 doesn't repro

non-determinism in unbound package

sneedle and ofrn match, scooby doesn't

what about ci gitian output?

checking

i don't remember seeing unbound in updated packages

it wasn't updated

ci matches sneedle and ofrn

i'll ask scooby to clear cache and rebuild

<robinho13:matrix.org> Hi, a quick question regarding refactoring.

<robinho13:matrix.org> The following line in wallet2.cpp uses designated initializers:

<robinho13:matrix.org> src.multisig_kLRki = {.k = {}, .L = {}, .R = {}, .ki = rct::ki2rct(td.m_key_image)};[... more lines follow, see mrelay.p2pool.observer/e/pOK3wYALYldveVRJ ]

If it currently compiles on all platforms then I would keep it as is, refactoring PRs are usually not accepted.

we could work towards switching to c++20 on master

need +1 on 10509

<jpk68:matrix.org> I approve of the changes in 10509

<jpk68:matrix.org> Not sure if it counts though :(

jpk68: it doesn't, i'm bound to the CoC. did you try making a support ticket to get your gh account unblocked? worked for me years ago when i signed up with tor.

where's this cache? i'm happy to rebuild. is it inside gitrun-bionic or base-bionic-amd64? am I docker exec'ing into a shell in a container and rm'ing a directory?

<jpk68:matrix.org> tobtoht: Yes, I have contacted their support several times over the course of two months. I receive the exact same templated response every time

jpk68: ugh..

<jpk68:matrix.org> I signed up with a legit resedential IP, verified with a real phone number, etc. The ban reason was for "abuse of GitHub actions" which I have never even used

<jpk68:matrix.org> I also tried to make a second account to get an actual support response, and it didn't do anything... plus contacting a Microsoft employee (who responded) but they couldn't do anything

<jpk68:matrix.org> I will stop ranting now :)

scoobybejesus: one sec, it's been a while since i touched gitian

jpk68: ok, well perhaps it's time for us to start thinking about if it makes sense to allow contributors to approve of prs via logged channels / alternative ways

<jpk68:matrix.org> tobtoht: Thanks, that would be hugely appreciated. Clearly I am annoyed about this primarily because it's personally inconvenient, but I think that not having other options besides GH is going to be... not a good idea in the future. But that's just me

<321bob321> There is an easier fix

<321bob321> Selfhostimg doesn't have this issue

<ofrnxmr:xmr.mx> @jpk68:matrix.org: Do like the bots - just create a new acct

<ofrnxmr:xmr.mx> scoobybejesus are you using dockrun?

yep

<jpk68:matrix.org> @ofrnxmr:xmr.mx: The stupid CAPTCHA doesn't work, no matter what I try. This has apparently been a thing since 2023

<jpk68:matrix.org> github.com/orgs/community/discussions/177816

<jpk68:matrix.org> github.com/orgs/community/discussions/57317

<jpk68:matrix.org> They also block Tutanota emails

<jpk68:matrix.org> *for new signups

Someone could revive tippero and add a module to allow a ack/nak command to github :D

(that was an old python IRC bot)

<jpk68:matrix.org> I pray to the gods of Core that a self-hosted Forgejo instance will be considered someday ;)

<jpk68:matrix.org> Hey, we might have more GitLab disk space soon anyways...

scoobybejesus: there should be a container called gitrun? can you delete it and rerun dockrun

indeed. maybe that's easiest. on it

<ravfx:xmr.mx> > > > monero: miniupnpc.patch

<ravfx:xmr.mx> > > > patching file src/p2p/net_node.inl

<ravfx:xmr.mx> > > > Hunk #1 FAILED at 2989.

<ravfx:xmr.mx> > > > Hunk #2 FAILED at 3057.

<ravfx:xmr.mx> [... more lines follow, see mrelay.p2pool.observer/e/7b2hw4ALZ18wR2xZ ]

ravfx: yeah miniupnp is gone

<ravfx:xmr.mx> Perfect!

jpk68: if github continues its steady decline in reliability we might not have a choice lol

someone would need to set up the infra for self-hosted runners and be willing and able to maintain it reliably tho

<jpk68:matrix.org> True

<ofrnxmr:xmr.mx> So far weve had a few self-hosted solutions come and die or decline faster than github

<jpk68:matrix.org> It wasn't run by Core though

<ofrnxmr:xmr.mx> Example: repo.getmonero.org has one job and its offline due to spam and a full disk

<jpk68:matrix.org> Also, Forgejo actions syntax is very similar to GitHub's and can probably be converted quickly

<ofrnxmr:xmr.mx> Repo.getmonero.org is managed by core

<ofrnxmr:xmr.mx> Well, the server is.

<vtnerd> In fairness to GH, the spammers are relentless as shown by the repo.getmonero.org issues

<jpk68:matrix.org> GH clearly doesn't care about that, considering the amount of vibecoded garbage they push

<ofrnxmr:xmr.mx> There are millions of accounts of repo, and unless you manually approve each one you'll end up with people storing unsightly shit on it

<jpk68:matrix.org> See: GH trending page

<ofrnxmr:xmr.mx> Of account on repo/gitlab

<jpk68:matrix.org> They also apparently suspended sech1's CI usage for a few weeks last year

ofrnxmr: There's a reason why the big projects use a mailing list to coordinate patches

And just leave the self-hosted repository as invite-only

<ofrnxmr:xmr.mx> i agree

<jpk68:matrix.org> Is there something that can be set up, like a #monero-prs channel, for sending pastebin links

<jpk68:matrix.org> So it doesn't clutter up this one

pastebin banned me for uploading the uri of a base64 image (but the data redacted) so that term is triggering to me and we can't mention it anymore

debians*

<jpk68:matrix.org> In the meantime I guess I'll just put them here then (??)

What's wrong with sending patches over mail?

<jpk68:matrix.org> I don't have any problem with that. But AFAIK there isn't a mailing list

IRC is designed to be ephemeral so it would be harder to keep track of

Oh uhh Matrix too

They both aren't designed for this

<jpk68:matrix.org> I know, I was just thinking, as a temporary fix, it could work

<jpk68:matrix.org> I have like 6 different PRs that I want to submit right now, and it would be annoying for me to randomly paste links here

<jpk68:matrix.org> Annoying to others, I mean

<ofrnxmr:xmr.mx> Yes > <@jpk68:matrix.org> In the meantime I guess I'll just put them here then (??)

<monero.arbo:matrix.org> so this error is from trying to build on a fresh ubuntu 26.04 install: Error response from daemon: client version 1.41 is too old. Minimum supported API version is 1.44, please upgrade your client to a newer version

<ofrnxmr:xmr.mx> Using what method

<monero.arbo:matrix.org> dockrun.sh

<321bob321> Docker daemon version

<monero.arbo:matrix.org> yeah seems like

how weird. different i686 hash from before and still doesn't match. secpaste.dev/show/qenQT4ut

<monero.arbo:matrix.org> @monero.arbo:matrix.org: same error on fresh 24.04 actually

you gotta sudo vim /etc/docker/daemon.json and then { "min-api-version": "1.41" } and then sudo systemctl restart docker

we'll see if your workaround will similarly result in different i686 hashes...

<monero.arbo:matrix.org> okay yeah that did the trick, we will see

<jpk68:matrix.org> common: make timespan more readable paste.debian.net/hidden/a2e373c4

jpk68: we have a no code rewrite policy

so unless it fixes a bug it won't be merged

debian wont even let me view the diff cus banned lol

<jpk68:matrix.org> Just to be clear, it's improving the readability of the output, not how the function is written

will take closer look