whoo pr bot is back, nice.

.merge+ #1198

Added

.merges

ErCiccione[m]: can you just let me know when 1198 gets merged?

I'll rebase from master and avoid any other MR conflicts

Sure. Luigi1111 could you merge those PR when you can?

fwiw 1198 shouldn't conflict with the PR with the new paper. If that's your concern

Ah ok

In that case I'll remove the WIP from it

and do a later MR after all are merged

done

Will the presence of the colon in the title be a problem?

Do I need to quote-escape it?

I didn't even consider that

Can someone test the build for MR 1197 for that colon question?

sarang: colons in the titles must be escaped, yes

just p[ut the whole sentence in quotes

Bah, ok

Don't merge 1197 until I fix that quickly

i don't see any colon btw. looks fine to me

sarang what colon are you refering to?

Triptych: ...

stupid me. i got confused by the color. Yeah the whole string must be between quotes

I'll update. Marked WIP to avoid accidental merge

lesson learned :/

Single vs double quotes?

^ ErCiccione[m]

Does the parser care? Is there a style preference?

double quotes pls

roger

ErCiccione[m]: gitlab-ci file needs a `.` in the beginning

`gitlab-ci.yml` -> `.gitlab-ci.yml`

can you PR? I don’t have the repo setup

Oh, i must have messed up. Thanks selsta, will PR later today

wait, can make a quick PR now

ok

selsta: repo.getmonero.org/monero-project/monero-site/merge_requests/1200

could you rubberstamp it? so maybe luigi can merge it straight away

done

thanks! luigi1111 ^

cool CI seems to be working now

yep, seems to be going fine

Liquid Exception: invalid byte sequence in US-ASCII in resources/user-guides/How-to-mine-Monero-XMR-without-a-mining-equipment.md

aaand failed. I think i saw that before, something about how the docker container is configured

looks fixable :P

yeah, nothing bad. Moneromooo will be happy about it :P

OHAI. Do I need to read the backlog ?

nah. Pipeline failed, but should be a problem of encoding

That's the thing you got me to add a couple weeks ago ?

Yeah, the runner will probably need some tweaking

i'm about to go afk for some hours. I will look into it later this evening

multiple ways to solve it

moneromooo: you have to do nothing it’s not setup related

MR 1197 has been fixed ErCiccione[m]

selsta: it's not? taking a look now

oh i see your PR now. Is it just about enforcing the encoding?

sarang, thanks i will test it in a bit

ErCiccione[m]: AFAIK yes but my PR is untested

Also I don’t understand yet, every user has to run their own runner?

It would be good if the runner would also run on merge requests.

i'm testing your patch on gitlab using their shared runners. Seems to be building for now

if for "users" you mean "self hosted platform", yes

and they should run on PR as well

It does not. I think `only: master` means only on push to master.

And when I pushed on my forked to master it asked me to setup a runner.

Yes, only: master will only trigger when it's auctually merged to master

i see the pipeline on your PR: repo.getmonero.org/monero-project/m…-site/merge_requests/1201/pipelines

I tested that configuration file opening a PR on gitlab, and it run on the PR as well

Yes, it created the pipleline but it asked me to setup a runner.

Odd, it shouldn't be. Lemme pull one of my confs.

i made the tests on gitlab.com using their shared runners, maybe that's the difference.

I think the structure's wrong, which is why it's doing that. Structure in my envs are only: refs: -<branch>

I think both it valid.

they have a master example without `refs:`

I added `merge_requests` to only but then it asked me to setup a runner so it appears that mooos runner is only for the monero-project/monero-site repo.

And I would have to setup my own. I’m sure this can be changed.

damn, i was wromg, it doesn't build PRs also on gitlab.com :/

selsta: yes, the runner can be shared to build PR, but at this point i'll just shut up and ask snipa for suggestions on how to set things up

ooh here it is: docs.gitlab.com/ce/ci/merge_request_pipelines/index.html

> Currently, those pipelines are created in a forked project, not in the parent project. This means you cannot completely trust the pipeline result, because, technically, external contributors can disguise their pipeline results by tweaking their GitLab Runner in the forked project.

and forked projects don’t have a runner setup, that’s the problem

moneromooo: Do you have options which projects have access to the Gitlab runner?

I can add it to more IIRC.

Which ones do you need it forn ?

Every person would need access I guess else the on merge request feature is useless.

Appears that Gitlab does not support this yet. gitlab.com/gitlab-org/gitlab-foss/issues/23902

"I convinced my superiors that we do not need Jenkins, that the current GitLab can do all the same and even more, but it turned out that you do not have one of the base features." i can empathize with this guy :/

yep :/

i can see why it can be tricky to implement this but this is probably the most important feature for us

Travis-CI also does not support Gitlab.

well, if we cannot have runners working in a safe way against MRs (having runners running freely on all forks sounds like a security nightmare, and gitlab acknowledges that). This is the time i stop advocating for moving to gitlab. They moved that issue areound for years and now appears like they want to make it available only for enterprise users

whicb sounds like a big middle finger to open source projects

i mean. there are workarounds, but using APIs and external repos to run pipelines against MRs sounds ridiculus to me

I mean, did anyone apply to get the commerical license yet?

As it sounds like that's still an option.

Commercial license also does not support this.

Ah, that's annoying.

yeah, they put it in the milestone for a release 1 year ago, then 6 months ago, then 3 months ago -_-

If it's also 1.5 months ago, I think we can expect it in 1.5 months.

Wait. No. Now.

so... never

lol

sarang: the html of 1197 needs a couple of fixes. I left 2 comments

That was sloppy of me. Fixed

Thanks. Now it's ok.

.merge+ #1197 #1201

Added

.merges

pigeons please update the server when you have a chance

and maybe binaryfate? I lost track of who have access to the server

I can do now. Deploy current master?

ErCiccione[m] ^

Great. Yep :)

I see the updates. Thanks binaryfate

great! First time I did it so that's good :)

pigeons to be clear, you don't need to do anything

<ErCiccione[m]> and maybe binaryfate? I lost track of who have access to the server <--- fluffypony pigeons and me.

Ok, good to know. Thanks.

and yeah everything seems fine 🙂

btw i think the caching should be adjusted binaryfate. When there are heavy changes some parts take a lot of time to load or don't load at all without cleaning the cache first (which was the case with the new download page)

pigeons maybe could take a look? Don't know if it's nginx or cloudflare.

I still don't have access to cloudlfare (fluffypony has to follow up). So maybe it is because I couldn't flush stuff there.

I'd assume they detect changes on css/html and the slow time is only on first hit? Not really sure

Also, when you said "I see the updates", it was actually still building. So maybe it was just that

mmh, maybe, but the spreedsheet with the logos is not updated yet. See how gitlab's one is missing: web.getmonero.org/community/hangouts

*spritesheet

and the old one is still showing: web.getmonero.org/img/monero-spritesheet.png

could be just cloudflare slowly catching up, but i'm just guessing

Caching issue yes. I did flush the CDN, so that must be cloudflare.

I can't help this until fluffypony gives me and luigi1111 access to CF as discussed, sorry

Got it. Thanks for looking that up.