2023-09-01 07:35:24 <ikke> ptrc: it should be 2
2023-09-01 07:36:02 <ikke> But the only thing I could not set via an argument / env variable was the max concurrency, so I have to fix that by hand if it's incorrect
2023-09-01 07:53:33 <clandmeter> ikke: added you to th e infra room
2023-09-01 07:53:39 <ikke> clandmeter: ok
2023-09-01 07:53:45 <clandmeter> thougt you as admin could chantge things
2023-09-01 07:54:32 <ikke> I don't see rooms not explicitly shared
2023-09-01 07:55:59 <ikke> Or maybe I can and didn't look at the right location
2023-09-01 07:56:01 <clandmeter> you dont see them here: https://meet.alpinelinux.org/b/admins/rooms
2023-09-01 07:56:59 <ikke> Yeah, I do see the rooms there
2023-09-01 07:57:54 <clandmeter> :)
2023-09-01 08:23:56 <ikke> ptrc: which arch had less concurrent runners?
2023-09-01 11:52:30 <ikke> clandmeter: fyi, I've enabled actively disallowing gitlab posts (issues / snippets) in the spamcheck service if the threshold is high enough. I haven't seen any genuine post come come close to reaching the threshold, but if things are being blocked, you know why it happens
2023-09-01 12:00:54 <raspbeguy> ikke, could I have a test VM please?
2023-09-01 12:19:18 <ikke> raspbeguy: root@gbr-app-4.alpinelinux.org
2023-09-01 12:19:48 <raspbeguy> thanks
2023-09-01 12:20:55 <raspbeguy> ikke, but it doesn't have acces to the VPN though
2023-09-01 12:21:09 <raspbeguy> I mean, it cannot sole zabbix.alpin.pw
2023-09-01 12:21:14 <raspbeguy> solve*
2023-09-01 12:21:19 <ikke> zabbix.alpinelinux.org works
2023-09-01 12:21:34 <raspbeguy> even in the agent config?
2023-09-01 12:22:45 <ikke> trapper.zabbix.alpinelinux.org would, but we'd only use that in case a host cannot connect to the VPN, and we'd use PSK encryption
2023-09-01 12:23:37 <raspbeguy> so apart from changing the ServerActive parameter, should I change something else?
2023-09-01 12:25:27 <ikke> https://www.zabbix.com/documentation/current/en/manual/appendix/config/zabbix_agent2 TLSConnect, TLSPSKFile
2023-09-01 12:26:28 <ptrc> ikke: hm, is it possible that they were constantly running jobs outside of alpine/aports for a while
2023-09-01 12:26:45 <ptrc> because it was all of them
2023-09-01 12:29:42 <ikke> I don't see any long running jobs around that theim
2023-09-01 12:29:44 <ikke> time
2023-09-01 12:56:23 <ikke> ncopa: I want to upgrade gitlab, and do an OS upgrade as well. I don't expect the outage to be long, but any objection?
2023-09-01 13:02:04 <ncopa> no problem
2023-09-01 13:02:18 <ncopa> do you need my help?
2023-09-01 13:02:53 <ikke> Not for now, thanks :)
2023-09-01 13:03:12 <ikke> I'll make an anouncement on gitlab
2023-09-01 13:07:42 <ikke> ncopa: do you see a notification? It's not appearing for me
2023-09-01 13:08:47 <ikke> ah, now I see it
2023-09-01 13:08:52 <raspbeguy> ikke, so the best practive when adding a machine to Zabbix is to connect it to the VPN right?
2023-09-01 13:09:07 <ikke> yes
2023-09-01 13:09:13 <ikke> I can add that machine to VPN later
2023-09-01 13:09:45 <raspbeguy> is that possible for me to add the machine to the VPN ?
2023-09-01 13:10:01 <raspbeguy> I guess we should have a ansible role for that
2023-09-01 13:10:13 <ikke> the most work is in generating the cert
2023-09-01 13:10:41 <raspbeguy> yeah, but on security side, am I allowed to do that?
2023-09-01 13:10:50 <ikke> the rest is installing the dmvpn package and running setup-dmvpn <cert>
2023-09-01 13:11:34 <ikke> No
2023-09-01 13:12:25 <raspbeguy> ok then the cert should be provided from your side, right?
2023-09-01 13:12:42 <ikke> correct
2023-09-01 13:13:29 <raspbeguy> ok, then it would be nice that you do that when you have time please
2023-09-01 13:13:54 <ikke> I'll arrange it
2023-09-01 13:14:13 <raspbeguy> I'll write a new role that expect a certificate to be given by the admin
2023-09-01 13:27:04 <raspbeguy> ikke, also, could you please give me the link to dmvpn doc again? I can't seem to find it anymore
2023-09-01 13:27:34 <ikke> https://gitlab.alpinelinux.org/alpine/dmvpn-tools
2023-09-01 13:27:36 <ikke> this?
2023-09-01 13:28:03 <raspbeguy> yes thanks
2023-09-01 13:28:14 <raspbeguy> didn't remember that was on gitlab
2023-09-01 13:28:36 <ikke> Going to reboot the gitlab server soon
2023-09-01 13:54:56 <ikke> raspbeguy: I've added a pfx file in /root/
2023-09-01 13:55:25 <ikke> The name of the file is significant 
2023-09-01 13:55:35 <raspbeguy> ok
2023-09-01 14:03:11 <raspbeguy> so each time you wan to connect to the VPN you have to run setup-dmvpn, right?
2023-09-01 14:03:29 <raspbeguy> is there a clean way to make it persistent?
2023-09-01 14:04:26 <raspbeguy> I mean I wan write a rc service file for sure, but isn't there anything you use already on existing machines?
2023-09-01 14:05:44 <ikke> it is persistent
2023-09-01 14:06:25 <ikke> You don't need to rerun it after reboot
2023-09-01 14:09:54 <raspbeguy> where did it put its config then?
2023-09-01 14:10:13 <ikke> different places
2023-09-01 14:10:16 <ikke> /etc/network/interfaces
2023-09-01 14:10:19 <ikke> /etc/dmvpn.conf
2023-09-01 14:10:37 <ikke> quagga, strongswan
2023-09-01 14:10:48 <raspbeguy> nothing in /etc/dmvpn.conf
2023-09-01 14:10:56 <raspbeguy> file don't even exist
2023-09-01 14:14:31 <ikke> /etc/quagga/*
2023-09-01 14:14:52 <ikke> /etc/strongswan.d/*
2023-09-01 14:15:14 <ikke> /etc/swanctl/*
2023-09-01 14:49:09 <ikke> :)
2023-09-01 14:49:14 <ikke> testing a new trigger, it works
2023-09-01 16:57:15 <ikke> nice, the spamcheck service is blocking spam
2023-09-01 17:00:50 <ikke> https://gitlab.alpinelinux.org/admin/spam_logs
2023-09-02 07:34:37 <ikke> clandmeter: do you know if minio on gitlab is still used for something?
2023-09-02 07:35:07 <clandmeter> Not by me
2023-09-04 17:18:31 <mps> should we upgrade postfix for stable releases?
2023-09-04 17:19:38 <mps> I will post MR for 3.18
2023-09-04 17:21:20 <mps> gitlab is slow for me
2023-09-04 19:27:56 <ikke> mps: I didn't see any recent load spikes on the gitlab server
2023-09-04 19:29:22 <mps> ikke: it was temporary, after few minutes it become normal
2023-09-05 01:35:55 <pj> I had issues with gitlab yesterday but I disregarded that as my network, though everything else worked fine
2023-09-05 12:12:00 <dpward> rsync.alpinelinux.org refuses connections from mirror.math.princeton.edu for 7 days, according to the admin I contacted. He said it affects their entire network and asked if someone else triggered a block? (I believe I sent this to the wrong channel yesterday)
2023-09-05 12:25:23 <ikke> dpward: We don't block anything ourselves 
2023-09-05 12:25:38 <ikke> So may be some network issue 
2023-09-05 12:25:56 <ikke> raync.a.o points to an anycast address 
2023-09-05 12:26:17 <ikke> dpward: but I can check in the logs later see if I notice something 
2023-09-05 12:29:57 <dpward> ikke: Thanks. He said it's resolving to 147.75.40.42 (same as I am getting)
2023-09-05 12:37:00 <ikke> Yes, that's the anycast address 
2023-09-05 12:37:36 <ikke> Can you ask them to try usa.t1.alpinelinux.org?
2023-09-05 12:38:03 <dpward> ikke: Sure I'm sending them a note back now, thanks
2023-09-05 17:46:00 <dpward> @iike usa.t1.a.o is working - they have reconfigured the mirror to use that
2023-09-05 17:46:20 <dpward> ikke: ^
2023-09-07 14:27:15 <Kladky> Hello, it seems that the MTA for Alpine's Gitlab does not support TLS (or at least doesn't advertise it). Is there a reason for this?
2023-09-07 14:31:48 <minimal> Kladky: BTW are you trying to deliver over IPv6 or IPv4?
2023-09-07 14:36:06 <Kladky> Idk, I can see if I can find out.
2023-09-07 14:41:18 <Kladky> Yea I got no idea, I haven't got logging setup.
2023-09-07 14:42:26 <Kladky> Most likely ipv4 though.
2023-09-07 14:43:05 <Kladky> Since that is what other mailservers are telling me that my mailserver is sending them messages using a v4 address.
2023-09-07 14:45:13 <Kladky> Another interesting thing is that the mail servers that send me the emails do support tls (gbr-app-1.alpinelinux.org), but the ones that should be recieving emails (gitlab.alpinelinux.org) don't.
2023-09-07 14:45:42 <Kladky> minimal: does the email reply feature work for you or not?
2023-09-07 14:46:00 <minimal> I have never used it
2023-09-07 14:46:14 <Kladky> If you can, please give it a go.
2023-09-08 09:34:30 <clandmeter> Kladky: i just added the TLS config, but I am unable to apply it as I have some trouble with the docker networking. ive asked ikke to look into it and try to enable it.
2023-09-08 18:46:07 <Kladky> clandmeter: Thank you.
2023-09-09 04:47:57 <F88888965> hi
2023-09-09 16:33:38 <clandmeter> ikke: i think i enabled TLS for gitlab postfix
2023-09-09 16:33:50 <clandmeter> im not sure if it works for ipv6
2023-09-09 16:34:17 <clandmeter> i added a build option to use host networking for postfix
2023-09-09 16:35:01 <clandmeter> and i named the container to make it easier to restart it from our acme cron script
2023-09-09 16:35:38 <clandmeter> but the build network option is a hack, i am not sure whats going on.
2023-09-09 16:36:33 <clandmeter> i can ping ipv6 in the RUN cmd, so what blocks things is a mystery for me
2023-09-10 19:59:26 <ptrc> who should i ask here for access to dev.a.o? :p
2023-09-10 19:59:37 <ptrc> would love to merge !51383
2023-09-10 20:03:39 <ikke> ptrc: I can arrange that in a bit
2023-09-10 20:16:15 <ikke> ptrc: you should be able to login to ptrc@dev.alpinelinux.org
2023-09-10 20:16:41 <ptrc> yup, works, any tl;dr on how the server actually works?
2023-09-10 20:17:12 <ptrc> ah, i can see
2023-09-10 20:17:22 <ikke> Everything is in /var/www/localhost/htdocs/archive/
2023-09-10 20:17:45 <ikke> I made a symlink in your home
2023-09-10 20:18:32 <ptrc> apparently there's also a symlink in /archive
2023-09-10 20:18:42 <ikke> ah ok
2023-09-10 20:18:53 <ptrc> which was made there for putting files there over scp, i guess? :p
2023-09-10 20:19:02 <ptrc> but yeah, got it, thanks!
2023-09-10 20:19:05 <ikke> cool
2023-09-12 06:39:49 <ikke> 🤨
2023-09-12 13:59:10 <ncopa> is it possible to get the source tarball that was used for a specific CI build? https://gitlab.alpinelinux.org/Celeste/aports/-/pipelines/179596
2023-09-12 13:59:25 <ncopa> it passed there, but now the source checksum fails
2023-09-12 14:26:50 <ikke> We don't store those at the moment 
2023-09-12 14:53:33 <ncopa> like I thought. so there is no way to find out what changed
2023-09-12 15:21:49 <ncopa> im stopping mqtt-exec.aports-build on build-edge-x86 to debug https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/49802
2023-09-12 15:22:59 <ikke> 👍
2023-09-12 15:34:02 <ikke> oom killed :(
2023-09-12 15:35:32 <ncopa> oom killed?
2023-09-12 15:35:58 <ikke> arm ci vms
2023-09-12 16:32:54 <ncopa> i restarted build-edge-x86. will have to continue tomorrow
2023-09-12 17:47:20 <ikke> One thing we should already look at is if we have enough space on the builders for the upcoming release
2023-09-12 23:07:22 <ptrc> ikke: we don't have enough space on ci, that's for sure
2023-09-12 23:07:29 <ptrc> tar: firefox-117.0.1/third_party/rust/unicode-ident/tests/fst/xid_continue.fst: Cannot write: No space left on device
2023-09-12 23:07:29 <ptrc> tar: firefox-117.0.1/third_party/rust/unicode-ident/tests/fst/xid_start.fst: Wrote only 3072 of 10240 bytes
2023-09-12 23:07:40 <ptrc> that's from a recent job that tried to build ff
2023-09-13 08:30:11 <ncopa> yup. we are running out of diskspace on arm CI. https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/51486#note_337031
2023-09-13 08:41:17 <ikke> Will fix it in a bit
2023-09-13 11:47:34 <mps> I'm thinking that the linux-edge should be moved to testing. anyone against this
2023-09-13 12:37:33 <pj> testing was supposed to keep packages for like 6 months at most
2023-09-13 13:00:24 <mps> pj: no, some are there for years, intentionally
2023-09-13 14:06:35 <mps> clandmeter: ^
2023-09-13 14:06:45 <mps> what is your opinion
2023-09-13 14:37:13 <clandmeter> What is your reasoning mps?
2023-09-13 14:40:22 <mps> clandmeter: it changes fast and could break for end users, because this I think it is not for alpine stable releases
2023-09-13 14:40:48 <mps> it broke few times my armv7 machines
2023-09-13 14:41:52 <mps> if it is in testing then users knows they use it at their risk
2023-09-13 14:44:57 <clandmeter> Testing is like staging, it should not stay long or get deleted. I think this was the outcome in TSC issue months ago 
2023-09-13 14:46:03 <ptrc> wasn't the outcome "there should be a split into staging and experimental", or something like that?
2023-09-13 14:46:04 <clandmeter> you specifically named it edge for that reason right?
2023-09-13 14:46:48 <clandmeter>  I’m not sure exactly, currently on holidays so can’t check it
2023-09-13 14:47:28 <clandmeter>  I remember I created a script to cleanup testing 
2023-09-13 14:47:36 <clandmeter> but never implemented it
2023-09-13 14:49:17 <mps> I keep some pkgs for years in testing
2023-09-13 14:50:16 <ptrc> yeah, firefox-developer-edition also isn't meant to go to stable
2023-09-13 14:50:20 <mps> please, don't dare to remove them before asking me ;-)
2023-09-13 14:51:24 <mercenary> <user opinion>default install has linux-lts, it is user's choice to install linux-edge, knowing full well it may break. And I don't really want to run the entire machine at testing level, just because of linux-edge</opinion>
2023-09-13 14:51:25 <mps> clandmeter: I can keep linux-edge in community for you
2023-09-13 14:52:09 <mps> but I feel that can cause problems for end users
2023-09-13 14:53:02 <mps> it doesn't make much time to cherry-pick it from alpine edge to latest stable
2023-09-13 14:53:12 <mps> s/make/take/
2023-09-13 14:54:13 <mercenary> hmm. testing repo is not rsync-able is it?
2023-09-13 14:54:23 <mps> and we keep linux-asahi and other things for apple silicon for more than a year in testing
2023-09-13 14:55:49 <ptrc> mercenary: it very much is
2023-09-13 14:59:21 <mercenary> ptrc: yeah, I realised after typing that - reason it is excluded here is that there is so much churn that it sometimes takes more than the allowed 7 hours to sync
2023-09-13 14:59:39 <mercenary> (don't ask. 3rd-world internet)
2023-09-13 15:28:21 <clandmeter> mps: don’t keep it for just me. It’s just my opinion. You are the maintainer. You can do what you want.
2023-09-13 15:30:41 <mps> maintainer yes, but not owner
2023-09-13 15:32:16 <pj> clandmeter: mps: testing  very much isn't supposed to keep packages, it has been made clear long time ago, it was discussed and I specifically remember talking with ariadne about this on gitlab
2023-09-13 15:32:37 <pj> and it's not "intentionally"
2023-09-13 15:32:47 <pj> unless you intentionally forget to clean it up
2023-09-13 15:32:59 <ikke> the main reason testing was added was to give package maintainers an opportunity to test their packages before they were being included in a stable release
2023-09-13 15:33:33 <ikke> But people now use that as a way to include packages that they don't want in a stable release
2023-09-13 15:33:39 <pj> it just became a dumpster for abandoned packages people add and forget
2023-09-13 15:33:52 <pj> or don't want to maintain anymore
2023-09-13 15:34:05 <pj> (although that specific situation is rare)
2023-09-13 15:34:39 <mps> testing is as it is because alpine doesn't have something like ubuntu PPA
2023-09-13 15:35:05 <pj> erm, no
2023-09-13 15:35:33 <pj> the distance between community and testing is just a single MR
2023-09-13 15:35:52 <mps> well
2023-09-13 15:37:37 <pj> I think if we have hard rule of "testing max. 6 months otherwise delete" it will very quickly sort out which packages are still maintained/used (as maintainers will move them to community) and which are not
2023-09-13 15:38:30 <mps> some pkgs are not main/community, take asahi as examples
2023-09-13 15:38:53 <mps> not for*
2023-09-13 15:38:53 <ikke> why is that so?
2023-09-13 15:39:01 <ikke> what if testing didn't exist?
2023-09-13 15:39:16 <pj> Ok, then we take asahi as example. Why asahi is not moving to community in 6 months or less?
2023-09-13 15:39:17 <mps> then I will fork alpine 
2023-09-13 15:39:34 <mps> ikke: ^
2023-09-13 15:39:52 <mps> ;-p
2023-09-13 15:41:30 <mps> pj: asahi is experimental and intended to merge to current upstream
2023-09-13 15:42:08 <mps> and one day these pkgs will disappear, though I think this will not be soooon
2023-09-13 15:42:48 <pj> then maybe it should have been on it's own package distribution point
2023-09-13 15:44:29 <mps> I already have such https://arvanta.net/mps/aports-local/aarch64/
2023-09-13 15:45:11 <mps> and more not public
2023-09-13 15:45:25 <pj> then I don't see an issue in removing the package
2023-09-13 15:45:38 <mps> and we have riscv64 starfive repo
2023-09-13 15:45:49 <ikke> also a custom repo
2023-09-13 15:46:13 <ikke> exactly because we felt they don't belong in aports
2023-09-13 15:46:27 <mps> pj: because out of repo pkgs are not alpine official ones
2023-09-13 15:46:56 <mps> and people want alpine on asahi and want to get official pkgs
2023-09-13 15:47:13 <mps> s/asahi/apple silicon/
2023-09-13 15:47:18 <pj> the only official packages are in main/
2023-09-13 15:47:30 <mps> no, all are
2023-09-13 15:47:44 <ikke> pj: we don't make such a distinction
2023-09-13 15:47:59 <pj> ikke: i know
2023-09-13 15:48:29 <pj> but I don't see how "#maintainer: some dude" is official package
2023-09-13 15:48:48 <mps> pj: with this I agree
2023-09-13 15:48:50 <pj> mps: btw. you have invalid #Maintainer comment in asahi-scripts
2023-09-13 15:49:02 <pj> missing space after :
2023-09-13 15:49:09 <mps> pj: ah thanks, will look
2023-09-13 15:50:41 <mps> pj: I would prefer if maintainer have 'real name' and mail address
2023-09-13 15:51:07 <mps> but I'm not one who decide about this
2023-09-13 15:51:35 <pj> ikke: realistically, are you (you as in project) maintaining (actively) any of the packages that you are not maintainer of that reside in community/testing
2023-09-13 15:52:37 <pj> and this is not about alpine stable support dates but about actually doing anything on those packages and keeping them up-to-date
2023-09-13 15:52:55 <pj> and I know the answer, it's "no"
2023-09-13 15:53:19 <ikke> but the same counts for main
2023-09-13 15:54:19 <pj> if i recall correctly, main was supposed to be maintained purely by alpine team
2023-09-13 15:54:27 <pj> or is in majority
2023-09-13 15:54:52 <pj> (speaking after the fact that I took over "rust" in main and I'm not part of Alpine project)
2023-09-13 15:55:34 <mps> bbl
2023-09-13 15:57:00 <ikke> I don't think that's a useful distinction, and again too much work for just us to maintain
2023-09-13 15:57:13 <ikke> and in practice also not the case, like you mention
2023-09-13 15:59:12 <pj> then it's a bit weird that the contributing concern for including systemd package was maintenance 
2023-09-13 15:59:58 <pj> s/for/against
2023-09-13 16:03:08 <pj> mps: i don't really care about 'real name' or not, if people are not comfortable using it, then that's fine, I care about result, if there will be demand of using real names then: 1) count of active maintainers will drop 2) people will start using fake names
2023-09-13 16:06:12 <pj> anyway: https://gitlab.alpinelinux.org/alpine/tsc/-/issues/2#note_231114
2023-09-13 16:09:07 <pj> I already forgot it's been so long since that issue
2023-09-13 16:12:16 <ikke> pj: that was not about the package being in main or not, but about it being included at all
2023-09-13 16:16:13 <pj> yes, but there was concern about maintenance, why is that a concern if it's not a concern in main/ or anywhere else
2023-09-13 16:17:37 <pj> anyway, about the tsc issue, is there any progress or blocker for that?
2023-09-13 16:17:51 <ikke> blocker: time
2023-09-13 16:18:20 <pj> time for what
2023-09-13 16:18:48 <ikke> clandmeter was working on it
2023-09-13 16:18:48 <pj> if it's about implementing it, can't it be stated what has to be done so anyone can pick up the issue
2023-09-13 16:19:41 <ikke> 14:47 <@clandmeter>  I remember I created a script to cleanup testing
2023-09-13 16:19:44 <ikke> 14:47 <@clandmeter> but never implemented it
2023-09-13 16:21:38 <pj> yeah but it would be nice/good to know in that specific issue "I (clandmeter) created a script but I don't have time to implement it. Can someone take care of that"
2023-09-13 16:22:28 <pj> (and point to the script)
2023-09-13 16:23:29 <ikke> In an ideal world, that would've happened, in practice, life happens and you forget
2023-09-13 16:24:06 <pj> maybe there should be project manager
2023-09-13 16:24:22 <pj> \>:)
2023-09-13 20:12:36 <ptrc> hm
2023-09-13 20:12:41 <ptrc> is irc supposed to leak gitlab user emails
2023-09-13 20:12:44 <ptrc> err
2023-09-13 20:12:45 <ptrc> mqtt
2023-09-13 20:20:25 <ikke> I would not expect it, but might be possible if gitlab sends it unredacted (the logs shows it as ["REDACTED"]
2023-09-13 20:21:11 <ptrc> well, i've got mosquitto_sub running in the background and noticed stuff like
2023-09-13 20:21:45 <ptrc> gitlab/user_create/noah.friedman {"created_at":"2023-09-13T20:07:00Z","email":"<redacted by me, not gitlab>","event_name":"user_create","name":"Noah Friedman","updated_at":"2023-09-13T20:07:00Z","user_id":27726,"username":"noah.friedman"}
2023-09-13 20:22:14 <ptrc> and as far as i can see, the email is not public
2023-09-13 20:25:25 <ikke> Hm, not good
2023-09-13 20:34:12 <ikke> I've disabled most hooks, and tried to set the user_email field to nil for the remaining events
2023-09-13 20:34:18 <ikke> see if that helps
2023-09-13 20:45:25 <ptrc> will update on next user signup
2023-09-13 20:47:58 <ikke> Only push and tag_push events should be left
2023-09-13 20:48:09 <ikke> I don't think we ever used anything else
2023-09-13 21:19:30 <ptrc> aarch64 CI is down?
2023-09-13 21:19:32 <ptrc> https://gitlab.alpinelinux.org/alpine/aports/-/jobs?statuses=RUNNING
2023-09-13 21:19:38 <ptrc> nothing running
2023-09-13 21:19:43 <ptrc> but aarch64 jobs are pending
2023-09-13 21:20:42 <ptrc> ...ah, it doesn't show CI running for non-developers, because it runs in the forks' scopes instead
2023-09-14 17:20:22 <usrhere> ikke: hi, are the recordings from https://alpinelinux.org/conf/ available somewhere else ? looks like those links are dead 
2023-09-14 18:40:00 <ikke> Sadly, the storage it was on had died 
2023-09-14 19:15:18 <raspbeguy> Too bad this wasn't seeded as a torrent
2023-09-14 20:17:19 <ptrc> do you think someone would keep on seeding it?
2023-09-14 20:17:37 <ptrc> i'd say there would be exactly one seed and it would die along with the storage
2023-09-14 22:07:41 <raspbeguy> ptrc: I would
2023-09-14 22:09:03 <raspbeguy> And I'm sure there are plenty of other datahoarders in Alpine community or even in the Linux community
2023-09-15 04:59:32 <ikke> nginx was not properly started because it complained it could not find a backend hostname
2023-09-15 05:03:11 <ikke> reloaded*
2023-09-15 05:29:10 <ikke> "It's always dns"
2023-09-15 06:41:25 <numono> >storage died
2023-09-15 06:43:58 <numono> o no :< wish i had mirrored them. i have plenty space on s3, and if anything else is in danger, lmk i can give you credentials/storage
2023-09-16 06:50:28 <mps> algitbot: retry master
2023-09-16 13:41:23 <ikke> Seems like the spamcheck service is quite effective, the amount of spam we receive in the first place reduced quite and most that we still get is blocked
2023-09-16 13:44:07 <ikke> And importantly, no false positives as of yet
2023-09-18 10:53:09 <ikke> nginx is in an uninterruptable sleep
2023-09-18 10:53:21 <ikke> (disk sleep)
2023-09-18 10:54:58 <ikke> rebooting deu5
2023-09-18 11:15:54 <ikke> It's back, had to reboot it another time 
2023-09-18 19:03:40 <mps> looks like reply by mail doesn't work to merge request comments?
2023-09-18 19:12:31 <ikke> mps: in what sense?
2023-09-18 19:13:29 <mps> I replied by mail to comment in MR and got answer that it is unacceptable
2023-09-18 19:14:21 <mps> 'Recipient address rejected: User unknown in local recipient table (in reply to RCPT TO command)'
2023-09-18 19:15:44 <ikke> clandmeter recently changed something, maybe it's a result of that
2023-09-18 19:16:46 <clandmeter> I set the servername
2023-09-18 19:16:55 <clandmeter> maybe that is causing it?
2023-09-18 19:18:09 <mps> here is what I got https://tpaste.us/qERw
2023-09-18 19:20:03 <ikke> fyi, the e-mail address in there is private, anyone can pretent to be you using that address
2023-09-18 19:20:54 <mps> how it is private?
2023-09-18 19:21:03 <mps> I used it a lot of times
2023-09-18 19:21:22 <ikke> I mean, private to you
2023-09-18 19:21:42 <ikke> Not your personal the incoming+<hash> email address
2023-09-18 19:22:32 <ikke> Hmm, or maybe that one is not
2023-09-18 19:22:49 <mps> why then it is allowed to send from it
2023-09-18 19:23:20 <mps> i.e. set 'Reply-To: alpine / aports <incoming+d8b609d52eef5fde399724a39e1c6b82@gitlab.alpinelinux.org>'
2023-09-18 19:24:54 <clandmeter> ikke: can you check if i messed up the entrypoint script?
2023-09-18 19:25:39 <clandmeter>  There is a if then for the imap connection
2023-09-18 19:26:24 <ikke> if [ "$VIRTUAL_MAILBOX_DOMAINS" ]; then
2023-09-18 19:26:59 <clandmeter> notnsure where it lookup the receipient 
2023-09-18 19:27:11 <ikke> virtual_mailbox_domains = gitlab.alpinelinux.org
2023-09-18 19:27:13 <ikke> virtual_transport = lmtp:dovecot:2525
2023-09-18 19:27:19 <ikke> that's in main.cf
2023-09-18 19:27:26 <clandmeter> right 
2023-09-18 19:27:35 <clandmeter> that’s should allow it
2023-09-18 19:27:43 <clandmeter> if it’s enabled
2023-09-18 19:29:32 <ikke> "postfix/trivial-rewrite[12051]: warning: do not list domain gitlab.alpinelinux.org in BOTH mydestination and virtual_mailbox_domains"
2023-09-18 19:29:50 <ikke> I guess that's the issue
2023-09-18 19:31:59 <clandmeter> Change it to the real hostnane?
2023-09-18 19:33:20 <ikke> which one?
2023-09-18 19:33:35 <ikke> mydestination?
2023-09-18 19:33:41 <clandmeter> I’m confused
2023-09-18 19:34:22 <clandmeter>  what is mydestinarion set to?
2023-09-18 19:36:26 <ikke> mydestination = $myhostname, localhost.$mydomain, localhost
2023-09-18 19:36:30 <ikke> myhostname = gitlab.alpinelinux.org
2023-09-18 19:37:12 <clandmeter> Did I change that?
2023-09-18 19:37:30 <clandmeter> got diff should tell
2023-09-18 19:37:48 <ikke> No, but you did add the virtual_mailbox_domain
2023-09-18 19:38:13 <clandmeter>  no that’s there already
2023-09-18 19:38:20 <clandmeter> just the if 
2023-09-18 19:38:37 <clandmeter> so that’s weird
2023-09-18 19:39:02 <ikke> was the image update?
2023-09-18 19:39:08 <clandmeter> i did set the hostname
2023-09-18 19:39:08 <ikke> newer postfix version?
2023-09-18 19:39:42 <ikke> in .env POSTFIX_MYHOSTNAME
2023-09-18 19:40:08 <ikke> https://serverfault.com/questions/491601/postfix-do-not-list-domain-example-com-in-both-mydestination-and-virtual-alias
2023-09-18 19:40:36 <ikke> Should we set the hostname to the hostname of the server?
2023-09-18 19:43:29 <clandmeter> Yup
2023-09-18 19:48:19 <clandmeter> Or we need to explicitly set mydestination 
2023-09-18 19:49:03 <ikke> mps: can you try again?
2023-09-18 19:55:07 <mps> ikke: will do in few minutes
2023-09-18 19:55:26 <ikke> In a few minutes I need to restart gitlab :p
2023-09-18 19:57:41 <mps> ikke: it passed now https://gitlab.alpinelinux.org/alpine/aports/-/merge_requests/51822#note_338326
2023-09-18 19:57:51 <ikke> good
2023-09-18 20:03:09 <clandmeter> ikke: thx
2023-09-18 20:04:43 <ikke> applied security upgrade for gitlab (got a notification popup)
2023-09-18 20:43:31 <ptrc> ikke: https://gitlab.alpinelinux.org/alpine/cloud/alpine-cloud-images/-/issues/133#note_338385
2023-09-18 20:44:42 <ikke> poof
2023-09-18 20:57:01 <ptrc> thanks :3
2023-09-19 05:15:12 <ikke> https://about.gitlab.com/releases/2023/09/18/security-release-gitlab-16-3-4-released
2023-09-19 10:49:10 <mps> 9e13d85582f7d0ade9b7d5e7cc31974baa2c06bb commit is mistake imo
2023-09-19 10:49:26 <mps> DB is still in repo
2023-09-19 10:50:39 <ikke> an older version that stil has the old license
2023-09-19 10:53:38 <ikke> but yes, it's still available
2023-09-19 10:53:43 <mps> I know and remember why and what happened
2023-09-19 10:54:14 <ikke> It depends on whether postgrey can still work with bdb 5
2023-09-19 10:54:14 <mps> DB is not upgraded because of license but it is not removed from aports
2023-09-19 10:54:30 <mps> ikke: I'm just testing
2023-09-19 10:54:41 <mps> just now*
2023-09-19 10:55:53 <mps> I think we should revert removal of perl-db and postgrey (maybe something else also)
2023-09-19 10:57:24 <ikke> The question is how long we can keep bdb 5 in the repos
2023-09-19 11:00:11 <mps> idk, but revdep show openldap, opensmtpd, postfix-stats, bitcoin, apk, isync, netatalk
2023-09-19 12:57:08 <ncopa> i think we should remove berkley db and perl-db. but we could replace db with something else for postgrey if the app is needed and there are no good alternatives
2023-09-19 13:05:40 <mps> not only postgrey depends on db but also some pkgs still in aports
2023-09-19 14:57:06 <ncopa> they should ideally migrate to other db backend
2023-09-19 16:12:52 <ikke> more spam blocked :)
2023-09-20 10:26:14 <ikke> telmich: anything up with vigir23?
2023-09-20 11:16:47 <pj> is there a way to plug own gitlab-runner with docker and specific arch?
2023-09-20 11:17:23 <pj> or make the job for specific arch run with that arch :>
2023-09-20 11:34:58 <ikke> Usecase
2023-09-20 11:35:00 <ikke> ?
2023-09-20 11:35:27 <ikke> Is this related to aports, or just in general?
2023-09-20 11:35:46 <pj> aports
2023-09-20 11:36:06 <pj> I wanted to use own machine for building some arches, like for example aarch64 is off now
2023-09-20 11:37:35 <pj> afaik, it's not possible to give arch to gitlab-runner, and it would either have to be native arch or configured in .gitlab-ci.yml
2023-09-20 11:37:57 <ikke> Yes, correct 
2023-09-20 11:38:19 <pj> sad
2023-09-20 11:38:44 <ikke> The platform in docker terms is not exposed in gitlab-ci
2023-09-20 11:39:30 <ikke> So we have runners setup to build a specific arch and use tags to target them
2023-09-20 11:39:33 <pj> i thought i could cheat it by registering runner with different arch
2023-09-20 11:40:13 <ikke> The arch is not part of the registration 
2023-09-20 11:40:51 <pj> No, as in register runner on different architecture than host
2023-09-20 12:01:38 <chereskata> ERROR: Job failed: failed to pull image "alpinelinux/gitlab-runner-helper:latest" with specified policies [always]: Error response from daemon: Get "https://registry-1.docker.io/v2/": dial tcp: lookup registry-1.docker.io on 172.16.27.1:53: read udp 172.16.27.65:52525->172.16.27.1:53: i/o timeout (manager.go:237:20s)
2023-09-20 12:01:56 <chereskata> for aarch64
2023-09-20 12:02:30 <ikke> Which runner?
2023-09-20 12:02:56 <chereskata> https://gitlab.alpinelinux.org/chereskata/aports/-/jobs/1118758
2023-09-20 12:13:16 <ikke> chereskata: I've paused those runners, so it should not pick up jobs 
2023-09-20 12:16:10 <chereskata> okey, thanks
2023-09-20 12:38:31 <telmich> <ikke> "nico🇨🇭: anything up with vigir23..." <- Nothing I am aware of
2023-09-20 15:27:39 <mps> hm, arm boxes are unavailable again
2023-09-20 15:46:01 <ikke> telmich: since last night, vigir23 is (partially) unreachable for us. Partially, because at least one CI runner apparently still had contact with gitlab
2023-09-20 15:46:58 <ikke> Cannot ping 2a0a:e5c1:517::42 is not pingable
2023-09-21 07:41:55 <ncopa> is there some way I can connect to che-bld-1 or che-ci-1? I'd like to debug curl on on 32 bit arm
2023-09-21 07:45:49 <ikke> There are connection issues atm, but usa-bld-1 is reachable 
2023-09-21 08:18:15 <ncopa> ok
2023-09-21 08:58:59 <ptrc> https://gitlab.alpinelinux.org/alpine/aports/-/jobs/1119608 looks stuck :/
2023-09-21 09:00:52 <ikke> Despite being cancelled?
2023-09-21 09:04:19 <ptrc> i *just* cancelled it
2023-09-21 09:04:26 <ptrc> it was running for 100+ minutes
2023-09-21 09:04:50 <ptrc> most of that stuck on.. cleaning up srcdir?
2023-09-21 09:07:07 <ikke> Stunned 
2023-09-21 09:07:21 <ikke> stunnel is known to get stuck on the builders 
2023-09-21 09:11:59 <ptrc> hm, i suppose it forked itself and somehow refused to get killed?
2023-09-21 09:18:59 <nmeum> ikke: would it be possible to give me access to an aarch64 lxc? access over the wireguard VPN would be fine. I would like to debug https://github.com/golang/go/issues/62556 further
2023-09-21 09:20:31 <nmeum> (i would like to figure out if this is a regression in Go or in binutils and for this purpose I need to compile an old binutils/go version + git bisect potentially and that is kind of annoying in qemu)
2023-09-21 10:38:22 <ikke> nmeum: root@nmeum-edge-aarch64.usa-bld-1.alpin.pw
2023-09-21 11:53:18 <nmeum> ikke: how do I resolve that domain?
2023-09-21 11:53:56 <ikke> 172.16.6.3
2023-09-21 11:54:13 <nmeum> ah thanks
2023-09-21 12:00:29 <nmeum> ikke: which pubkey did you use for root@? https://gitlab.alpinelinux.org/nmeum.keys are the ones I currently use
2023-09-21 12:01:37 <ikke> I used those
2023-09-21 12:02:13 <ikke> typo in the authorized_keys file name
2023-09-21 12:03:05 <nmeum> yep, works now. thank you!
2023-09-21 12:07:01 <mps> is it possible to use rustup in makedepends on builders and/or CIs
2023-09-21 12:08:16 <ikke> I don't think so, why do you need rustup?
2023-09-21 12:09:04 <mps> to build linux-asahi GPU driver, need to use specific rust versions
2023-09-21 12:09:44 <mps> current rust 1.72 works but not sure will the next one work
2023-09-21 12:32:35 <pj> why wouldn't it?
2023-09-21 12:33:42 <mps> pj: you asking me?
2023-09-21 12:33:56 <mps> or ikke 
2023-09-21 12:35:32 <pj> you, since you're implying that next version won't build
2023-09-21 12:46:57 <mps> pj: because specific kernel versions require specific rust version
2023-09-21 12:48:57 <pj> sounds like something that can be easily patched to accept any version
2023-09-21 12:51:13 <mps> not worth effort and results would be risky
2023-09-21 12:51:52 <mps> solution could be rustup or have different versioned rust in repos
2023-09-21 12:52:05 <pj> rustup would not be solution
2023-09-21 12:52:40 <ikke> why would a newer rust version break those features?
2023-09-21 12:52:42 <pj> it's completely different compiler than then one we have in aports
2023-09-21 12:54:59 <mps> ikke: idk, but guess why and will not tell publicly
2023-09-21 12:56:07 <pj> I don't see any reason why it wouldn't build
2023-09-21 12:56:33 <mps> pj: try and tell if you know solution
2023-09-21 12:56:54 <pj> solution to what
2023-09-21 12:56:57 <mps> I'm fighting with this for nearly full year
2023-09-21 12:57:24 <mps> and not only I
2023-09-21 12:57:41 <mps> different people on different distros have same problem
2023-09-21 12:57:49 <pj> what problem
2023-09-21 12:58:10 <mps> rust kernel relation
2023-09-21 12:58:58 <mps> try to build kernel 6.6-rc2 with rust 1.71 for example
2023-09-21 13:00:10 <pj> or you could send an actual issue
2023-09-21 13:00:15 <pj> build log
2023-09-21 13:00:16 <pj> error
2023-09-21 13:02:48 <mps> don't have time. Just asked could rustup  work on builders. if it can't then this is not important
2023-09-21 13:03:03 <ikke> conclusion: no
2023-09-21 13:03:21 <mps> ikke: thanks
2023-09-21 13:04:11 <mps> I'll continue to keep on local site kernel with GPU for users who wants to use it
2023-09-21 13:04:40 <mps> anyway, I'm keeping mesa-asahi on my http server
2023-09-21 16:10:18 <ptrc> https://gitlab.alpinelinux.org/alpine/aports/-/jobs/1119737
2023-09-21 16:10:37 <ptrc> aaaand we got another hanging stunnel build, this time for the past 350 minutes :p
2023-09-21 16:10:52 <ptrc> https://gitlab.alpinelinux.org/alpine/aports/-/jobs/1119735
2023-09-21 16:10:58 <ptrc> both aarch64 and armv7
2023-09-21 16:11:24 <ikke> It's about time we find out what is causing it to stuck
2023-09-21 16:11:32 <ikke> instead of just killing the test and letting it continue
2023-09-21 16:21:43 <ptrc> https://gitlab.alpinelinux.org/chereskata/aports/-/jobs/1119851
2023-09-21 16:21:45 <ptrc> also this!
2023-09-21 16:21:50 <ptrc> that's not even stunnel
2023-09-21 16:22:10 <ptrc> i assume the proper way of finding out would be logging into the runner machine/container?
2023-09-21 16:23:36 <ikke> Seems like that one is doing something over and over?
2023-09-21 16:23:52 <ikke> I'm looking on the host 
2023-09-21 16:24:06 <ptrc> it is, with the timeout getting increasingly longer for some reason
2023-09-21 16:24:25 <ikke> bind(3, {sa_family=AF_INET6, sin6_port=htons(20616), sin6_flowinfo=htonl(0), inet_pton(AF_INET6, "::1", &sin6_addr), sin6_scope_id=0}, 28) = -1 EADDRNOTAVAIL (Address not available) <0.000013>
2023-09-21 16:25:52 <ikke> Seems like ipv6 is not enabled in the containers
2023-09-21 16:45:42 <ikke> ptrc: enabled ipv6 on the arm* ci runners
2023-09-21 16:57:47 <ikke> hmm, high ping loss means that at least something is coming through
2023-09-21 16:59:06 <ikke> was just a momentary blip
2023-09-21 17:00:03 <ikke> telmich: for the record, we have no contact with vigir23 or the builders at the moment
2023-09-22 15:26:40 <ikke> gitlab 16.4 released
2023-09-24 11:36:25 <mps> algitbot: retry master
2023-09-25 06:26:19 <ikke> telmich: fyi, our infra is still unreachable. 
2023-09-26 18:14:54 <mps> anyone of pmOS people have time to look at !51827
2023-09-26 18:20:54 <ptrc> you could try asking in #-devel
2023-09-26 18:21:49 <mps> ptrc: thanks, just asked on #pmOS -dev
2023-09-26 18:22:08 <ptrc> that works too, i guess :p
2023-09-26 18:22:31 <mps> hah, you mean alpine-devel? no thanks
2023-09-27 17:05:09 <celie> It seems 3.15 and 3.16 aren't getting updated packages for ARM, @mbentley on gitlab.a.o mentioned that curl is still at version 8.2.1-r0 for aarch64/armv7/armhf
2023-09-27 20:07:05 <ikke> celie: those builders are on the HW that is unavailble. I'd hoped they would be back before that became relevant
2023-09-27 20:18:14 <mps> ikke: you don't have access to move them?
2023-09-27 20:18:31 <ikke> No
2023-09-27 20:18:47 <ikke> For the other builders, we created new builders the previous time
2023-09-27 20:19:13 <mps> huh
2023-09-27 20:20:37 <mps> now I have optical link so could keep builders here
2023-09-28 14:52:49 <ptrc> https://git.alpinelinux.org/cgit/dmvpn-tools//snapshot/dmvpn-tools-1.4.0.tar.bz2 ( the source for dmvpn ) returns 400
2023-09-28 14:53:04 <ptrc> ah, should've read the page, nevermind
2023-09-28 14:53:13 <ptrc> ( .tar.bz2 got apparently disabled )
2023-09-28 14:53:38 <ptrc> nevermind
2023-09-28 14:53:41 <ptrc> .tar.gz also returns 400
2023-09-28 15:07:29 <ikke> I think psykose disabled the snapshot feature entirely
2023-09-28 15:22:24 <ikke> ptrc: I guess we should change it to source it from https://gitlab.alpinelinux.org/alpine/dmvpn-tools
2023-09-28 15:28:46 <ncopa> I think we need document how to unsubscribe to mailing lists somewhere. https://lists.alpinelinux.org/~alpine/announce
2023-09-28 15:29:56 <ncopa> i dont remember the exact unsubscribe email for alpine announce
2023-09-28 15:30:11 <ncopa> and i dont get emails i post myself, so I cannot check any headers
2023-09-28 15:30:13 <ikke> ncopa: it's at least in the headers
2023-09-28 15:30:52 <ncopa> could you please paste me the exact email?
2023-09-28 15:30:54 <ikke> It's the mailing list address with '+unsubscribe'
2023-09-28 15:31:19 <ikke> ~alpine/announce+unsubscribe@lists.alpinelinux.org if I'm not mistaken
2023-09-28 15:32:35 <ikke> ncopa: ~alpine/aports+subscribe@lists.alpinelinux.org for aports, so should be correct
2023-09-28 15:32:54 <ikke> sorry, ~alpine/aports+unsubscribe@lists.alpinelinux.org
2023-09-28 15:33:36 <ncopa> thanks!
2023-09-28 20:07:26 <ikke> Applied security update to gitlab
2023-09-28 21:08:49 <ptrc> ikke: the runners are messed up
2023-09-28 21:08:52 <ptrc> https://gitlab.alpinelinux.org/alpine/aports/-/jobs/1127946
2023-09-28 21:08:59 <ptrc> though it's a temporary issue
2023-09-28 21:09:03 <ptrc> but it doesn't resolve itself after retrying
2023-09-28 21:11:34 <ptrc> as in, they sometimes go through, and sometimes don't
2023-09-29 04:54:11 <ikke> ptrc: https://www.dockerstatus.com/pages/incident/533c6539221ae15e3f000031/6515dbc89fabff05350ae18d
2023-09-29 04:54:14 <ikke> upstream docker issue
2023-09-29 10:28:20 <ikke> lol: %spinfile-names.dat%%spinfile-lnames.dat%%random-1-100%@hujk.carmanial.com