| golinux | Is there any reason to still have libflashplayer.so installed? | 00:13 |
|---|---|---|
| golinux | It was accessed in July via firemenu. | 00:14 |
| golinux | That would have been after the upgrade to chimaera | 00:15 |
| oliv3 | hi there, i'd like to give devuan a try, mostly because i'm a debian user, DM for some projects, and really don't like to have systemd around. so quick question, how far is devuan behind debian ? from what i see on https://repology.org/project/lebiniou/versions, only devuan/unstable is up-to-date | 00:17 |
| oliv3 | is there some place where syncing with official debian is described ? | 00:18 |
| golinux | oliv3: Greetings! Have a look here https://git.devuan.org/devuan/amprolla3 | 00:20 |
| fsmithred | I have no libflashplayer on beowulf. | 00:35 |
| fsmithred | oliv3, devuan is normally about four minutes behind debian for packages. | 00:39 |
| fsmithred | I don't understand the chart you linked. Debian 11 = Devuan 4. Same versions. | 00:41 |
| * rrq interesting site | 00:47 | |
| rwp | That site is out of date for Devuan as it lists Debian 12 Testing "bookworm" but does not list Devuan 5 Testing "Deadalus". Missing it. | 00:53 |
| rwp | And by listing Debian 12 "Bookworm" without saying "Testing" it makes it appear to be released there. But it is not. | 00:54 |
| rwp | Clearly the author of that site thinks the rolling release model is the only way to do things. :-( | 00:54 |
| oliv3 | maybe the author would like to hear about misses | 00:55 |
| rrq | yes, it conflates the conepts of "repository", "release" and "soure list point" | 01:03 |
| rrq | "sources.list point" | 01:03 |
| rrq | still quite interesting I think | 01:04 |
| oliv3 | funny you guys never heard of it | 01:05 |
| oliv3 | it's well known | 01:05 |
| oliv3 | even Debian uses it as a reference to "other distros" | 01:06 |
| oliv3 | see e.g. the "other distros" link on https://tracker.debian.org/pkg/lebiniou :) | 01:06 |
| oliv3 | anyway | 01:06 |
| oliv3 | as i said maybe the author would like to hear about issues | 01:07 |
| rrq | its "docs" page has Copyright (C) 2016-2022 Dmitry Marakasov <amdmi3@amdmi3.ru>. | 01:08 |
| oliv3 | open-source is not about yelling at projects, rather contributing to them | 01:09 |
| oliv3 | rrq: he's also a freebsd commiter https://reviews.freebsd.org/p/AMDmi3/ | 01:15 |
| oliv3 | just sayin' | 01:17 |
| rrq | yes, I didin't complain.. just tried to find some identity behind the site | 01:17 |
| rrq | .. just because I didn't know it before | 01:17 |
| oliv3 | nw | 01:19 |
| Ooft | https://www.youtube.com/watch?v=mgtpdgGYbV4 | 08:02 |
| Ooft | "Is this my next laptop? - HP Dev One" | 08:02 |
| gnarface | take it to #devuan-offtopic | 08:03 |
| Ooft | What for? nobody is talking | 08:05 |
| Ooft | But okay | 08:05 |
| Ooft | My job here is done. Bye again! | 08:11 |
| u-amarsh04 | the new mmap changes post-6.0.0-rc7 kernel seem to be working some magic, I have 1920*1080 video playing, Discord and Firefox with over 1,000 tabs open behaving | 16:53 |
| bb|hcb | oliv3: Maybe I am late to the pary, just wanted to share that from my experience the person maintaining repology is helpful and responsive. Just let them know about the problem you are seeing and it will get fixed soon | 17:50 |
| bb|hcb | s/pary/party | 17:51 |
| oliv3 | bb|hcb: hi, i'm fine with repology :) it's rwp/rrq who had remarks about it | 18:09 |
| Afdal | All right, I'm thinkin of advancing past Chimaera | 21:21 |
| Afdal | Someone explain the difference between Daedalus and Ceres to me... | 21:21 |
| fsmithred | ceres = Unstable. New versions go into there first. | 21:24 |
| fsmithred | ceres = sid | 21:24 |
| fsmithred | daedalus is testing, which is currently the same as bullseye. | 21:25 |
| fsmithred | stuff moves into the testing suite after a week. | 21:25 |
| bb|hcb | chimaera=bullseye, daedalus=bookworm | 21:25 |
| fsmithred | when bullseye is released as stable, daedalus will still be in testing for a little while. That a | 21:26 |
| bb|hcb | Confusing, eh? ;) | 21:26 |
| fsmithred | yeah | 21:26 |
| fsmithred | three B's in a row is fucking stupid. | 21:26 |
| fsmithred | anyway, that asynchronous state is why we say to use codenames and not "stable" or "testing" in sources.list | 21:27 |
| rwp | Testing (Daedalus at this moment) is similar to a rolling release but also not quite a rolling release. | 21:27 |
| rwp | But if a bug is not found until after it rolls into Testing then the report prevents upgrades to that package in Testing until it is fixed in Unstable first. | 21:27 |
| rwp | If a bug is found in Unstable *and reported* then it will prevent a package from rolling into Testing. | 21:27 |
| rwp | For a long time there was no explicit security support for Testing. So you can see the problem. But due to popular demand Testing now has a security support path. | 21:28 |
| fsmithred | and in either testing or unstable, you can expect a lot of updates until testing goes into freeze | 21:28 |
| Afdal | Okay, so Daedalus should generally be less prone to catastrophic bugs in Daedalus than Ceres | 21:28 |
| fsmithred | huh? | 21:28 |
| rwp | fsmithred, Was the huh directed at my comment about testing security support? | 21:28 |
| fsmithred | afdal, if you want to run daedais, it's probably a good idea to add ceres and pin it to a low priority | 21:29 |
| fsmithred | rwp yes | 21:29 |
| rwp | I'll double check.... | 21:29 |
| Afdal | What do you mean? | 21:29 |
| fsmithred | daedalus | 21:29 |
| Afdal | "add ceres and pin it to a low priority"? | 21:29 |
| fsmithred | see apt_preferences for info on pinning. You can have both daedalus and ceres in sources.list | 21:29 |
| Afdal | eh... | 21:29 |
| bb|hcb | Afdal: Debian's definition of stable does not mean bugless - only the known ones get fixed; same is valid for security | 21:30 |
| Afdal | What does that accomplish? | 21:30 |
| fsmithred | but packages will only install from daedalus unless you pin ceres low | 21:30 |
| Afdal | Okay, what's the point of doing that? | 21:30 |
| fsmithred | then if something in testing is broken, you can install the fix when it gets into ceres | 21:30 |
| Afdal | "can"? | 21:30 |
| fsmithred | apt -t ceres install <fixed version> | 21:30 |
| fsmithred | instead of waiting a week | 21:30 |
| Afdal | hmm | 21:31 |
| Afdal | So prioritizing repos with apt is how you get away with having multiple version repos on your sources.list eh... | 21:31 |
| fsmithred | yeah, pinning | 21:31 |
| Afdal | Why don't I already know this stuff :'c | 21:31 |
| Afdal | I'm such a rube... | 21:32 |
| fsmithred | never needed it? | 21:32 |
| fsmithred | historically, debian has always "just worked" | 21:32 |
| fsmithred | well, maybe not always, but for a long time | 21:32 |
| Afdal | I've definitely wanted to grab never software versions on more recent repos before without upgrading all my stuff | 21:32 |
| Afdal | Usually I just add a repo temporarily and cross my fingers that everything will work out all right | 21:32 |
| fsmithred | yeah, well you can mix testing and unstable, but you can't mix stable and testing (or unstable) | 21:32 |
| Afdal | Oh, I see | 21:33 |
| bb|hcb | Stable means that no intrusive changes are done (e.g. new version of something with incompatible config), so applying security and updates to stable is like 99% guaranteeing that you will not need to touch anything after update. Unless your config relied on a bug that got fixed ;) | 21:33 |
| Afdal | testing and unstable actually aren't all that far behind one another | 21:33 |
| Afdal | that's why that works | 21:33 |
| fsmithred | right | 21:33 |
| fsmithred | and one flows into the other. They are part of the same process | 21:34 |
| fsmithred | stable is a finished product that gets no new versions other than security and bug fixes | 21:34 |
| fsmithred | stable/unstable refers to how fast versions change, not stability of the system | 21:35 |
| fsmithred | people running daedalus or ceres are pretty happy with it. If you're ok with the possiblity of occasional breakage, then you might be happy with it, too. | 21:37 |
| bb|hcb | fsmithred: Right, but it is good to know that the testing migration goes both ways. E.g. if a package migrated to testing from unstable after 5 days wait, and a RC bug is filed, it will get removed from testing. For a user who is updating regularly this may mean broken dependencies or uninstallable packages. About 1y back virtualbox was in that state | 21:37 |
| fsmithred | who decides if a bug is RC? | 21:38 |
| bb|hcb | The one who files it... There is also automatic RC bug filing based on automated checks | 21:38 |
| rwp | fsmithred, Debian definitely has security upgrades for Testing. Devuan appears not to support security for Testing at this time that I can tell. | 21:39 |
| bb|hcb | But that does no happen immediately and there is time for others to react and reclassify the bug's severity | 21:40 |
| bb|hcb | Another unpleasant scenario is when for example a dependency goes FTBFS - then after e.g. 30days the whole tree of dependants can get removed | 21:40 |
| rwp | Regarding running Testing, See the "Best practices for Testing users" here https://wiki.debian.org/DebianTesting | 21:43 |
| rwp | And for Unstable users also https://wiki.debian.org/DebianUnstable#What_are_some_best_practices_for_testing.2Fsid_users.3F | 21:44 |
| rwp | Since Devuan is an overlay those best practices also apply after being mapped into Devuan resources. | 21:44 |
| rwp | This applies to Debian but unfortunately apparently not to Devuan, but useful information regardless: https://www.debian.org/doc/manuals/securing-debian-manual/ch10.en.html#security-support-testing | 21:51 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!