| bgstack15 | Where are the mailing list archives now? https://mailinglists.dyne.org/pipermail/devuan-dev/ is 404 | 16:29 |
|---|---|---|
| fsmithred_ | bgstack15, https://lists.dyne.org/lurker/list/devuan-dev.en.html | 16:50 |
| bgstack15 | Thanks, fsmithred_! We need to get certain pages updated. Our page https://www.devuan.org/os/community links to https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/devuan-dev which links to the 404-ing Archives page. | 16:53 |
| golinux | https://lists.dyne.org/lurker/mindex/dng@20380101.000000.00000000.en.html | 17:07 |
| golinux | Probably need to be changed on the website too | 17:07 |
| golinux | Xenguy: ^^^ | 17:08 |
| golinux | bgstack15: Thanks for finding that! :) | 17:08 |
| golinux | I have always used that address | 17:09 |
| golinux | And also https://lists.dyne.org/lurker/mindex/devuan-dev@20380101.000000.00000000.en.html | 17:09 |
| Xenguy | Sounds like mailing list migration woes | 17:17 |
| fsmithred_ | I use this one for dng: https://lists.dyne.org/lurker/list/dng.en.html | 17:42 |
| golinux | Xenguy: Pipermail has been mia for a very long time. | 18:03 |
| golinux | Dyne still hosts the dng list and the mail archives. | 18:04 |
| golinux | devuan-dev has been migrated. At least that's how I understand things . . . | 18:05 |
| Xenguy | Looks like a search/replace operation, so I would just need the canonical URL to use to replace | 18:07 |
| Xenguy | Seems like fsmithred_ has one | 18:07 |
| Xenguy | Thanks for the bug report bgstack15 | 18:07 |
| amesser | LeePenn: are you there? | 19:05 |
| LeePen | amesser: Hi | 19:07 |
| amesser | Hi! Hope you're well! | 19:10 |
| LeePen | OK thanks, and you? | 19:11 |
| LeePen | Thanks for the new apt and libvirt | 19:11 |
| amesser | I got an email related to libvirtd with missing symbol in lxc/fuse3. It seemed to me, that libvirtd was build for testing,experimental against fuse3 3.13.0-1 | 19:12 |
| amesser | But in testing we have only fuse3 3.12.0-1 in the repos? | 19:12 |
| amesser | I just got back from work by bike. (Midlife Crisis sports :-D) | 19:13 |
| LeePen | :) | 19:13 |
| amesser | But besides that im fine. Thanks! | 19:13 |
| LeePen | Let me look. It might need a binNMU/binary rebuild -- which we can now do! | 19:13 |
| amesser | (I also started a project to build a modular Eurorack Synth last year :-) | 19:14 |
| LeePen | Excellent. How is teh theremin? | 19:14 |
| amesser | A bit dusty. Did not yet get into the mood to learn to play it well. But maybe I can connect it with the synth :-) | 19:15 |
| amesser | However, a relative of me was for a visit here. It took him about an hour to play a song. | 19:16 |
| LeePen | amesser: there is no sign that there is an API change which might require a transition between fuse3 3.12 and 3.13: https://tracker.debian.org/pkg/fuse3 | 19:29 |
| LeePen | That doesn't mean there isn't, just that nobody has noticed so far. | 19:30 |
| LeePen | I don't see libvirt-daemon depending on fuse at all. Is it just passing an error through? | 19:30 |
| amesser | its probably libvirt-daemon-driver-lxc | 19:31 |
| amesser | Let me forward the mail to you | 19:31 |
| LeePen | Or submit it to the BTS so we can track it properly? | 19:32 |
| amesser | ok, thats probably sensible | 19:32 |
| bgstack15 | Devuan meet tomorrow, Thursday, Jan 26, at 20:30 UTC. Pad is here: https://pad.dyne.org/code/#/2/code/edit/Qo90Nxdiruh1VY4ZzA3R99LD/ | 19:44 |
| amesser | LeePen: Besides this issue. There is something else: I just recovered that the war also an libvirt update for chimaera/bullseye. I have packaged it but I don't know how to trigger a build for it: https://git.devuan.org/devuan/libvirt/src/branch/suites/chimaera-proposed-updates | 19:46 |
| bgstack15 | Xenguy: I've tried searching git.devuan.org for a repo that manages the website but I couldn't find any. Would it be possible to add me to the list of "Devuan developers past and present" on page https://www.devuan.org/os/team? | 19:48 |
| LeePen | amesser: I will do that now. | 19:51 |
| amesser | Thanks! | 19:52 |
| LeePen | amesser: the fuse3 symbol thing looks odd. | 19:57 |
| LeePen | There is an archlinux report from yesterday as well: https://bbs.archlinux.org/viewtopic.php?pid=2081000 | 19:57 |
| amesser | Saw that, but didn't look at the date. | 20:04 |
| LeePen | amesser: just noticed the version on the chimaera-pu build. | 20:10 |
| amesser | Thanks! | 20:11 |
| LeePen | Because it is still using the old +devuanN numbers, dpkg sorts +deb11u1devuan1 earlier than +devuan3. | 20:12 |
| LeePen | I think it will have to be +devuan4 to allow upgrades. | 20:12 |
| LeePen | A pain, I know. | 20:12 |
| amesser | I can push another version? | 20:17 |
| amesser | I found something in libfuse 3.13.0 changelog: "libfuse can now be build explicitly for C libraries without symbol versioning support." | 20:18 |
| amesser | Maybe different build settings are needed for libfuse now to have it backwards compatible? So this could be an issue with Debian as well. | 20:19 |
| amesser | libfuse 3.13.0 is just about two weeks old. | 20:19 |
| LeePen | Yes, me to. I have just followed up to #738. | 20:32 |
| LeePen | Do push another version for chimaera-pu and I will build again. | 20:32 |
| bb|hcb | amesser: objdump -T blabla.so will tell what is there... If the new one is now versioned, I see no proper way to manage without soname bump. It is possible, but is not worth the clutter. Maybe they tried to avoid the new queue? FWIW now is the season of transition freeze... | 20:33 |
| bb|hcb | Just checked libfuse3.so.3.10.3 is versioned... | 20:37 |
| LeePen | 3.13 is due to migrate to testing in 2 days. | 20:38 |
| LeePen | That should fix it. | 20:38 |
| LeePen | If not, Debian will have to sort out another way. | 20:39 |
| bb|hcb | 3.13.0(-1/-2) are also properly versioned with backward compatible symbols, most probably it is something minor | 20:48 |
| amesser | I think libfuse it self is broken. During compilation of the package itself, it sets "HAVE_LIBC_VERSIONED_SYMBOLS". This define however, is not set in any of its development headers. | 20:52 |
| amesser | Thus apps using libfuse will not have the define set. | 20:52 |
| amesser | This leads to apps expecting "fuse_new_31" symbol while libfuse3.13 only provides "fuse_new" | 20:53 |
| amesser | I believe the same linker error will occur with libfuse 3.13 | 20:54 |
| amesser | So its probably something to forward to ebian & upstream as well | 20:55 |
| amesser | Hmm, seems they do really awkwards thing in this library | 21:03 |
| amesser | 3.13 seems to have the symbol indeed. | 21:03 |
| bb|hcb | That is a strange concept - normally fuse_new_xy should be hidden and only avail to binaries compiled against ver x.y, not redefined in headers... | 21:03 |
| bb|hcb | Time for me to call it a day | 21:04 |
| amesser | LeePen: I have just pushed libvirt 7.0.0-3+devuan4 for chimaera-po | 21:18 |
| LeePen | amesser: there were libxlxml2domconfigtest failures on arm64 and armhf on the last one. | 21:29 |
| amesser | Hmm. It built fine for me here. I can not test arm64 builds however :-) | 21:39 |
| LeePen | I just run it again. | 21:47 |
| * amesser crosses fingers :-D | 21:52 | |
| amesser | Have to go now, just to tired to stay up. Please drop me an e-mail if it still failes. | 21:52 |
| LeePen | OK. Check on #devuan-ci too. | 21:55 |
| LeePen | Thanks. | 21:55 |
| golinux | bgstack15: https://git.devuan.org/devuan/www.devuan.org | 22:04 |
| Xenguy | bgstack15, I would think definitely, I had just assumed you were already listed | 23:32 |
| bgstack15 | Ha, I'm sending up a merge request right now! | 23:32 |
| bgstack15 | I'm working on my visibility for open source contributions. | 23:33 |
| Xenguy | np, I have not been active recently, so I'll try to de-rust the machinery | 23:34 |
| bgstack15 | Hope that's the right branch: https://git.devuan.org/bgstack15/www.devuan.org/pulls/1 | 23:34 |
| Xenguy | Me too, been awhile | 23:34 |
| bgstack15 | Grr, it didn't keep "devuan:master" in there. | 23:34 |
| golinux | Shouldn't you be working on your own branch? | 23:35 |
| bgstack15 | Yes, and I am trying to get a merge request created that wants to merge my branch into 'master' of the devuan/ repo. | 23:35 |
| bgstack15 | Hm, maybe master isn't the right one... is "new-beta" the correct target branch? That's what it created: add bgstack15 to team list <https://git.devuan.org/devuan/www.devuan.org/pulls/22> | 23:37 |
| Xenguy | new-beta is the usual target IIUC | 23:38 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!