[Mageia-dev] Mageia-dev] mageia 2 and systemd
Oddly enough, I have four machines running Mageia - starting with a fresh Mageia 1 install and updating to Cauldron (keeping old /home from Mandriva) Every time I have tried systemd - something breaks - mostly sound and/or networking (eth0) - When I disable allow interface to be controlled by network-manager - I can then use ifup to manually start eth0 - and revert back to sysvinit (urpmi sysvinit) Unfortunately, I don't have the time to troubleshoot the sound issues, and have found the easiest solution is to fallback to sysvinit (so long that it works!) Cheers, R.Fox
Re: [Mageia-dev] Mageia-dev] mageia 2 and systemd
2011/10/30 Robert Fox l...@foxconsult.net: Unfortunately, I don't have the time to troubleshoot the sound issues, and have found the easiest solution is to fallback to sysvinit (so long that it works!) You can have a look here: https://bugs.mageia.org/show_bug.cgi?id=2982 Cheers, R.Fox Regards, Rémi / Akien
Re: [Mageia-dev] Mageia-dev] mageia 2 and systemd
'Twas brillig, and Robert Fox at 30/10/11 09:38 did gyre and gimble: Unfortunately, I don't have the time to troubleshoot the sound issues, Honestly I've mentioned this so many times now it's getting ridiculous. It's been covered about half a dozen times on this list and I have a few bugs opened (and subsequently closed) about it too. Please read up properly on the reasons why this is the case and either give appropriate debug output or be happy the problem is fixed. Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] Mageia-dev] mageia 2 and systemd
On Sun, 2011-10-30 at 11:29 +, Colin Guthrie wrote: 'Twas brillig, and Robert Fox at 30/10/11 09:38 did gyre and gimble: Unfortunately, I don't have the time to troubleshoot the sound issues, Honestly I've mentioned this so many times now it's getting ridiculous. It's been covered about half a dozen times on this list and I have a few bugs opened (and subsequently closed) about it too. Please read up properly on the reasons why this is the case and either give appropriate debug output or be happy the problem is fixed. Col Sorry Colin - wasn't trying to irk you. R.Fox
Re: [Mageia-dev] Mageia-dev] mageia 2 and systemd
'Twas brillig, and Robert Fox at 30/10/11 11:52 did gyre and gimble: Sorry Colin - wasn't trying to irk you. Yeah I know, sorry for being grumpy. I do appreciate we need to handle pam.d stuff better but I've tried raising this discussion a few times and no-ones really doing anything about it :( Guess I'll just have to pull my finger out and do it myself. Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] Mageia-dev] mageia 2 and systemd
On Sun, 2011-10-30 at 12:16 +, Colin Guthrie wrote: 'Twas brillig, and Robert Fox at 30/10/11 11:52 did gyre and gimble: Sorry Colin - wasn't trying to irk you. Yeah I know, sorry for being grumpy. I do appreciate we need to handle pam.d stuff better but I've tried raising this discussion a few times and no-ones really doing anything about it :( Guess I'll just have to pull my finger out and do it myself. Col No problem - I wanted to change back to systemd, but now I get a strange conflict: [root@foxbase rfox]# urpmi systemd-sysvinit installing systemd-sysvinit-37-7.mga2.x86_64.rpm from /var/cache/urpmi/rpms Preparing... # Installation failed:file /sbin/halt from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/init from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/poweroff from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/reboot from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/runlevel from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/shutdown from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/telinit from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 [root@foxbase rfox]# Do I have to use --force here?? Thx, R.Fox
Re: [Mageia-dev] Mageia-dev] mageia 2 and systemd
'Twas brillig, and Robert Fox at 30/10/11 12:27 did gyre and gimble: On Sun, 2011-10-30 at 12:16 +, Colin Guthrie wrote: 'Twas brillig, and Robert Fox at 30/10/11 11:52 did gyre and gimble: Sorry Colin - wasn't trying to irk you. Yeah I know, sorry for being grumpy. I do appreciate we need to handle pam.d stuff better but I've tried raising this discussion a few times and no-ones really doing anything about it :( Guess I'll just have to pull my finger out and do it myself. Col No problem - I wanted to change back to systemd, but now I get a strange conflict: [root@foxbase rfox]# urpmi systemd-sysvinit installing systemd-sysvinit-37-7.mga2.x86_64.rpm from /var/cache/urpmi/rpms Preparing... # Installation failed: file /sbin/halt from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/init from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/poweroff from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/reboot from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/runlevel from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/shutdown from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 file /sbin/telinit from install of systemd-sysvinit-37-7.mga2.x86_64 conflicts with file from package sysvinit-2.87-10.mga1.x86_64 [root@foxbase rfox]# Do I have to use --force here?? Hmm, not sure, I guess it's not conflicting or obsoleting sysvinit or something. You can just pass init=/sbin/systemd to the kernel command line to boot with systemd tho' - you don't absolutely need the systemd-sysvinit package installed. Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] Mageia-dev] mageia 2 and systemd
On Sun, Oct 30, 2011 at 10:38 AM, Robert Fox l...@foxconsult.net wrote: Oddly enough, I have four machines running Mageia - starting with a fresh Mageia 1 install and updating to Cauldron (keeping old /home from Mandriva) Every time I have tried systemd - something breaks - mostly sound and/or networking (eth0) - When I disable allow interface to be controlled by network-manager - I can then use ifup to manually start eth0 - and revert back to sysvinit (urpmi sysvinit) Unfortunately, I don't have the time to troubleshoot the sound issues, and have found the easiest solution is to fallback to sysvinit (so long that it works!) Cheers, R.Fox have you disabled speedboot ? it fixed sound and network pb for me
Re: [Mageia-dev] Mageia-dev] mageia 2 and systemd
'Twas brillig, and D.Morgan at 30/10/11 12:47 did gyre and gimble: On Sun, Oct 30, 2011 at 10:38 AM, Robert Fox l...@foxconsult.net wrote: Oddly enough, I have four machines running Mageia - starting with a fresh Mageia 1 install and updating to Cauldron (keeping old /home from Mandriva) Every time I have tried systemd - something breaks - mostly sound and/or networking (eth0) - When I disable allow interface to be controlled by network-manager - I can then use ifup to manually start eth0 - and revert back to sysvinit (urpmi sysvinit) Unfortunately, I don't have the time to troubleshoot the sound issues, and have found the easiest solution is to fallback to sysvinit (so long that it works!) Cheers, R.Fox have you disabled speedboot ? it fixed sound and network pb for me Keep in mind that the latest packages will disable speedboot automagically. Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] mageia 2 and systemd
'Twas brillig, and philippe makowski at 27/10/11 20:13 did gyre and gimble: 2011/10/13 Guillaume Rousse guillomovi...@gmail.com: Le 13/10/2011 10:08, philippe makowski a écrit : what do we do with %_post_service and %_preun_service ? Those macros will have to be enhanced to handle new behaviours, without having to modify packages. hum an how to manage sockets ? moving from xinetd to systemd sockets ? In an ideal world we would use systemd for traditional xinetd stuff, but being diplomatic, I think we should wait until after mga2 for that. Sure there is arguably additional overhead of running xinetd but while we support both systemd and sysvinit, xinetd serves a purpose. I propose that we should ultimately deprecated xinted in mga3, but that discussion can happen later. Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] mageia 2 and systemd
2011/10/13 Guillaume Rousse guillomovi...@gmail.com: Le 13/10/2011 10:08, philippe makowski a écrit : what do we do with %_post_service and %_preun_service ? Those macros will have to be enhanced to handle new behaviours, without having to modify packages. hum an how to manage sockets ? moving from xinetd to systemd sockets ?
Re: [Mageia-dev] mageia 2 and systemd
On Wed, 12 Oct 2011 21:00:58 +0200 philippe makowski makowski.mag...@gmail.com wrote: Hi, how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? some doc from Fedora for packaging : http://fedoraproject.org/wiki/Packaging:Guidelines:Systemd http://fedoraproject.org/wiki/Packaging:ScriptletSnippets#Systemd I've had a very bad experience upgrading to systemd-sysvinit on my desktop Core i3 machine. The boot took forever, and I ended up giving up, booting from a LiveCD, and restoring the sysvinit package, which made booting the system OK again. Here are the relevant logs from #systemd on Freenode (I am rindolf): Oct 13 09:30:40 rindolf Hi all. Oct 13 09:31:02 rindolf Today systemd-sysvinit replaced sysvinit (on my Mageia 2 system after a urpmi upgrade), but after I rebooted my system won't boot. It got hang at the udev step (the first one). I recall that when I rebooted, it warned me that it couldn't find libsystemd-daemons.so.1 or something like that. How can I restore booting? Oct 13 09:45:28 rindolf Anyone? Oct 13 09:46:37 MK_FG rindolf, I suspect it just hangs not being able to detect your hard drives for some reason Oct 13 09:46:49 MK_FG rindolf, In that case, it should time out after a while Oct 13 09:47:07 MK_FG rindolf, But in any case, it should help to enable debug output Oct 13 09:47:10 rindolf MK_FG: well, now after I hit Ctrl+C it continues, but hangs again. Oct 13 09:47:17 rindolf MK_FG: how can I enable debug output? Oct 13 09:47:21 MK_FG http://fedoraproject.org/wiki/How_to_debug_Systemd_problems Oct 13 09:48:01 MK_FG But it's all in the mans, too, if you'll end up w/o a browser Oct 13 09:49:14 rindolf MK_FG: I have a second computer. Oct 13 10:01:29 rindolf MK_FG: now it gets stuck on loading shorewall. Oct 13 10:01:36 rindolf MK_FG: it got stuck several times before Oct 13 10:01:51 rindolf MK_FG: and no matter how many times I press Ctrl+C it won't stop that. Oct 13 10:01:59 rindolf MK_FG: booting is slower than ever this way. Oct 13 10:02:07 MK_FG It should timeout eventually Oct 13 10:02:07 rindolf I thought systemd was supposed to make booting faster. Oct 13 10:02:17 rindolf MK_FG: eventually is the key word here. Oct 13 10:02:17 sztanpet works for me Oct 13 10:02:38 MK_FG Yeah, a minute or two by default Oct 13 10:02:38 rindolf sztanpet: what does work for you? Oct 13 10:02:49 sztanpet systemd operating as expected Oct 13 10:03:00 sztanpet but then again, i dont have your distro Oct 13 10:03:20 MK_FG But it's abnormal behavior, something should be fixed, not timeout and be killed on every boot Oct 13 10:05:48 rindolf sztanpet: someone once told me that works for me is the oldest excuse in the programmer's book. Oct 13 10:05:52 rindolf And it's not very helpful. Oct 13 10:06:15 sztanpet indeed Oct 13 10:06:26 MK_FG Neither is unconstructive criticism like nothing works! Oct 13 10:06:38 rindolf MK_FG: well, I told you what doesn't work for me. Oct 13 10:06:55 MK_FG I think you know what I mean Oct 13 10:07:15 rindolf MK_FG: no, I don't. Oct 13 10:07:25 MK_FG Well, nevermind then ;) Oct 13 10:09:50 bochecha rindolf, the default timeout can be 5 minutes (for services with legacy init scripts), did you wait that long? Oct 13 10:16:55 rindolf bochecha: some of the times. Oct 13 10:17:20 rindolf bochecha: I now tried a “failsafe” boot and systemd gets stuck on an infinite Trying to load D-Bus loop. Oct 13 10:18:29 bochecha possibly dbus doesn't start, and systemd tries again, and again and... ? Regards, Shlomi Fish -- - Shlomi Fish http://www.shlomifish.org/ What does Zionism mean? - http://shlom.in/def-zionism I learned a lot from my teachers, and from my friends more than my teachers, and from my pupils the most. — Rabbi Hanina Please reply to list if it's a mailing list post - http://shlom.in/reply .
Re: [Mageia-dev] mageia 2 and systemd
Le 14/10/2011 08:53, Shlomi Fish a écrit : On Wed, 12 Oct 2011 21:00:58 +0200 philippe makowskimakowski.mag...@gmail.com wrote: Hi, how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? some doc from Fedora for packaging : http://fedoraproject.org/wiki/Packaging:Guidelines:Systemd http://fedoraproject.org/wiki/Packaging:ScriptletSnippets#Systemd I've had a very bad experience upgrading to systemd-sysvinit on my desktop Core i3 machine. The boot took forever, and I ended up giving up, booting from a LiveCD, and restoring the sysvinit package, which made booting the system OK again. Here are the relevant logs from #systemd on Freenode (I am rindolf): But why did you the change this way ? Using 'init=/bin/systemd' on kernel command line is enough to switch to systemd, and allows you to fallback to sysinit if needed. -- BOFH excuse #100: IRQ dropout
Re: [Mageia-dev] mageia 2 and systemd
'Twas brillig, and Guillaume Rousse at 14/10/11 08:51 did gyre and gimble: Le 14/10/2011 08:53, Shlomi Fish a écrit : On Wed, 12 Oct 2011 21:00:58 +0200 philippe makowskimakowski.mag...@gmail.com wrote: Hi, how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? some doc from Fedora for packaging : http://fedoraproject.org/wiki/Packaging:Guidelines:Systemd http://fedoraproject.org/wiki/Packaging:ScriptletSnippets#Systemd I've had a very bad experience upgrading to systemd-sysvinit on my desktop Core i3 machine. The boot took forever, and I ended up giving up, booting from a LiveCD, and restoring the sysvinit package, which made booting the system OK again. Here are the relevant logs from #systemd on Freenode (I am rindolf): But why did you the change this way ? Using 'init=/bin/systemd' on kernel command line is enough to switch to systemd, and allows you to fallback to sysinit if needed. It's the default for the moment is it not - to try and encourage more testing? That said, for the case like this, it's a bit of a pain to recover when something goes horribly wrong! Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] mageia 2 and systemd
'Twas brillig, and Shlomi Fish at 14/10/11 07:53 did gyre and gimble: On Wed, 12 Oct 2011 21:00:58 +0200 philippe makowski makowski.mag...@gmail.com wrote: Hi, how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? some doc from Fedora for packaging : http://fedoraproject.org/wiki/Packaging:Guidelines:Systemd http://fedoraproject.org/wiki/Packaging:ScriptletSnippets#Systemd I've had a very bad experience upgrading to systemd-sysvinit on my desktop Core i3 machine. The boot took forever, and I ended up giving up, booting from a LiveCD, and restoring the sysvinit package, which made booting the system OK again. Here are the relevant logs from #systemd on Freenode (I am rindolf): Oct 13 09:30:40 rindolf Hi all. Oct 13 09:31:02 rindolf Today systemd-sysvinit replaced sysvinit (on my Mageia 2 system after a urpmi upgrade), but after I rebooted my system won't boot. It got hang at the udev step (the first one). I recall that when I rebooted, it warned me that it couldn't find libsystemd-daemons.so.1 or something like that. How can I restore booting? Oct 13 09:45:28 rindolf Anyone? Oct 13 09:46:37 MK_FG rindolf, I suspect it just hangs not being able to detect your hard drives for some reason Oct 13 09:46:49 MK_FG rindolf, In that case, it should time out after a while Oct 13 09:47:07 MK_FG rindolf, But in any case, it should help to enable debug output Oct 13 09:47:10 rindolf MK_FG: well, now after I hit Ctrl+C it continues, but hangs again. Oct 13 09:47:17 rindolf MK_FG: how can I enable debug output? Oct 13 09:47:21 MK_FG http://fedoraproject.org/wiki/How_to_debug_Systemd_problems Oct 13 09:48:01 MK_FG But it's all in the mans, too, if you'll end up w/o a browser Oct 13 09:49:14 rindolf MK_FG: I have a second computer. Oct 13 10:01:29 rindolf MK_FG: now it gets stuck on loading shorewall. Oct 13 10:01:36 rindolf MK_FG: it got stuck several times before Oct 13 10:01:51 rindolf MK_FG: and no matter how many times I press Ctrl+C it won't stop that. Oct 13 10:01:59 rindolf MK_FG: booting is slower than ever this way. Oct 13 10:02:07 MK_FG It should timeout eventually Oct 13 10:02:07 rindolf I thought systemd was supposed to make booting faster. Oct 13 10:02:17 rindolf MK_FG: eventually is the key word here. Oct 13 10:02:17 sztanpetworks for me Oct 13 10:02:38 MK_FG Yeah, a minute or two by default Oct 13 10:02:38 rindolf sztanpet: what does work for you? Oct 13 10:02:49 sztanpetsystemd operating as expected Oct 13 10:03:00 sztanpetbut then again, i dont have your distro Oct 13 10:03:20 MK_FG But it's abnormal behavior, something should be fixed, not timeout and be killed on every boot Oct 13 10:05:48 rindolf sztanpet: someone once told me that works for me is the oldest excuse in the programmer's book. Oct 13 10:05:52 rindolf And it's not very helpful. Oct 13 10:06:15 sztanpetindeed Oct 13 10:06:26 MK_FG Neither is unconstructive criticism like nothing works! Oct 13 10:06:38 rindolf MK_FG: well, I told you what doesn't work for me. Oct 13 10:06:55 MK_FG I think you know what I mean Oct 13 10:07:15 rindolf MK_FG: no, I don't. Oct 13 10:07:25 MK_FG Well, nevermind then ;) Oct 13 10:09:50 bochecharindolf, the default timeout can be 5 minutes (for services with legacy init scripts), did you wait that long? Oct 13 10:16:55 rindolf bochecha: some of the times. Oct 13 10:17:20 rindolf bochecha: I now tried a “failsafe” boot and systemd gets stuck on an infinite Trying to load D-Bus loop. Oct 13 10:18:29 bochechapossibly dbus doesn't start, and systemd tries again, and again and... ? Well I'm afraid that as the guys on #systemd said, it's nigh on impossible to say what your problem actually is without more debugging etc. For example if you manage to get the boot to work by waiting patiently enough then the output from systemd-analyze might point to the cause. I suspect to be honest that it relates to some binfmt stuff or other modprobe setups that just ultimately stall things. It's obviously not a problem that affects other people other than yourself so you have to ask yourself why your system is different, try with a clean install and then add back your config until you find the issue. Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] mageia 2 and systemd
On Fri, 14 Oct 2011 09:51:00 +0200 Guillaume Rousse guillomovi...@gmail.com wrote: Le 14/10/2011 08:53, Shlomi Fish a écrit : On Wed, 12 Oct 2011 21:00:58 +0200 philippe makowskimakowski.mag...@gmail.com wrote: Hi, how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? some doc from Fedora for packaging : http://fedoraproject.org/wiki/Packaging:Guidelines:Systemd http://fedoraproject.org/wiki/Packaging:ScriptletSnippets#Systemd I've had a very bad experience upgrading to systemd-sysvinit on my desktop Core i3 machine. The boot took forever, and I ended up giving up, booting from a LiveCD, and restoring the sysvinit package, which made booting the system OK again. Here are the relevant logs from #systemd on Freenode (I am rindolf): But why did you the change this way ? Using 'init=/bin/systemd' on kernel command line is enough to switch to systemd, and allows you to fallback to sysinit if needed. Because that's what urpmi --auto --auto-select wanted me to do. Regards, Shlomi Fish -- - Shlomi Fish http://www.shlomifish.org/ Escape from GNU Autohell - http://www.shlomifish.org/open-source/anti/autohell/ E‐mail, web feeds, and doing something productive — choose two. Please reply to list if it's a mailing list post - http://shlom.in/reply .
Re: [Mageia-dev] mageia 2 and systemd
Hello Colin, On Fri, 14 Oct 2011 09:35:07 +0100 Colin Guthrie mag...@colin.guthr.ie wrote: 'Twas brillig, and Shlomi Fish at 14/10/11 07:53 did gyre and gimble: On Wed, 12 Oct 2011 21:00:58 +0200 philippe makowski makowski.mag...@gmail.com wrote: Hi, how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? some doc from Fedora for packaging : http://fedoraproject.org/wiki/Packaging:Guidelines:Systemd http://fedoraproject.org/wiki/Packaging:ScriptletSnippets#Systemd I've had a very bad experience upgrading to systemd-sysvinit on my desktop Core i3 machine. The boot took forever, and I ended up giving up, booting from a LiveCD, and restoring the sysvinit package, which made booting the system OK again. Here are the relevant logs from #systemd on Freenode (I am rindolf): Oct 13 09:30:40 rindolf Hi all. Oct 13 09:31:02 rindolf Today systemd-sysvinit replaced sysvinit (on my Mageia 2 system after a urpmi upgrade), but after I rebooted my system won't boot. It got hang at the udev step (the first one). I recall that when I rebooted, it warned me that it couldn't find libsystemd-daemons.so.1 or something like that. How can I restore booting? Oct 13 09:45:28 rindolf Anyone? Oct 13 09:46:37 MK_FG rindolf, I suspect it just hangs not being able to detect your hard drives for some reason Oct 13 09:46:49 MK_FG rindolf, In that case, it should time out after a while Oct 13 09:47:07 MK_FG rindolf, But in any case, it should help to enable debug output Oct 13 09:47:10 rindolf MK_FG: well, now after I hit Ctrl+C it continues, but hangs again. Oct 13 09:47:17 rindolf MK_FG: how can I enable debug output? Oct 13 09:47:21 MK_FG http://fedoraproject.org/wiki/How_to_debug_Systemd_problems Oct 13 09:48:01 MK_FG But it's all in the mans, too, if you'll end up w/o a browser Oct 13 09:49:14 rindolf MK_FG: I have a second computer. Oct 13 10:01:29 rindolf MK_FG: now it gets stuck on loading shorewall. Oct 13 10:01:36 rindolf MK_FG: it got stuck several times before Oct 13 10:01:51 rindolf MK_FG: and no matter how many times I press Ctrl+C it won't stop that. Oct 13 10:01:59 rindolf MK_FG: booting is slower than ever this way. Oct 13 10:02:07 MK_FG It should timeout eventually Oct 13 10:02:07 rindolf I thought systemd was supposed to make booting faster. Oct 13 10:02:17 rindolf MK_FG: eventually is the key word here. Oct 13 10:02:17 sztanpet works for me Oct 13 10:02:38 MK_FG Yeah, a minute or two by default Oct 13 10:02:38 rindolf sztanpet: what does work for you? Oct 13 10:02:49 sztanpet systemd operating as expected Oct 13 10:03:00 sztanpet but then again, i dont have your distro Oct 13 10:03:20 MK_FG But it's abnormal behavior, something should be fixed, not timeout and be killed on every boot Oct 13 10:05:48 rindolf sztanpet: someone once told me that works for me is the oldest excuse in the programmer's book. Oct 13 10:05:52 rindolf And it's not very helpful. Oct 13 10:06:15 sztanpet indeed Oct 13 10:06:26 MK_FG Neither is unconstructive criticism like nothing works! Oct 13 10:06:38 rindolf MK_FG: well, I told you what doesn't work for me. Oct 13 10:06:55 MK_FG I think you know what I mean Oct 13 10:07:15 rindolf MK_FG: no, I don't. Oct 13 10:07:25 MK_FG Well, nevermind then ;) Oct 13 10:09:50 bochecha rindolf, the default timeout can be 5 minutes (for services with legacy init scripts), did you wait that long? Oct 13 10:16:55 rindolf bochecha: some of the times. Oct 13 10:17:20 rindolf bochecha: I now tried a “failsafe” boot and systemd gets stuck on an infinite Trying to load D-Bus loop. Oct 13 10:18:29 bochecha possibly dbus doesn't start, and systemd tries again, and again and... ? Well I'm afraid that as the guys on #systemd said, it's nigh on impossible to say what your problem actually is without more debugging etc. Then it may be a limitation of systemd's architecture. For example if you manage to get the boot to work by waiting patiently enough then the output from systemd-analyze might point to the cause. OK. I'll try it now. I suspect to be honest that it relates to some binfmt stuff or other modprobe setups that just ultimately stall things. How can I investigate that and fix that? It's obviously not a problem that affects other people other than yourself so you have to ask yourself why your system is different, try with a clean install and then add back your config until you find the issue. A clean install of what? There's no way I can “cleanly” install Mageia Cauldron because it doesn't have an installation medium. And I'm not overwriting my current system. No way, Jose. Regards, Shlomi Fish -- - Shlomi
Re: [Mageia-dev] mageia 2 and systemd
On 13/10/11 14:26, Michael Altizer wrote: On 10/13/2011 05:29 AM, Colin Guthrie wrote: 'Twas brillig, and Guillaume Rousse at 13/10/11 09:44 did gyre and gimble: Le 13/10/2011 10:08, philippe makowski a écrit : what do we do with %_post_service and %_preun_service ? Those macros will have to be enhanced to handle new behaviours, without having to modify packages. /usr/share/rpm-helper/add-service already has some systemd support what's still missing? Col The del-service helper has some support, but it gets mighty unhappy when you're not actually using systemd. :) # urpme ipsec-tools removing ipsec-tools-0.7.3-4.mga1.x86_64 process 8398: arguments to dbus_connection_send_with_reply_and_block() were incorrect, assertion connection != NULL failed in file dbus-connection.c line 3483. This is normally a bug in some application using the D-Bus library. D-Bus not built with -rdynamic so unable to print a backtrace /usr/share/rpm-helper/del-service: line 32: 8398 Aborted (core dumped) /bin/systemctl daemon-reload process 8431: arguments to dbus_connection_send_with_reply_and_block() were incorrect, assertion connection != NULL failed in file dbus-connection.c line 3483. This is normally a bug in some application using the D-Bus library. D-Bus not built with -rdynamic so unable to print a backtrace /usr/share/rpm-helper/del-service: line 32: 8431 Aborted (core dumped) /bin/systemctl daemon-reload Yes, I reported that here :- https://bugs.mageia.org/show_bug.cgi?id=2865 No response from anyone though :\
Re: [Mageia-dev] mageia 2 and systemd
2011/10/12 Thierry Vignaud thierry.vign...@gmail.com: Would you have tried to install Cauldron, you would know it's enabled by default. ;) ok so let me express this another way : are we taking the decision to have mandatory move from SysV initscript to systemd unit file for our packages that use SysV initscript ? any doc, recommandation on how we manage this migration in our spec ?
Re: [Mageia-dev] mageia 2 and systemd
philippe makowski skrev 13.10.2011 09:58: 2011/10/12 Thierry Vignaudthierry.vign...@gmail.com: Would you have tried to install Cauldron, you would know it's enabled by default. ;) ok so let me express this another way : are we taking the decision to have mandatory move from SysV initscript to systemd unit file for our packages that use SysV initscript ? any doc, recommandation on how we manage this migration in our spec ? No, it's not mandatory. (but yes we forced it for now on Cauldron to get more systemd testers) For Mageia 2 we will support both initscripts and systemd Mageia 3 will probably be systemd only. -- Thomas
Re: [Mageia-dev] mageia 2 and systemd
2011/10/13 Thomas Backlund t...@mageia.org:
No, it's not mandatory.
(but yes we forced it for now on Cauldron to get more systemd testers)
For Mageia 2 we will support both initscripts and systemd
ok
any howto available ?
where do we put systemd files ?
is %{_unitdir} the correct place ?
any doc, scriptlet on how to deal with intscripts removing and systemd
settings ?
what do we do with %_post_service and %_preun_service ?
is there in cauldron a spec that already implemented the move from
initscripts to systemd ?
Re: [Mageia-dev] mageia 2 and systemd
Le 13/10/2011 10:08, philippe makowski a écrit :
2011/10/13 Thomas Backlundt...@mageia.org:
No, it's not mandatory.
(but yes we forced it for now on Cauldron to get more systemd testers)
For Mageia 2 we will support both initscripts and systemd
ok
any howto available ?
no
where do we put systemd files ?
is %{_unitdir} the correct place ?
yes
any doc, scriptlet on how to deal with intscripts removing and systemd
settings ?
no
what do we do with %_post_service and %_preun_service ?
Those macros will have to be enhanced to handle new behaviours, without
having to modify packages.
is there in cauldron a spec that already implemented the move from
initscripts to systemd ?
Not really moved from initscripts, rather with additional systemd
support: apache, openssh, mysql, ntpd... BTW, the easiest way is to
look in fedora git repository:
pkgs.fedoraproject.org/gitweb
--
BOFH excuse #404:
Sysadmin accidentally destroyed pager with a large hammer.
Re: [Mageia-dev] mageia 2 and systemd
'Twas brillig, and Kira at 13/10/11 06:08 did gyre and gimble: 在 Thu, 13 Oct 2011 04:08:48 +0800, Thierry Vignaud thierry.vign...@gmail.com寫道: On 12 October 2011 21:00, philippe makowski makowski.mag...@gmail.com wrote: how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? Would you have tried to install Cauldron, you would know it's enabled by default. I know colin had explained before, but my cauldron still lost sound in the last update and I got no idea how to get it back. Any idea, guys? I looked into /etc/pam.d and renamed system-auth.rpmnew to system-auth, but after doing so I still got no sound. Then you need to check systemd-loginctl to see if your user is listed. If it's not then that is the primary problem. If you are not listed, then you need say which DM you use (e.g. gdm or kdm) and we can take it further. If you are listed, make sure you have permissions on the sound device nodes (getfacl /dev/snd/pcm*) Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] mageia 2 and systemd
what do we do with %_post_service and %_preun_service ? Those macros will have to be enhanced to handle new behaviours, without having to modify packages. ??? is there in cauldron a spec that already implemented the move from initscripts to systemd ? Not really moved from initscripts, rather with additional systemd support: apache, openssh, mysql, ntpd... I will take a look looking at openssh both scripts are provided ? I'm lost BTW, the easiest way is to look in fedora git repository: I know but I guess that Mageia have some rules that Fedora don't have and vice versa and about Fedora, for next 16 coming, they make the move https://bugzilla.redhat.com/show_bug.cgi?id=713562
Re: [Mageia-dev] mageia 2 and systemd
在 Thu, 13 Oct 2011 17:23:57 +0800, Colin Guthrie mag...@colin.guthr.ie寫道: Then you need to check systemd-loginctl to see if your user is listed. If it's not then that is the primary problem. If you are not listed, then you need say which DM you use (e.g. gdm or kdm) and we can take it further. I got nothing listed in the output of the command. I am using KDM without auto-login.
Re: [Mageia-dev] mageia 2 and systemd
'Twas brillig, and Guillaume Rousse at 13/10/11 09:44 did gyre and gimble: Le 13/10/2011 10:08, philippe makowski a écrit : what do we do with %_post_service and %_preun_service ? Those macros will have to be enhanced to handle new behaviours, without having to modify packages. /usr/share/rpm-helper/add-service already has some systemd support what's still missing? Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] mageia 2 and systemd
'Twas brillig, and Kira at 13/10/11 10:28 did gyre and gimble: 在 Thu, 13 Oct 2011 17:23:57 +0800, Colin Guthrie mag...@colin.guthr.ie寫道: Then you need to check systemd-loginctl to see if your user is listed. If it's not then that is the primary problem. If you are not listed, then you need say which DM you use (e.g. gdm or kdm) and we can take it further. I got nothing listed in the output of the command. I am using KDM without auto-login. Well that is the problem. That combination works for me (or did when I last tested it a while ago). It'll be a pam issue in some capacity. Make sure you have the systemd registration in the relevant pam.d file used (I presume /etc/pam.d/kdm these days?) Here, it has both the pam_loginuid.so and the system-auth include which are both needed. Make sure your system-auth has: -sessionoptional pam_systemd.so in it. Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] mageia 2 and systemd
Le 13/10/2011 11:29, Colin Guthrie a écrit : 'Twas brillig, and Guillaume Rousse at 13/10/11 09:44 did gyre and gimble: Le 13/10/2011 10:08, philippe makowski a écrit : what do we do with %_post_service and %_preun_service ? Those macros will have to be enhanced to handle new behaviours, without having to modify packages. /usr/share/rpm-helper/add-service already has some systemd support what's still missing? I wasn't aware. Looking at the code, I don't see any call to 'systemctl daemon-reload' in add-service, whereas it is present in del-service. Shouldn't we use a filetrigger instead to automatically manage this kind of issue, so as to not rely on the package ? Also, they are other kind of systemd units to handle. -- BOFH excuse #334: 50% of the manual is in .pdf readme files
Re: [Mageia-dev] mageia 2 and systemd
在 Thu, 13 Oct 2011 17:33:22 +0800, Colin Guthrie mag...@colin.guthr.ie寫道: Well that is the problem. That combination works for me (or did when I last tested it a while ago). It'll be a pam issue in some capacity. Make sure you have the systemd registration in the relevant pam.d file used (I presume /etc/pam.d/kdm these days?) Here, it has both the pam_loginuid.so and the system-auth include which are both needed. Make sure your system-auth has: -sessionoptional pam_systemd.so in it. Col Okay, it works now. Adding -session optional pam_system.so into system-auth works.
Re: [Mageia-dev] mageia 2 and systemd
'Twas brillig, and Kira at 13/10/11 10:50 did gyre and gimble: 在 Thu, 13 Oct 2011 17:33:22 +0800, Colin Guthrie mag...@colin.guthr.ie寫道: Well that is the problem. That combination works for me (or did when I last tested it a while ago). It'll be a pam issue in some capacity. Make sure you have the systemd registration in the relevant pam.d file used (I presume /etc/pam.d/kdm these days?) Here, it has both the pam_loginuid.so and the system-auth include which are both needed. Make sure your system-auth has: -sessionoptional pam_systemd.so in it. Col Okay, it works now. Adding -session optional pam_system.so into system-auth Hmm, OK, but the question is, where did it go? This should have been in the system-auth.rpmnew file... perhaps you purged it before without properly resolving it? Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
Re: [Mageia-dev] mageia 2 and systemd
On 10/13/2011 05:29 AM, Colin Guthrie wrote: 'Twas brillig, and Guillaume Rousse at 13/10/11 09:44 did gyre and gimble: Le 13/10/2011 10:08, philippe makowski a écrit : what do we do with %_post_service and %_preun_service ? Those macros will have to be enhanced to handle new behaviours, without having to modify packages. /usr/share/rpm-helper/add-service already has some systemd support what's still missing? Col The del-service helper has some support, but it gets mighty unhappy when you're not actually using systemd. :) # urpme ipsec-tools removing ipsec-tools-0.7.3-4.mga1.x86_64 process 8398: arguments to dbus_connection_send_with_reply_and_block() were incorrect, assertion connection != NULL failed in file dbus-connection.c line 3483. This is normally a bug in some application using the D-Bus library. D-Bus not built with -rdynamic so unable to print a backtrace /usr/share/rpm-helper/del-service: line 32: 8398 Aborted (core dumped) /bin/systemctl daemon-reload process 8431: arguments to dbus_connection_send_with_reply_and_block() were incorrect, assertion connection != NULL failed in file dbus-connection.c line 3483. This is normally a bug in some application using the D-Bus library. D-Bus not built with -rdynamic so unable to print a backtrace /usr/share/rpm-helper/del-service: line 32: 8431 Aborted (core dumped) /bin/systemctl daemon-reload
Re: [Mageia-dev] mageia 2 and systemd
在 Thu, 13 Oct 2011 21:20:59 +0800, Colin Guthrie mag...@colin.guthr.ie寫道: Hmm, OK, but the question is, where did it go? This should have been in the system-auth.rpmnew file... perhaps you purged it before without properly resolving it? Col I discovered that I foolishly modified system-auth.rpmnew, rather than system-auth in the first mail.
Re: [Mageia-dev] mageia 2 and systemd
'Twas brillig, and Kira at 13/10/11 17:29 did gyre and gimble: 在 Thu, 13 Oct 2011 21:20:59 +0800, Colin Guthrie mag...@colin.guthr.ie寫道: Hmm, OK, but the question is, where did it go? This should have been in the system-auth.rpmnew file... perhaps you purged it before without properly resolving it? Col I discovered that I foolishly modified system-auth.rpmnew, rather than system-auth in the first mail. Ahh :) Cool, that explains it. Col -- Colin Guthrie colin(at)mageia.org http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/
[Mageia-dev] mageia 2 and systemd
Hi, how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? some doc from Fedora for packaging : http://fedoraproject.org/wiki/Packaging:Guidelines:Systemd http://fedoraproject.org/wiki/Packaging:ScriptletSnippets#Systemd
Re: [Mageia-dev] mageia 2 and systemd
On 12 October 2011 21:00, philippe makowski makowski.mag...@gmail.com wrote: how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? Would you have tried to install Cauldron, you would know it's enabled by default.
Re: [Mageia-dev] mageia 2 and systemd
在 Thu, 13 Oct 2011 04:08:48 +0800, Thierry Vignaud thierry.vign...@gmail.com寫道: On 12 October 2011 21:00, philippe makowski makowski.mag...@gmail.com wrote: how and when will we make the move ? should we need to provide native systemd service files for Mageia 2 ? Would you have tried to install Cauldron, you would know it's enabled by default. I know colin had explained before, but my cauldron still lost sound in the last update and I got no idea how to get it back. Any idea, guys? I looked into /etc/pam.d and renamed system-auth.rpmnew to system-auth, but after doing so I still got no sound.
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
Le samedi 16 juillet 2011 à 16:37 +0200, Olivier Blin a écrit : Michael Scherer m...@zarb.org writes: People on Fedora list were quite reluctant to changes ( to say the least ), and I am pretty sure that someone will hit a unrelated corner case in the last minute. Rtp was not really fond of systemd on arm and embedded system either. Do you have more details about the issues on arm/embedded? How could it be worse than forking our dozens of shell scripts? It isn't (IMHO), it is better to have systemd (IMHO). But that was a attempt to make rtp speak on the list ( like speaking of his proposal about flags ) :) His point (IIRC) was that adding cgroup to the kernel would have a small overhead, and big overhead are made of lots of smaller one. And (maybe) that systemd take ressources all the time, while shell scripts only take them on boot. On the other hand, we do not really target specific ultra specific embedded system for now, so this point may be moot. -- Michael Scherer
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Fri, 15 Jul 2011, D.Morgan wrote: On Fri, Jul 15, 2011 at 3:23 PM, Colin Guthrie mag...@colin.guthr.ie wrote: 'Twas brillig, and Colin Guthrie at 15/07/11 11:01 did gyre and gimble: Now that it's compiled, I need to test it :D Well it boots. And I have network connections! I have a problem where bluetoothd does not start but I'll solve that one at some point (it was a problem when I last used systemd too!) The various issues can now be solved as we go! Col Now that we have latest systemd rpm, is it OK to enable the systemd switches in the other rpms ? udev, ... OK? it is necessary: I had to fix dbus, consolekit, and accountsservice to get a working system. But mediatomb now starts serving files from nfs on boot, so the new systemd seems to be an improvement. Christiaan
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
Michael Scherer m...@zarb.org writes: People on Fedora list were quite reluctant to changes ( to say the least ), and I am pretty sure that someone will hit a unrelated corner case in the last minute. Rtp was not really fond of systemd on arm and embedded system either. Do you have more details about the issues on arm/embedded? How could it be worse than forking our dozens of shell scripts? -- Olivier Blin - blino
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Thu, 14 Jul 2011, Eugeni Dodonov wrote: On Thu, Jul 14, 2011 at 20:08, Olivier Blin mag...@blino.org wrote: Option 2 seems good as well, but does it really have to uninstall sysvinit? Isn't it enough to put some alternatives symlinks for /sbin/init? This is actually a great idea! A perfect candidate for update-alternatives, thanks, I haven't thought on that! Alternatives is not the most reliable system so it should not be used for /sbin/init. You could make 2 conflicting packages that only contain a /sbin/init symlink and provide something which basesystem depends on. Christiaan
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
'Twas brillig, and Eugeni Dodonov at 14/07/11 23:40 did gyre and gimble: On Thu, Jul 14, 2011 at 18:26, Colin Guthrie mag...@colin.guthr.ie mailto:mag...@colin.guthr.ie wrote: Systemd 30 is out, with lots of nice changes, so I think we should use it now as we are quite early in the release cycle. It is working on my machine, but before doing something about it, I prefer to hear opinions :). Well, IMO (for what it matters) I'm massive for using it ASAP! Ok, systemd 30 is in svn, and (apparently) working, at least on my machine. However, as I will have a limited Internet connection between tomorrow and next week, I am not feeling that safe to send it to cooker. So if someone (colin?) would be able to do some additional testing with it before submitting, it would be great :). OK, I tried to compile it up locally, but it didn't build: checking for m4... /usr/bin/m4 configure: creating ./config.status config.status: creating Makefile config.status: creating po/Makefile.in config.status: creating config.h config.status: executing depfiles commands config.status: executing libtool commands config.status: executing po/stamp-it commands configure: WARNING: unrecognized options: --with-syslog-service systemd 30 Distribution:mageia SysV compatibility: yes SysV init scripts: /etc/rc.d/init.d SysV rc?.d directories: /etc/rc.d Gtk: yes libcryptsetup: no tcpwrap: yes PAM: yes AUDIT: yes SELinux: no ACL: yes binfmt: yes plymouth:yes prefix: /usr root dir: udev rules dir: /lib/udev/rules.d pam modules dir: /lib/../lib64/security dbus policy dir: /etc/dbus-1/system.d dbus session dir:/usr/share/dbus-1/services dbus system dir: /usr/share/dbus-1/services/../system-services dbus interfaces dir: /usr/share/dbus-1/services/../interfaces + make -j2 Makefile:7228: *** missing separator. Stop. error: Bad exit status from /home/colin/rpm/tmp/rpm-tmp.vn9NeJ (%build) RPM build errors: Bad exit status from /home/colin/rpm/tmp/rpm-tmp.vn9NeJ (%build) This appears to be the result of some tab/spaces mixup in the Makefile.am patch, so I've fixed that up. Next problem as a TARGET_MANDRIVA define rather than a TARGET_MAGEIA one (easily done), so I tweaked that too. Then there was a syntax error with an #ifndef line in service.c so fixed that up. Now that it's compiled, I need to test it :D (I've pushed the updated patch back to SVN, but you probably want to update your git checkout before regenerating and posting the patch upstream) Col -- Colin Guthrie mageia(at)colin.guthr.ie http://colin.guthr.ie/ Day Job: Tribalogic Limited [http://www.tribalogic.net/] Open Source: Mageia Contributor [http://www.mageia.org/] PulseAudio Hacker [http://www.pulseaudio.org/] Trac Hacker [http://trac.edgewall.org/]
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
'Twas brillig, and Colin Guthrie at 15/07/11 11:01 did gyre and gimble: Now that it's compiled, I need to test it :D Well it boots. And I have network connections! I have a problem where bluetoothd does not start but I'll solve that one at some point (it was a problem when I last used systemd too!) The various issues can now be solved as we go! Col -- Colin Guthrie mageia(at)colin.guthr.ie http://colin.guthr.ie/ Day Job: Tribalogic Limited [http://www.tribalogic.net/] Open Source: Mageia Contributor [http://www.mageia.org/] PulseAudio Hacker [http://www.pulseaudio.org/] Trac Hacker [http://trac.edgewall.org/]
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Fri, Jul 15, 2011 at 3:23 PM, Colin Guthrie mag...@colin.guthr.ie wrote: 'Twas brillig, and Colin Guthrie at 15/07/11 11:01 did gyre and gimble: Now that it's compiled, I need to test it :D Well it boots. And I have network connections! I have a problem where bluetoothd does not start but I'll solve that one at some point (it was a problem when I last used systemd too!) The various issues can now be solved as we go! Col Now that we have latest systemd rpm, is it OK to enable the systemd switches in the other rpms ? udev, ...
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
Le mercredi 13 juillet 2011 à 22:47 -0300, Eugeni Dodonov a écrit :
Almost finally, should the systemd files belong to the main package, the
same way as they do with initscripts-based one (e.g., the package would
provide /lib/systemd/system/%{name}.service together with
%_sysconfig/rc.d/init.d/%{name} for example), with no extra subpackages or
flags - or should all systemd-specific files go into %{name}-systemd package
for example? What do you think?
Everything in 1 package.
And finally, what does seems to be the best way of starting to use systemd
in cauldron? I have thought on 3 alternatives:
- easy way, only having it packaged, but not
providing/obsoleting/conflicting with sysvinit. This way, it will work when
kernel is booted with init=/bin/systemd (the least invasive way)
- compatible way (like in Mandriva) - it is available, systemd-sysvinit
conflicts with sysvinit, so if someone installs systemd-sysvinit, sysvinit
goes away and systemd is run by default. This seems to be the most sane way
to me (but I could be biased), and it is easiest one for testing
- ultimate way - systemd provides and obsoletes sysvinit and its goodies.
This way, systemd will be the only one (e.g., highlander style). This is how
fedora did it if I am not mistaken, but I am not sure if it the best way.
Option 2 for Mageia 2, and option 3 for Mageia 3.
People on Fedora list were quite reluctant to changes ( to say the least
), and I am pretty sure that someone will hit a unrelated corner case in
the last minute. Rtp was not really fond of systemd on arm and embedded
system either.
So we should keep the possibility of letting people choose for Mageia 2.
Also, systemd was reverted for f14, and while it was working fine for
me, I would rather make sure we have a reputation of being cautious and
choosing the quality rather than trying to steal the reputation of
Fedora.
--
Michael Scherer
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Wed, 13 Jul 2011, Eugeni Dodonov wrote: Firstly, systemdrequires udev = 172, what is the policy to update it? According to 'mgarepo maintdb get udev', it has no maintainers, does anyone objects if I grab/update it as well? Since you volunteered, could you take a look at: https://bugs.mageia.org/show_bug.cgi?id=1454 Hotplug works, coldplug doesn't, so I guess something goes wrong with udev. Updating packages to newer versions needs to be done and help with that is welcome, but we also need to fix reported problems. Unrelated: AFAIK you are registered as maintainer of both system-config-printer and hplip in mandriva linux, maybe you want to apply http://svnweb.mageia.org/packages/cauldron/hplip/current/SPECS/hplip.spec?r1=106333r2=106663 there to fix an old bug which mageia inherited. Christiaan
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Thu, Jul 14, 2011 at 06:25, Christiaan Welvaart c...@daneel.dyndns.orgwrote: On Wed, 13 Jul 2011, Eugeni Dodonov wrote: Firstly, systemdrequires udev = 172, what is the policy to update it? According to 'mgarepo maintdb get udev', it has no maintainers, does anyone objects if I grab/update it as well? Since you volunteered, could you take a look at: https://bugs.mageia.org/show_**bug.cgi?id=1454https://bugs.mageia.org/show_bug.cgi?id=1454 Hotplug works, coldplug doesn't, so I guess something goes wrong with udev. Updating packages to newer versions needs to be done and help with that is welcome, but we also need to fix reported problems. I think that updating udev could fix it as well, there were many different changes all around.. Unrelated: AFAIK you are registered as maintainer of both system-config-printer and hplip in mandriva linux, maybe you want to apply http://svnweb.mageia.org/**packages/cauldron/hplip/** current/SPECS/hplip.spec?r1=**106333r2=106663http://svnweb.mageia.org/packages/cauldron/hplip/current/SPECS/hplip.spec?r1=106333r2=106663there to fix an old bug which mageia inherited. Nope, I am not cups nor hplip maintainer, I had to give up on them because I don't have any printers around to do any testing at all :(. -- Eugeni Dodonov http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
'Twas brillig, and Eugeni Dodonov at 14/07/11 02:47 did gyre and gimble:
On Tue, Jul 12, 2011 at 09:48, Colin Guthrie mag...@colin.guthr.ie
mailto:mag...@colin.guthr.ie wrote:
'Twas brillig, and Eugeni Dodonov at 12/07/11 13:15 did gyre and gimble:
If nobody objects, I could help with that. Mandriva certainly gave a
large experience on how to integrate systemd into the system without
killing traditional sysvinit alternative.
It would also be extremely interested to have native systemd services
which use most of systemd features (like sound and alsa scripts, which
we discussed with Colin and Andrey Borzenkov some months ago but never
got to implement properly).
Massive +1 for systemd and massive +1 Eugeni wanting to help out! \o/
I'll try and help out in bits and bobs too, tho' time is always a
problem!
Ok, some n00b questions arise from my part, sorry if they seem too basic
- I am only catching up with mga style of development :).
Systemd 30 is out, with lots of nice changes, so I think we should use
it now as we are quite early in the release cycle. It is working on my
machine, but before doing something about it, I prefer to hear opinions :).
Well, IMO (for what it matters) I'm massive for using it ASAP!
Firstly, systemdrequires udev = 172, what is the policy to update it?
According to 'mgarepo maintdb get udev', it has no maintainers, does
anyone objects if I grab/update it as well?
I think that would be most welcome!
Secondly, what should be the correct way of supporting systemd in a
package? In Mandriva, I thought on adding a --with flag to
enable/disable systemd, but in most cases it does (almost) nothing. All
services which want to support systemd only need to place their files
into /lib/systemd - and that's it. Should we support opting-out of
systemd in specs? I believe fcrozat is having the same dilemma in SuSE
now as well, and he settled on some common packaging macros.
IMO, unless something goes horribly wrong with systemd (which I very
seriously doubt will happen - it's actually been amazing to watch this
project from birth to wide adoption) I'd say we just jump right in and
package the units unconditionally. They don't really take up much space
and do nothing if you choose to not use it for the time being. I think
by mga3 we'll likely *only* offer systemd (if not in mga2), so I'd say
putting in effort to opt out now is pretty much not going to be worth it.
If anyone else feels super strongly to the contrary, speak up now or
forever hold your peace :D
Almost finally, should the systemd files belong to the main package, the
same way as they do with initscripts-based one (e.g., the package would
provide /lib/systemd/system/%{name}.service together with
%_sysconfig/rc.d/init.d/%{name} for example), with no extra subpackages
or flags - or should all systemd-specific files go into %{name}-systemd
package for example? What do you think?
I vote to put them in the same place as the initscripts-based ones. As I
predicted above, I reckon it won't take too long for systemd to be the
only system offered and obsoleting all those subpackages would be a bit
of a PITA down the line. And if systemd isn't used those files do
nothing, so no harm done.
And finally, what does seems to be the best way of starting to use
systemd in cauldron? I have thought on 3 alternatives:
- easy way, only having it packaged, but not
providing/obsoleting/conflicting with sysvinit. This way, it will work
when kernel is booted with init=/bin/systemd (the least invasive way)
- compatible way (like in Mandriva) - it is available, systemd-sysvinit
conflicts with sysvinit, so if someone installs systemd-sysvinit,
sysvinit goes away and systemd is run by default. This seems to be the
most sane way to me (but I could be biased), and it is easiest one for
testing
- ultimate way - systemd provides and obsoletes sysvinit and its
goodies. This way, systemd will be the only one (e.g., highlander
style). This is how fedora did it if I am not mistaken, but I am not
sure if it the best way.
So, that's it for now from my part..
I'd like option 2 please! :) It doesn't exclude using option 1 along
with it (I was doing this for a while), so it's most flexible.
If all goes well for those testing it, we can vote on whether we jump in
with both feet :)
That's my take on it anyway!
:)
Col
--
Colin Guthrie
mageia(at)colin.guthr.ie
http://colin.guthr.ie/
Day Job:
Tribalogic Limited [http://www.tribalogic.net/]
Open Source:
Mageia Contributor [http://www.mageia.org/]
PulseAudio Hacker [http://www.pulseaudio.org/]
Trac Hacker [http://trac.edgewall.org/]
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Thu, Jul 14, 2011 at 18:26, Colin Guthrie mag...@colin.guthr.ie wrote: Systemd 30 is out, with lots of nice changes, so I think we should use it now as we are quite early in the release cycle. It is working on my machine, but before doing something about it, I prefer to hear opinions :). Well, IMO (for what it matters) I'm massive for using it ASAP! Ok, systemd 30 is in svn, and (apparently) working, at least on my machine. However, as I will have a limited Internet connection between tomorrow and next week, I am not feeling that safe to send it to cooker. So if someone (colin?) would be able to do some additional testing with it before submitting, it would be great :). -- Eugeni Dodonov http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Thu, 14 Jul 2011 22:26:37 +0100, Colin Guthrie mag...@colin.guthr.ie wrote: And finally, what does seems to be the best way of starting to use systemd in cauldron? I have thought on 3 alternatives: - easy way, only having it packaged, but not providing/obsoleting/conflicting with sysvinit. This way, it will work when kernel is booted with init=/bin/systemd (the least invasive way) - compatible way (like in Mandriva) - it is available, systemd-sysvinit conflicts with sysvinit, so if someone installs systemd-sysvinit, sysvinit goes away and systemd is run by default. This seems to be the most sane way to me (but I could be biased), and it is easiest one for testing - ultimate way - systemd provides and obsoletes sysvinit and its goodies. This way, systemd will be the only one (e.g., highlander style). This is how fedora did it if I am not mistaken, but I am not sure if it the best way. So, that's it for now from my part.. I'd like option 2 please! :) It doesn't exclude using option 1 along with it (I was doing this for a while), so it's most flexible. After my tries with mandriva, I vote for the highlander way (option 3). Systemd was almost working (for me, I suppose that with some more questions I could have got it working fine), but I always had problems with the mix of init and systemd scripts. You could disable something with chkconfig, but systemd still lanunched it, you never knew if you had to disable something in one place or another, and anyways, it looks like systemd is not really thought to disable services. So go for all and go systemd only, so problems will be discovered early and fixed early. As a side note, this reminds me of the mix of ifplugd and networkmanager. This was a point I wanted to throw in MGA2 specs: migrate completely to networkmanager and ditch net_applet and traditional network scripts. NetworkManger is the only sane way where I can have 5 wifi setups and system chooses the good one automagically, and connects on boot. Do you think this is worth a separate thread, or there is no point in discussion ? TIA -- J.A. Magallon jamagallon()ono!com \ Winter is coming...
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
2011/7/14 JA Magallón jamagal...@ono.com As a side note, this reminds me of the mix of ifplugd and networkmanager. This was a point I wanted to throw in MGA2 specs: migrate completely to networkmanager and ditch net_applet and traditional network scripts. NetworkManger is the only sane way where I can have 5 wifi setups and system chooses the good one automagically, and connects on boot. Do you think this is worth a separate thread, or there is no point in discussion ? I am all up for networkmanager since 2009, but let's hear Blino's opinion :). -- Eugeni Dodonov http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
Eugeni Dodonov eug...@dodonov.net writes: 2011/7/14 JA Magallón jamagal...@ono.com As a side note, this reminds me of the mix of ifplugd and networkmanager. This was a point I wanted to throw in MGA2 specs: migrate completely to networkmanager and ditch net_applet and traditional network scripts. NetworkManger is the only sane way where I can have 5 wifi setups and system chooses the good one automagically, and connects on boot. Do you think this is worth a separate thread, or there is no point in discussion ? I am all up for networkmanager since 2009, but let's hear Blino's opinion :). We can probably keep some compatibility (like the sysvinit/systemd duality). What is the issue with ifplugd and NM? If NM is used for an interface, ifplugd should not be started for the same interface. What's actually happening? If you have 5 wifi networks configured, wpa_supplicant is normally able to automatically select the right one at boot, and it used to work for me. What happens in your case? -- Olivier Blin - blino
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Thu, Jul 14, 2011 at 20:04, Olivier Blin mag...@blino.org wrote: Eugeni Dodonov eug...@dodonov.net writes: 2011/7/14 JA Magallón jamagal...@ono.com As a side note, this reminds me of the mix of ifplugd and networkmanager. This was a point I wanted to throw in MGA2 specs: migrate completely to networkmanager and ditch net_applet and traditional network scripts. NetworkManger is the only sane way where I can have 5 wifi setups and system chooses the good one automagically, and connects on boot. Do you think this is worth a separate thread, or there is no point in discussion ? I am all up for networkmanager since 2009, but let's hear Blino's opinion :). We can probably keep some compatibility (like the sysvinit/systemd duality). This is how they work now, except that in mageia networkmanager uses keyfile plugin instead of networkmanager-mdv (which is a bit outdated at the moment) so the network configuration is different among them. -- Eugeni Dodonov http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
Colin Guthrie mag...@colin.guthr.ie writes:
Secondly, what should be the correct way of supporting systemd in a
package? In Mandriva, I thought on adding a --with flag to
enable/disable systemd, but in most cases it does (almost) nothing. All
services which want to support systemd only need to place their files
into /lib/systemd - and that's it. Should we support opting-out of
systemd in specs? I believe fcrozat is having the same dilemma in SuSE
now as well, and he settled on some common packaging macros.
IMO, unless something goes horribly wrong with systemd (which I very
seriously doubt will happen - it's actually been amazing to watch this
project from birth to wide adoption) I'd say we just jump right in and
package the units unconditionally. They don't really take up much space
and do nothing if you choose to not use it for the time being. I think
by mga3 we'll likely *only* offer systemd (if not in mga2), so I'd say
putting in effort to opt out now is pretty much not going to be worth it.
I agree, it does not seem useful to avoid shipping systemd services.
Though, when systemd is enabled, what happens if one has both systemd +
sysvinit services files installed for a given package?
Will the initscript be run in addition to the systemd service?
Almost finally, should the systemd files belong to the main package, the
same way as they do with initscripts-based one (e.g., the package would
provide /lib/systemd/system/%{name}.service together with
%_sysconfig/rc.d/init.d/%{name} for example), with no extra subpackages
or flags - or should all systemd-specific files go into %{name}-systemd
package for example? What do you think?
I vote to put them in the same place as the initscripts-based ones. As I
predicted above, I reckon it won't take too long for systemd to be the
only system offered and obsoleting all those subpackages would be a bit
of a PITA down the line. And if systemd isn't used those files do
nothing, so no harm done.
Agreed as well.
And finally, what does seems to be the best way of starting to use
systemd in cauldron? I have thought on 3 alternatives:
- easy way, only having it packaged, but not
providing/obsoleting/conflicting with sysvinit. This way, it will work
when kernel is booted with init=/bin/systemd (the least invasive way)
- compatible way (like in Mandriva) - it is available, systemd-sysvinit
conflicts with sysvinit, so if someone installs systemd-sysvinit,
sysvinit goes away and systemd is run by default. This seems to be the
most sane way to me (but I could be biased), and it is easiest one for
testing
- ultimate way - systemd provides and obsoletes sysvinit and its
goodies. This way, systemd will be the only one (e.g., highlander
style). This is how fedora did it if I am not mistaken, but I am not
sure if it the best way.
So, that's it for now from my part..
I'd like option 2 please! :) It doesn't exclude using option 1 along
with it (I was doing this for a while), so it's most flexible.
Option 2 seems good as well, but does it really have to uninstall
sysvinit? Isn't it enough to put some alternatives symlinks for
/sbin/init?
--
Olivier Blin - blino
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Thu, Jul 14, 2011 at 20:08, Olivier Blin mag...@blino.org wrote: Option 2 seems good as well, but does it really have to uninstall sysvinit? Isn't it enough to put some alternatives symlinks for /sbin/init? This is actually a great idea! A perfect candidate for update-alternatives, thanks, I haven't thought on that! -- Eugeni Dodonov http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Thu, Jul 14, 2011 at 20:37, Balcaen John mik...@mageia.org wrote: On Thursday 14 July 2011 20:08:17 Eugeni Dodonov wrote: [...] This is how they work now, except that in mageia networkmanager uses keyfile plugin instead of networkmanager-mdv (which is a bit outdated at the moment) so the network configuration is different among them. Well for cauldron nm-mdv is not available at all because nm 0.9 is quite different than 0.8. From a fast look i guess it should not be too difficult to adapt your patch starting from the fedora/redhat plugin . Regarding a total switch to nm, if i'm not wrong the most problematic part is the installer integration. Maybe it's possible to adapt our draktools to edit/drop files directly in /etc/NetworkManager/system-connections/ if we're sticking with the native plugin (keyfile). I actually vote for keyfile plugin, because it is most stable and upstream. NM-mdv would be the best solution if it would be incorporated upstream, but it still needs lots of work to get there. One big issue with NM only by default is the installation via boot.iso and similar methods, and also overall installation via network - while it is certainly possible to configure networkmanager from the console, it is by no means easy. This was not the case for Mandriva desktop, but I believe that it is one nice feature to have in Mageia. -- Eugeni Dodonov http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Tue, Jul 12, 2011 at 09:48, Colin Guthrie mag...@colin.guthr.ie wrote:
'Twas brillig, and Eugeni Dodonov at 12/07/11 13:15 did gyre and gimble:
If nobody objects, I could help with that. Mandriva certainly gave a
large experience on how to integrate systemd into the system without
killing traditional sysvinit alternative.
It would also be extremely interested to have native systemd services
which use most of systemd features (like sound and alsa scripts, which
we discussed with Colin and Andrey Borzenkov some months ago but never
got to implement properly).
Massive +1 for systemd and massive +1 Eugeni wanting to help out! \o/
I'll try and help out in bits and bobs too, tho' time is always a problem!
Ok, some n00b questions arise from my part, sorry if they seem too basic - I
am only catching up with mga style of development :).
Systemd 30 is out, with lots of nice changes, so I think we should use it
now as we are quite early in the release cycle. It is working on my machine,
but before doing something about it, I prefer to hear opinions :).
Firstly, systemdrequires udev = 172, what is the policy to update it?
According to 'mgarepo maintdb get udev', it has no maintainers, does anyone
objects if I grab/update it as well?
Secondly, what should be the correct way of supporting systemd in a package?
In Mandriva, I thought on adding a --with flag to enable/disable systemd,
but in most cases it does (almost) nothing. All services which want to
support systemd only need to place their files into /lib/systemd - and
that's it. Should we support opting-out of systemd in specs? I believe
fcrozat is having the same dilemma in SuSE now as well, and he settled on
some common packaging macros.
Almost finally, should the systemd files belong to the main package, the
same way as they do with initscripts-based one (e.g., the package would
provide /lib/systemd/system/%{name}.service together with
%_sysconfig/rc.d/init.d/%{name} for example), with no extra subpackages or
flags - or should all systemd-specific files go into %{name}-systemd package
for example? What do you think?
And finally, what does seems to be the best way of starting to use systemd
in cauldron? I have thought on 3 alternatives:
- easy way, only having it packaged, but not
providing/obsoleting/conflicting with sysvinit. This way, it will work when
kernel is booted with init=/bin/systemd (the least invasive way)
- compatible way (like in Mandriva) - it is available, systemd-sysvinit
conflicts with sysvinit, so if someone installs systemd-sysvinit, sysvinit
goes away and systemd is run by default. This seems to be the most sane way
to me (but I could be biased), and it is easiest one for testing
- ultimate way - systemd provides and obsoletes sysvinit and its goodies.
This way, systemd will be the only one (e.g., highlander style). This is how
fedora did it if I am not mistaken, but I am not sure if it the best way.
So, that's it for now from my part..
Opinions?
--
Eugeni Dodonov
http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Wed, Jul 13, 2011 at 22:47, Eugeni Dodonov eug...@dodonov.net wrote: Firstly, systemdrequires udev = 172, what is the policy to update it? According to 'mgarepo maintdb get udev', it has no maintainers, does anyone objects if I grab/update it as well? (I forgot to mention that it is already up and running fine on my machine). -- Eugeni Dodonov http://eugeni.dodonov.net/
[Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
Hi there Following discussion about technical specifications, here is an interesting item: systemd Bug report to follow discussion and history: https://bugs.mageia.org/show_bug.cgi?id=2120 We need points of view, the way we could integrate it if we decide to go, planning (I guess it would need to start now to be full ready for final release) Comments, proposals, shout ? Cheers -- Anne http://www.mageia.org
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Tue, Jul 12, 2011 at 09:11, Anne nicolas enn...@mageia.org wrote: Hi there Following discussion about technical specifications, here is an interesting item: systemd Bug report to follow discussion and history: https://bugs.mageia.org/show_bug.cgi?id=2120 We need points of view, the way we could integrate it if we decide to go, planning (I guess it would need to start now to be full ready for final release) Comments, proposals, shout ? If nobody objects, I could help with that. Mandriva certainly gave a large experience on how to integrate systemd into the system without killing traditional sysvinit alternative. It would also be extremely interested to have native systemd services which use most of systemd features (like sound and alsa scripts, which we discussed with Colin and Andrey Borzenkov some months ago but never got to implement properly). -- Eugeni Dodonov http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Tue, Jul 12, 2011 at 2:15 PM, Eugeni Dodonov eug...@dodonov.net wrote: On Tue, Jul 12, 2011 at 09:11, Anne nicolas enn...@mageia.org wrote: Hi there Following discussion about technical specifications, here is an interesting item: systemd Bug report to follow discussion and history: https://bugs.mageia.org/show_bug.cgi?id=2120 We need points of view, the way we could integrate it if we decide to go, planning (I guess it would need to start now to be full ready for final release) Comments, proposals, shout ? If nobody objects, I could help with that. Mandriva certainly gave a large experience on how to integrate systemd into the system without killing traditional sysvinit alternative. It would also be extremely interested to have native systemd services which use most of systemd features (like sound and alsa scripts, which we discussed with Colin and Andrey Borzenkov some months ago but never got to implement properly). -- Eugeni Dodonov http://eugeni.dodonov.net/ If you can soon that would be awesome that way we will have plenty of time to test before our first alpha ( and because we will have to rebuild packages with systemd switch to on thanks for you help
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Tuesday 12 July 2011 09:15:18 Eugeni Dodonov wrote: [...] Comments, proposals, shout ? If nobody objects, I could help with that. Mandriva certainly gave a large experience on how to integrate systemd into the system without killing traditional sysvinit alternative. I don't have any objections ;o) Also since you work alot on it , why do we have some systemd files in initscripts ? (because i noticed that thoses files were moved back to systemd package) -- Balcaen John
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
On Tue, Jul 12, 2011 at 09:38, Balcaen John mik...@mageia.org wrote: On Tuesday 12 July 2011 09:15:18 Eugeni Dodonov wrote: [...] Comments, proposals, shout ? If nobody objects, I could help with that. Mandriva certainly gave a large experience on how to integrate systemd into the system without killing traditional sysvinit alternative. I don't have any objections ;o) Also since you work alot on it , why do we have some systemd files in initscripts ? (because i noticed that thoses files were moved back to systemd package) Each package is free to provide its .service files for systemd, they are not tied to any development libraries or headers and such. In case of initscripts, there are (were?) many mandriva-specifics tricks which used to run on startup via rc.sysinit and similar approaches, so initscripts package looked as the best place to put them at the time I think. -- Eugeni Dodonov http://eugeni.dodonov.net/
Re: [Mageia-dev] [Mageia 2 specifications] Systemd or not systemd
'Twas brillig, and Eugeni Dodonov at 12/07/11 13:15 did gyre and gimble: If nobody objects, I could help with that. Mandriva certainly gave a large experience on how to integrate systemd into the system without killing traditional sysvinit alternative. It would also be extremely interested to have native systemd services which use most of systemd features (like sound and alsa scripts, which we discussed with Colin and Andrey Borzenkov some months ago but never got to implement properly). Massive +1 for systemd and massive +1 Eugeni wanting to help out! \o/ I'll try and help out in bits and bobs too, tho' time is always a problem! Col -- Colin Guthrie mageia(at)colin.guthr.ie http://colin.guthr.ie/ Day Job: Tribalogic Limited [http://www.tribalogic.net/] Open Source: Mageia Contributor [http://www.mageia.org/] PulseAudio Hacker [http://www.pulseaudio.org/] Trac Hacker [http://trac.edgewall.org/]
