[opensuse-factory] [RFH] Anyone wiling to give systemd v232 a try ?
Hey, Newest version of systemd v232 has been integrated and is mostly ready to land in Factory. It can be found here: https://build.opensuse.org/package/show/home:fbui:systemd:next:openSUSE-Fact... We already gave it some testing internally and it also ran over the weekend on my own laptop (which is still alive): so far, so good :) However my setup is quite boring without any fancy configs (LVM, DM, encryption, ...). So in case some of you want to give it a shot, you're more than welcome. However please note that it's still "experimental" material (or rather call it a release candidate, which is probably a more appropriate wording) so make sure to test it on a testing/spare system only... A couple things that might be worth to be noted: - Make sure to use the latest version of dracut first. It should include the following change: * Tue Nov 08 2016 daniel@molkentin.de - systemd-initrd: Add initrd-root-device.target. Cherry-pick to get systemd v230 into factory (bsc1009089) * add 0451-systemd-initrd-add-initrd-root-device.target.patch - For those who rely on wicked: systemd is now making sure that local users are logged out before the network is shut down when the system goes down (commit 8c856804780681e). The consequence of this is that user can log in only after the network.target is activated. But for some reasons, wicked takes a while to do so (~15 secs), so you'll see a long delay before being able to log in. For now I don't know why wicked takes so long time, maybe that was already reported or discussed in the past... - The service that flushes the journal during boot looks to be stucked when the debug logs are enabled. I'll try to debug this but that shouldn't affect you for regular boot. - Here I could have noted a faster boot: 2 secs less spents in userspace :) Thanks ! -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 11/28/2016 09:00 AM, Franck Bui wrote:
But for some reasons, wicked takes a while to do so (~15 secs), so you'll see a long delay before being able to log in. This could be due to IPv4 Address Conflict Detection ( RFC 5227 [1]), something that wicked adheres to as I learned (It seems to annoy folks that expect fast boot). If that still doesn't account for the full time, my guess would be that a slow DHCP server.
Cheers, Daniel [1] https://www.ietf.org/rfc/rfc5227.txt -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Newest version of systemd v232 has been integrated and is mostly ready to land in Factory. It can be found here:
https://build.opensuse.org/package/show/home:fbui:systemd:next:openSUSE-Fact...
We already gave it some testing internally and it also ran over the weekend on my own laptop (which is still alive): so far, so good :)
Can we please make sure that this fix[1] is backported? v232 broke Docker and runC quite badly. [1]: https://github.com/systemd/systemd/pull/4628 -- Aleksa Sarai Software Engineer (Containers) SUSE Linux GmbH https://www.cyphar.com/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi, On 11/28/2016 10:13 AM, Aleksa Sarai wrote:
Newest version of systemd v232 has been integrated and is mostly ready to land in Factory. It can be found here:
https://build.opensuse.org/package/show/home:fbui:systemd:next:openSUSE-Fact...
We already gave it some testing internally and it also ran over the weekend on my own laptop (which is still alive): so far, so good :)
Can we please make sure that this fix[1] is backported? v232 broke Docker and runC quite badly.
It's already included so we should be fine (although it would be nice to give it a test). Thanks for the feedback. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 28 November 2016 at 16:00, Franck Bui <fbui@suse.de> wrote:
But for some reasons, wicked takes a while to do so (~15 secs), so you'll see a long delay before being able to log in.
For now I don't know why wicked takes so long time, maybe that was already reported or discussed in the past...
This isn't the issue that's been around for a while related to wicked and IPv6 right? I use NM but I think I've heard on the forums that wicked will slow down the boot if it can't find an IPv6 connection until it times out. - Karl Cheng (Qantas94Heavy) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Mon, Nov 28, 2016 at 12:30 PM, Karl Cheng <qantas94heavy@gmail.com> wrote:
On 28 November 2016 at 16:00, Franck Bui <fbui@suse.de> wrote:
But for some reasons, wicked takes a while to do so (~15 secs), so you'll see a long delay before being able to log in.
For now I don't know why wicked takes so long time, maybe that was already reported or discussed in the past...
This isn't the issue that's been around for a while related to wicked and IPv6 right? I use NM but I think I've heard on the forums that wicked will slow down the boot if it can't find an IPv6 connection until it times out.
The problem is not so much why it delays, as that there is no way to avoid this delay - there is no optional service wicked-wait-online like there is for NetworkManager or systemd-networkd. So everyone has to wait, even if I absolutely do not care about network availability at this point. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 11/28/2016 10:35 AM, Andrei Borzenkov wrote:
The problem is not so much why it delays, as that there is no way to avoid this delay - there is no optional service wicked-wait-online like there is for NetworkManager or systemd-networkd.
Correct. I talked to wicked maintainer and this should be fixed soon. Thanks. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 11/28/2016 10:35 AM, Andrei Borzenkov wrote:
The problem is not so much why it delays, as that there is no way to avoid this delay - there is no optional service wicked-wait-online like there is for NetworkManager or systemd-networkd.
Correct. I talked to wicked maintainer and this should be fixed soon. Thanks. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Franck Bui wrote:
- For those who rely on wicked: systemd is now making sure that local users are logged out before the network is shut down when the system goes down (commit 8c856804780681e). The consequence of this is that user can log in only after the network.target is activated.
But for some reasons, wicked takes a while to do so (~15 secs), so you'll see a long delay before being able to log in.
https://bugzilla.suse.com/show_bug.cgi?id=972471 "Slow boot: 33s waiting for wicked interfaces to start". Also, I think dhcp for ipv4 and ipv6 happen sequentially, adding more delay. -- Per Jessen, Zürich (2.6°C) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Monday, November 28, 2016 9:00:32 AM PST Franck Bui wrote:
Hey,
Newest version of systemd v232 has been integrated and is mostly ready to land in Factory. It can be found here:
https://build.opensuse.org/package/show/home:fbui:systemd:next:openSUSE-Fact ory/systemd
We already gave it some testing internally and it also ran over the weekend on my own laptop (which is still alive): so far, so good :)
However my setup is quite boring without any fancy configs (LVM, DM, encryption, ...). So in case some of you want to give it a shot, you're more than welcome.
However please note that it's still "experimental" material (or rather call it a release candidate, which is probably a more appropriate wording) so make sure to test it on a testing/spare system only...
A couple things that might be worth to be noted:
- Make sure to use the latest version of dracut first. It should include the following change:
* Tue Nov 08 2016 daniel@molkentin.de - systemd-initrd: Add initrd-root-device.target. Cherry-pick to get systemd v230 into factory (bsc1009089) * add 0451-systemd-initrd-add-initrd-root-device.target.patch
- For those who rely on wicked: systemd is now making sure that local users are logged out before the network is shut down when the system goes down (commit 8c856804780681e). The consequence of this is that user can log in only after the network.target is activated.
But for some reasons, wicked takes a while to do so (~15 secs), so you'll see a long delay before being able to log in.
For now I don't know why wicked takes so long time, maybe that was already reported or discussed in the past...
- The service that flushes the journal during boot looks to be stucked when the debug logs are enabled. I'll try to debug this but that shouldn't affect you for regular boot.
- Here I could have noted a faster boot: 2 secs less spents in userspace :)
Thanks !
https://git.io/v1qWS <- i took the systemd-232 challenge, i am alive, thnks System: Host: linux-s4s1 Kernel: 4.8.10-1-default x86_64 (64 bit gcc: 6.2.1) Desktop: KDE Plasma 5.8.4 (Qt 5.7.0) dm: sddm,sddm Distro: openSUSE Tumbleweed -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 11/29/2016 12:01 PM, emanuel wrote:
https://git.io/v1qWS <- i took the systemd-232 challenge, i am alive, thnks System: Host: linux-s4s1 Kernel: 4.8.10-1-default x86_64 (64 bit gcc: 6.2.1) Desktop: KDE Plasma 5.8.4 (Qt 5.7.0) dm: sddm,sddm Distro: openSUSE Tumbleweed
Thanks for your time ! -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 28 November 2016 at 16:00, Franck Bui <fbui@suse.de> wrote:
However my setup is quite boring without any fancy configs (LVM, DM, encryption, ...). So in case some of you want to give it a shot, you're more than welcome.
I've tested this on my own laptop and it seems pretty much fine to me. During installation, I did have some issues with a file conflict:
Detected 1 file conflict: File /usr/lib/systemd/libsystemd-shared-232.so from install of systemd-232.suse.20161124.gabef684-89.1.x86_64 (systemd) conflicts with file from install of systemd-32bit-232.suse.20161124.gabef684-89.1.x86_64 (systemd)
File conflicts happen when two packages attempt to install files with the same name but different contents. If you continue, conflicting files will be replaced losing the previous content. Continue? [yes/no] (no):
I typed "yes" in here, then realised that this was installing the 32-bit version when errors appeared during install about wrong ELF type. After that, I performed "zypper in -f systemd", then restarted the computer to test. System seemed to boot up without issue, but when checking journalctl it seems that postfix fails to load and xembedsniproxy dumps core. I don't know whether they are related to the systemd upgrade or not though. Thanks for the work! - Karl Cheng (Qantas94Heavy) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Am 2016-11-29 um 12:35 schrieb Karl Cheng:
On 28 November 2016 at 16:00, Franck Bui <fbui@suse.de> wrote:
During installation, I did have some issues with a file conflict:
Detected 1 file conflict: File /usr/lib/systemd/libsystemd-shared-232.so from install of systemd-232.suse.20161124.gabef684-89.1.x86_64 (systemd) conflicts with file from install of systemd-32bit-232.suse.20161124.gabef684-89.1.x86_64 (systemd)
File conflicts happen when two packages attempt to install files with the same name but different contents. If you continue, conflicting files will be replaced losing the previous content. Continue? [yes/no] (no):
i had the same conflict. it seems that the package "systemd" should contain a file /usr/lib64/systemd/libsystemd-shared-232.so and package "systemd-32bit" should contain /usr/lib/systemd/libsystemd-shared-232.so. reverting back to 228. -- Best Regards | Freundliche Grüße | Cordialement | Cordiali Saluti | *DI Rainer Klier* Research & Development, Technical Sales Consultant Namirial GmbH -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi, On 11/29/2016 12:35 PM, Karl Cheng wrote:
During installation, I did have some issues with a file conflict:
Detected 1 file conflict: File /usr/lib/systemd/libsystemd-shared-232.so from install of systemd-232.suse.20161124.gabef684-89.1.x86_64 (systemd) conflicts with file from install of systemd-32bit-232.suse.20161124.gabef684-89.1.x86_64 (systemd)
File conflicts happen when two packages attempt to install files with the same name but different contents. If you continue, conflicting files will be replaced losing the previous content. Continue? [yes/no] (no):
I typed "yes" in here, then realised that this was installing the 32-bit version when errors appeared during install about wrong ELF type.
I'll fix this up thanks for spotting ! Just out of curiosity why are both systemd and systemd-32bit installed on your system ?
System seemed to boot up without issue, but when checking journalctl it seems that postfix fails to load and xembedsniproxy dumps core. I don't know whether they are related to the systemd upgrade or not though.
I assume that those issues were not present before you upgraded systemd, right ? In this case could you provide the journal logs with debug logs enabled ?
Thanks for the work!
Thanks to you for taking time to test the beast ;) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 29 November 2016 at 21:39, Franck Bui <franck.bui@gmail.com> wrote:
Just out of curiosity why are both systemd and systemd-32bit installed on your system ?
I think that's how it is by default(?!), all I did was perform "zypper dup --from systemd_repo". I loaded up a pretty much fresh Leap 42.2 VM and it also has both systemd and systemd-32bit installed for some reason.
I assume that those issues were not present before you upgraded systemd, right ?
In this case could you provide the journal logs with debug logs enabled ?
I'm actually not sure, I haven't really had any reason to check journalctl in a while. I'll look back into the logs if I can and see if it's been a persistent issue or not. - Karl Cheng (Qantas94Heavy) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Tuesday 2016-11-29 15:01, Karl Cheng wrote:
On 29 November 2016 at 21:39, Franck Bui <franck.bui@gmail.com> wrote:
Just out of curiosity why are both systemd and systemd-32bit installed on your system ?
I think that's how it is by default(?!), all I did was perform "zypper dup --from systemd_repo".
I loaded up a pretty much fresh Leap 42.2 VM and it also has both systemd and systemd-32bit installed for some reason.
That "some reason" is, again, Recommends. patterns-openSUSE/patterns-openSUSE.spec:Recommends: systemd-32bit patterns-openSUSE/patterns-openSUSE.spec:Recommends: systemd-32bit patterns-openSUSE/patterns-openSUSE.spec:Recommends: systemd-32bit -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Tuesday, November 29, 2016 3:07:51 PM PST Jan Engelhardt wrote:
patterns-openSUSE/patterns-openSUSE.spec:Recommends: systemd-32bit
I would not have noticed systemd-32bit if i had not tried to install systemd-232 , an error occurred , since then fixed, still i am left to wonder what is the purpose of systemd-32bit in a 64bit install? -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Nov 30 2016, emanuel <emanuel.castelo@gmail.com> wrote:
what is the purpose of systemd-32bit in a 64bit install?
It contains the PAM module. Andreas. -- Andreas Schwab, SUSE Labs, schwab@suse.de GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7 "And now for something completely different." -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Hi! Am 30.11.2016 um 10:43 schrieb Andreas Schwab:
On Nov 30 2016, emanuel <emanuel.castelo@gmail.com> wrote:
what is the purpose of systemd-32bit in a 64bit install? It contains the PAM module. Interesting. There are 64-bit pam modules...
Why are the 32-bit ones needed on a 64-bit system? Herbert -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wednesday 2016-11-30 10:53, Herbert Graeber wrote:
Hi!
Am 30.11.2016 um 10:43 schrieb Andreas Schwab:
On Nov 30 2016, emanuel <emanuel.castelo@gmail.com> wrote:
what is the purpose of systemd-32bit in a 64bit install? It contains the PAM module. Interesting. There are 64-bit pam modules...
Why are the 32-bit ones needed on a 64-bit system?
Well, just in case you ever™ run a 32-bit /bin/su, you of course will need 32-bit pam plugins. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, Nov 30, 2016 at 10:43:02AM +0100, Andreas Schwab wrote:
On Nov 30 2016, emanuel <emanuel.castelo@gmail.com> wrote:
what is the purpose of systemd-32bit in a 64bit install?
It contains the PAM module.
systemd-32bit already contains a packageand(systemd:pam-32bit), why isn't this enough? Cheers, Michael. -- Michael Schroeder mls@suse.de SUSE LINUX GmbH, GF Jeff Hawn, HRB 16746 AG Nuernberg main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
systemd-32bit already contains a packageand(systemd:pam-32bit),
What do you mean with "contains"? Andreas. -- Andreas Schwab, SUSE Labs, schwab@suse.de GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7 "And now for something completely different." -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, Nov 30, 2016 at 11:05:21AM +0100, Andreas Schwab wrote:
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
systemd-32bit already contains a packageand(systemd:pam-32bit),
What do you mean with "contains"?
It was supposed to read "Supplements: packageand(systemd:pam-32bit)" $ rpm -q --supplements systemd-32bit packageand(systemd:pam-32bit) M. -- Michael Schroeder mls@suse.de SUSE LINUX GmbH, GF Jeff Hawn, HRB 16746 AG Nuernberg main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
On Wed, Nov 30, 2016 at 11:05:21AM +0100, Andreas Schwab wrote:
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
systemd-32bit already contains a packageand(systemd:pam-32bit),
What do you mean with "contains"?
It was supposed to read "Supplements: packageand(systemd:pam-32bit)"
$ rpm -q --supplements systemd-32bit packageand(systemd:pam-32bit)
How is that relevant? Andreas. -- Andreas Schwab, SUSE Labs, schwab@suse.de GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7 "And now for something completely different." -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, 2016-11-30 at 11:21 +0100, Andreas Schwab wrote:
$ rpm -q --supplements systemd-32bit
packageand(systemd:pam-32bit)
How is that relevant?
It's similar to a recommends: IF systemd AND pam-32bit are being installed, then systemd-32bit is recommended for installation, So, yes - it could probably be dropped from the pattern and not be recommended there. Cheers Dominique
On Wed, Nov 30, 2016 at 11:21:21AM +0100, Andreas Schwab wrote:
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
On Wed, Nov 30, 2016 at 11:05:21AM +0100, Andreas Schwab wrote:
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
systemd-32bit already contains a packageand(systemd:pam-32bit),
What do you mean with "contains"?
It was supposed to read "Supplements: packageand(systemd:pam-32bit)"
$ rpm -q --supplements systemd-32bit packageand(systemd:pam-32bit)
How is that relevant?
So you think the recommends systemd-32bit in the pattern is correct and all other pam packages that use a supplements are broken? M. -- Michael Schroeder mls@suse.de SUSE LINUX GmbH, GF Jeff Hawn, HRB 16746 AG Nuernberg main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
So you think the recommends systemd-32bit in the pattern is correct and all other pam packages that use a supplements are broken?
Why do you think so? Andreas. -- Andreas Schwab, SUSE Labs, schwab@suse.de GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE 1748 E4D4 88E3 0EEA B9D7 "And now for something completely different." -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 30.11.2016 11:56, Michael Schroeder wrote:
On Wed, Nov 30, 2016 at 11:21:21AM +0100, Andreas Schwab wrote:
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
On Wed, Nov 30, 2016 at 11:05:21AM +0100, Andreas Schwab wrote:
On Nov 30 2016, Michael Schroeder <mls@suse.de> wrote:
systemd-32bit already contains a packageand(systemd:pam-32bit),
What do you mean with "contains"?
It was supposed to read "Supplements: packageand(systemd:pam-32bit)"
$ rpm -q --supplements systemd-32bit packageand(systemd:pam-32bit)
How is that relevant?
So you think the recommends systemd-32bit in the pattern is correct and all other pam packages that use a supplements are broken?
Can you come down and talk about facts? The recommends for systemd-32bit are in rest_cd_* patterns that no user will have installed and Ludwig explained their purpose in detail on this mailing list this week. And there is a comment before this recommends explaining its need: # sssd-32bit pulls in some pam-32bit, which results in the need to also have # systemd-32bit available (or pam-config -a --systemd will fail later on) Recommends: systemd-32bit Greetings, Stephan -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, Nov 30, 2016 at 12:36:54PM +0100, Stephan Kulow wrote:
The recommends for systemd-32bit are in rest_cd_* patterns that no user will have installed and Ludwig explained their purpose in detail on this mailing list this week.
That wasn't clear from this thread and all the answers at all.
And there is a comment before this recommends explaining its need:
# sssd-32bit pulls in some pam-32bit, which results in the need to also have # systemd-32bit available (or pam-config -a --systemd will fail later on) Recommends: systemd-32bit
(This should be already handled with the Supplements in systemd-32bit.) M. -- Michael Schroeder mls@suse.de SUSE LINUX GmbH, GF Jeff Hawn, HRB 16746 AG Nuernberg main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);} -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 30.11.2016 13:14, Michael Schroeder wrote:
On Wed, Nov 30, 2016 at 12:36:54PM +0100, Stephan Kulow wrote:
The recommends for systemd-32bit are in rest_cd_* patterns that no user will have installed and Ludwig explained their purpose in detail on this mailing list this week.
That wasn't clear from this thread and all the answers at all.
And there is a comment before this recommends explaining its need:
# sssd-32bit pulls in some pam-32bit, which results in the need to also have # systemd-32bit available (or pam-config -a --systemd will fail later on) Recommends: systemd-32bit
(This should be already handled with the Supplements in systemd-32bit.)
It obviously didn't exist or didn't work at the point of time the recommend was added. Greetings, Stephan -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Wed, 2016-11-30 at 01:35 -0800, emanuel wrote:
I would not have noticed systemd-32bit if i had not tried to install systemd-232 , an error occurred , since then fixed, still i am left to wonder what is the purpose of systemd-32bit in a 64bit install?
This is the 32bit pam module... pam-config is nasty enough to bark if the 64it and 32-bit pam stack do not match Some pam maintainer might shed more light on the reason why and when the 32-bit pam modules might be needed. Cheers, Dominique
Hi, On 11/29/2016 12:35 PM, Karl Cheng wrote:
During installation, I did have some issues with a file conflict:
Detected 1 file conflict: File /usr/lib/systemd/libsystemd-shared-232.so from install of systemd-232.suse.20161124.gabef684-89.1.x86_64 (systemd) conflicts with file from install of systemd-32bit-232.suse.20161124.gabef684-89.1.x86_64 (systemd)
File conflicts happen when two packages attempt to install files with
the same name but different contents. If you continue, conflicting files will be replaced losing the previous content.
Continue? [yes/no] (no):
I typed "yes" in here, then realised that this was installing the 32-bit version when errors appeared during install about wrong ELF type.
I'll fix this up thanks for spotting ! Just out of curiosity why are both systemd and systemd-32bit installed on your system ?
System seemed to boot up without issue, but when checking journalctl it seems that postfix fails to load and xembedsniproxy dumps core. I don't know whether they are related to the systemd upgrade or not though.
I assume that those issues were not present before you upgraded systemd, right ? In this case could you provide the journal logs with debug logs enabled ?
Thanks for the work!
Thanks to you for taking time to test the beast ;) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Tuesday 2016-11-29 14:45, Franck Bui wrote:
Just out of curiosity why are both systemd and systemd-32bit installed on your system ?
42.1: systemd-32bit < libpulse0-32bit < libSDL*32bit < wine-32bit < wine Retaining systemd-32bit is possible if an older system was upgraded to TW at some point (where the split to libsystemd0-32bit was made). -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 11/28/2016 09:00 AM, Franck Bui wrote:
Hey,
Newest version of systemd v232 has been integrated and is mostly ready to land in Factory. It can be found here:
https://build.opensuse.org/package/show/home:fbui:systemd:next:openSUSE-Fact...
Packages have been updated to fix the following issues: - /usr/lib/systemd/libsystem-shared-232.so was shipped by both 32bit and 64bit packages which led to a file conflict when both packages were installed at the same time. - systemd was broken on i586 arch. This is due to: https://github.com/systemd/systemd/issues/4575. A workaround has been added until a proper fix is merged upstream. They are still available at the same place. Thanks a bunch for all of those who gave it a test, it really helps. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 11/28/2016 09:00 AM, Franck Bui wrote:
Hey,
Newest version of systemd v232 has been integrated and is mostly ready to land in Factory. It can be found here:
https://build.opensuse.org/package/show/home:fbui:systemd:next:openSUSE-Fact...
Packages have been updated (commit 4751a4d) to fix the following issues: - /usr/lib/systemd/libsystem-shared-232.so was shipped by both 32bit and 64bit packages which led to a file conflict when both packages were installed at the same time. - systemd was broken on i586 arch. This is due to: https://github.com/systemd/systemd/issues/4575. A workaround has been added until a proper fix is merged upstream. They are still available at the same place. Thanks a bunch for all of those who gave it a test, it really helps. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Am 29.11.2016 um 18:32 schrieb Franck Bui:
On 11/28/2016 09:00 AM, Franck Bui wrote:
Packages have been updated (commit 4751a4d) to fix the following issues:
- systemd was broken on i586 arch. This is due to: https://github.com/systemd/systemd/issues/4575. A workaround has been added until a proper fix is merged upstream.
The new i586 version works. The first one got me an unbootable system. Hendrik -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Am 28.11.2016 um 09:00 schrieb Franck Bui:
Hey,
Newest version of systemd v232 has been integrated and is mostly ready to land in Factory. It can be found here:
https://build.opensuse.org/package/show/home:fbui:systemd:next:openSUSE-Fact... Works for me. No issues on installation, no regression found. Cheers, Bernhard -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (17)
-
Aleksa Sarai -
Andreas Schwab -
Andrei Borzenkov -
Bernhard Held -
Daniel Molkentin -
Dominique Leuenberger / DimStar -
emanuel -
Franck Bui -
Franck Bui -
Hendrik Woltersdorf -
Herbert Graeber -
Jan Engelhardt -
Karl Cheng -
Michael Schroeder -
Per Jessen -
Rainer Klier -
Stephan Kulow