Re: [Mageia-dev] [changelog] [RPM] cauldron core/release kernel-3.8.3-2.mga3
On 03/19/2013 12:15 AM, tmb wrote: Name: kernel Relocations: (not relocatable) Version : 3.8.3 Vendor: Mageia.Org Release : 2.mga3Build Date: Mon Mar 18 22:32:12 2013 ... - PCI: Remove not needed check in disable aspm link (should fix ath5k problems) I can confirm this, it works fine now. Thanks !! -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Corrupt packets with ath5k
On 02/25/2013 10:03 AM, JA Magallón wrote: On 02/21/2013 08:02 PM, Thomas Backlund wrote: JA Magallón skrev 21.2.2013 02:28: On 02/20/2013 08:14 PM, JA Magallón wrote: Hi all... I have a strange problem with latest kernels. When I updatw my netbook I get many "Installation failed, bad rpms:" mesages. First I thougt that the oldie cheap ssd was failing (it is an Aspire AOA110). But the updates work always fine when plugged to ethernet. Now i tried to copy a couple RPMS via ssh over wifi and got: mplayer-1.1-11.r35916.1.mga3.tainted.i586.rpm 100% 2192KB 2.1MB/s 00:00 Received disconnect from 192.168.1.51: 2: Packet corrupt lost connection I try scp'ing a folder with 3 cauldron kernel rpms. Possible clues: - 3.7.1: works fine - warm boot in 3.8.0: transfer stalls and scp looks hanged, speed drops to zero, but no error message - cold boot in 3.8.0: transfer stops with Packet corrupt error or sometimes stalls.. Will ask in LKML. Does kernel-linus-3.8.0-1.mga3 work ? I finally tested it and no, it doesn't work either... I found a possible clue, it references a patch: http://blog.gmane.org/gmane.linux.drivers.ath5k.devel https://bugzilla.kernel.org/show_bug.cgi?id=55211 -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Release critical bugs for Mageia 3
On 03/15/2013 03:26 PM, Anne Nicolas wrote: Hi there We are in the last run for Mageia 3 final release. We have postponed the final release and added a beta 4 to give us some more time. The problem is we *still* have 59 bugs in list... We *must* all focus on it. Fixing bugs is for sure less fun than testing new versions and developing new functionnalities. But we are also working for Mageia users and we want to deliver a new version and make everybody happy with it. So here is a review of these bugs: https://docs.google.com/spreadsheet/ccc?key=0Ao3phYOTRNeQdEEtSjBSSWkxTmdIcmJZcGtfYjN1NVE&usp=sharing I have sorted it by BugID. Don't know if it is permanent or just for my view (not so expert in Google Docs), but it helps to find a bug you know you posted and have its ID... See the comments for each of them. We need help on testing, debugging and fixing bugs. Everybody can help. Please add your name if you want to help on one of them. Do not hesitate to ask any question here or on irc. The first reviews are rather good on beta 3. Keep on the good work Thanks all! -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Problem with console keymaps
Hi... I finally have discovered why my kbd never worked as it should in console: werewolf:~# loadkeys -v -u es | grep Loading Loading /usr/lib/kbd/keymaps/i386/olpc/es.map.gz It picks the keyboard layout from the OLCP ! If I delete/move the /usr/lib/kbd/keymaps/i386/olpc directory, everything works ok. But it should not be necessary. It also works if I do: werewolf:~# loadkeys -v -u qwerty/es | grep Loading Loading /usr/lib/kbd/keymaps/i386/qwerty/es.map.gz Solution ? What should I change in my /etc/sysconfig/keyboard: werewolf:/etc/sysconfig# cat keyboard XkbModel=pc105 XkbLayout=es KEYBOARD=es XkbOptions=compose:rwin KEYTABLE=es-latin1 What could break with, for example, KEYBOARD=qwerty/es ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release initscripts-9.41-11.mga3
On 02/27/2013 12:11 PM, colin wrote: Name: initscripts Relocations: (not relocatable) Version : 9.41 Vendor: Mageia.Org Release : 11.mga3 Build Date: Wed Feb 27 12:10:12 2013 Install Date: (not installed) Build Host: jonund.mageia.org Group : System/Base Source RPM: (none) Size: 2724125 License: GPLv2 and GPLv2+ Signature : (none) Packager: colin URL : http://www.mageia.org Summary : The inittab file and the /etc/init.d scripts Description : The initscripts package contains the basic system scripts used to boot your Mageia system, change run levels, and shut the system down cleanly. Initscripts also contains the scripts that activate and deactivate most network interfaces. colin 9.41-11.mga3: + Revision: 400506 - Bail out of ifup/down if NM is used and we cannot find the UUID/device mga#9028 - Fix sysvinit symlink cleanup filter again (thanks to Dan Fandrich) - Buildroot macro changes - Quote the file name in the cleanup filter. - Add a filetrigger to clean up broken sysvinit script links. - Reapply csh backspace fix added by Remmy lost on mgaconf rebase (mga#5007) Could you also remove the 'dm' script, that is useless nowadays ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Corrupt packets with ath5k
On 02/21/2013 08:02 PM, Thomas Backlund wrote: JA Magallón skrev 21.2.2013 02:28: On 02/20/2013 08:14 PM, JA Magallón wrote: Hi all... I have a strange problem with latest kernels. When I updatw my netbook I get many "Installation failed, bad rpms:" mesages. First I thougt that the oldie cheap ssd was failing (it is an Aspire AOA110). But the updates work always fine when plugged to ethernet. Now i tried to copy a couple RPMS via ssh over wifi and got: mplayer-1.1-11.r35916.1.mga3.tainted.i586.rpm 100% 2192KB 2.1MB/s 00:00 Received disconnect from 192.168.1.51: 2: Packet corrupt lost connection I try scp'ing a folder with 3 cauldron kernel rpms. Possible clues: - 3.7.1: works fine - warm boot in 3.8.0: transfer stalls and scp looks hanged, speed drops to zero, but no error message - cold boot in 3.8.0: transfer stops with Packet corrupt error or sometimes stalls.. Will ask in LKML. Does kernel-linus-3.8.0-1.mga3 work ? I finally tested it and no, it doesn't work either... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Corrupt packets with ath5k
On 02/20/2013 08:14 PM, JA Magallón wrote: Hi all... I have a strange problem with latest kernels. When I updatw my netbook I get many "Installation failed, bad rpms:" mesages. First I thougt that the oldie cheap ssd was failing (it is an Aspire AOA110). But the updates work always fine when plugged to ethernet. Now i tried to copy a couple RPMS via ssh over wifi and got: mplayer-1.1-11.r35916.1.mga3.tainted.i586.rpm 100% 2192KB 2.1MB/s 00:00 Received disconnect from 192.168.1.51: 2: Packet corrupt lost connection I try scp'ing a folder with 3 cauldron kernel rpms. Possible clues: - 3.7.1: works fine - warm boot in 3.8.0: transfer stalls and scp looks hanged, speed drops to zero, but no error message - cold boot in 3.8.0: transfer stops with Packet corrupt error or sometimes stalls.. Will ask in LKML. Ist that hardware failing ? Is driver failing ? Other laptop/androids update fine with the router via wifi. Any idea ? Will Google this while I expect it rings any bell for anyone in the list... TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Corrupt packets with ath5k
On 02/20/2013 08:14 PM, JA Magallón wrote: Hi all... I have a strange problem with latest kernels. When I updatw my netbook I get many "Installation failed, bad rpms:" mesages. First I thougt that the oldie cheap ssd was failing (it is an Aspire AOA110). But the updates work always fine when plugged to ethernet. Now i tried to copy a couple RPMS via ssh over wifi and got: mplayer-1.1-11.r35916.1.mga3.tainted.i586.rpm 100% 2192KB 2.1MB/s 00:00 Received disconnect from 192.168.1.51: 2: Packet corrupt lost connection Oops, sorry, forgot to spec the harware/driver: 03:00.0 Ethernet controller: Atheros Communications Inc. AR242x / AR542x Wireless Network Adapter (PCI-Express) (rev 01) Subsystem: Foxconn International, Inc. Device e008 Kernel driver in use: ath5k Ist that hardware failing ? Is driver failing ? Other laptop/androids update fine with the router via wifi. Any idea ? Will Google this while I expect it rings any bell for anyone in the list... TIA -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Corrupt packets with ath5k
Hi all... I have a strange problem with latest kernels. When I updatw my netbook I get many "Installation failed, bad rpms:" mesages. First I thougt that the oldie cheap ssd was failing (it is an Aspire AOA110). But the updates work always fine when plugged to ethernet. Now i tried to copy a couple RPMS via ssh over wifi and got: mplayer-1.1-11.r35916.1.mga3.tainted.i586.rpm 100% 2192KB 2.1MB/s 00:00 Received disconnect from 192.168.1.51: 2: Packet corrupt lost connection Ist that hardware failing ? Is driver failing ? Other laptop/androids update fine with the router via wifi. Any idea ? Will Google this while I expect it rings any bell for anyone in the list... TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] nss-ldap missing ?
On 02/12/2013 11:23 AM, Luca Olivetti wrote: Al 09/02/13 01:36, En/na JA Magallón ha escrit: I looked at some boxes on Uni with CentOS and use also nss-pam-ldapd (and CentOS is a not-so-up-to-date distro...). I migrated my Mageia boxes to that (both mga2 and mga3) and all is working fine. Would you be so kind to explain the migration procedure? TIA Sorry for the late answer... It really simple, just save your current ldap.conf/ldap.secret, config can be copied to new config files, syntax is similar. Install nss-pam-ldapd, urpmi will delete nss_ldap and pam_ldap. Copy the URI lines and secret to nslcd.conf, and start nslcd service. The only different thing (for what I used, I have a fairly simple setup) is the way of filtering users for machine access, it is even simpler in nss-pam-ldapd, you have 'filter' lines for passwd/shadow etc.. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] nss-ldap missing ?
On 02/20/2013 12:21 AM, David Walser wrote: David Walser writes: Thierry Vignaud writes: good luck with nslcd (if you use TLS you'll have to tell SELinux about it), Even if you don't use SELinux? I advise you sssd, it just work directly, one single config file to write for both PAM & NSS... Is that not the case for nss-pam-ldapd? I haven't investigated them yet. I did find out that there's some security bugs in sssd though (fixed upstream): https://bugs.mageia.org/show_bug.cgi?id=9027 Incidentally there's a security bug in nss-pam-ldapd too: https://bugs.mageia.org/show_bug.cgi?id=9113 I'm not sure which of the three patches (two linked by upstream advisory, one used by Debian in their update) is the right one to fix it. I think it is simpler to move to nss-pam-ldapd, specially if you want the minimal chages to drax* tool that sets up LDAP authentication (if there is any, I always did it by hand...): - same changes needed in nsswitch.conf, insert (or keep) the 'ldap' entry - syntax of nslcd.conf is similar to ldap.conf - just need to enable the service in systemd -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] nss-ldap missing ?
On 02/09/2013 12:31 AM, JA Magallón wrote: On 02/08/2013 01:41 AM, JA Magallón wrote: On 02/07/2013 11:44 PM, Guillaume Rousse wrote: Le 07/02/2013 18:07, Pascal Terjan a écrit : Is nss_ldap in the non-building package list, or is it deprecated and there is now another method to use ldap for authentication ? It was broken (new automake + undefined reference to `__libc_lock_*) and when investigating I found that it is dead and removed from several distros with suggestions to use nss-pam-ldapd (Fedora commit removing it also mentions sssd) BTW, if nss_ldap has been dropped, we could also drop pam_ldap, I don't much interest of allowing to authenticate users unknown from the system... Not _could_, but _should_. Currently pam_ldap requires nss_ldap, that is not in the repos no more (not only not rebuilt, but absent) And if mcc/drakxtools can be used to configure ldap, migrate everything to nss-pam-ldapd... Any decission on this ? I can move my servers to nss-pam-ldapd if thats the way to go... I looked at some boxes on Uni with CentOS and use also nss-pam-ldapd (and CentOS is a not-so-up-to-date distro...). I migrated my Mageia boxes to that (both mga2 and mga3) and all is working fine. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] nss-ldap missing ?
On 02/08/2013 01:41 AM, JA Magallón wrote: On 02/07/2013 11:44 PM, Guillaume Rousse wrote: Le 07/02/2013 18:07, Pascal Terjan a écrit : Is nss_ldap in the non-building package list, or is it deprecated and there is now another method to use ldap for authentication ? It was broken (new automake + undefined reference to `__libc_lock_*) and when investigating I found that it is dead and removed from several distros with suggestions to use nss-pam-ldapd (Fedora commit removing it also mentions sssd) BTW, if nss_ldap has been dropped, we could also drop pam_ldap, I don't much interest of allowing to authenticate users unknown from the system... Not _could_, but _should_. Currently pam_ldap requires nss_ldap, that is not in the repos no more (not only not rebuilt, but absent) And if mcc/drakxtools can be used to configure ldap, migrate everything to nss-pam-ldapd... Any decission on this ? I can move my servers to nss-pam-ldapd if thats the way to go... bran:~# urpmi pam_ldap A requested package cannot be installed: pam_ldap-186-4.mga3.i586 (due to unsatisfied nss_ldap[>= 217]) Continue installation anyway? (Y/n) While some packages may have been installed, there were failures. A requested package cannot be installed: pam_ldap-186-4.mga3.i586 (due to unsatisfied nss_ldap[>= 217]) Continue installation anyway? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] nss-ldap missing ?
On 02/07/2013 11:44 PM, Guillaume Rousse wrote: Le 07/02/2013 18:07, Pascal Terjan a écrit : Is nss_ldap in the non-building package list, or is it deprecated and there is now another method to use ldap for authentication ? It was broken (new automake + undefined reference to `__libc_lock_*) and when investigating I found that it is dead and removed from several distros with suggestions to use nss-pam-ldapd (Fedora commit removing it also mentions sssd) BTW, if nss_ldap has been dropped, we could also drop pam_ldap, I don't much interest of allowing to authenticate users unknown from the system... Not _could_, but _should_. Currently pam_ldap requires nss_ldap, that is not in the repos no more (not only not rebuilt, but absent) And if mcc/drakxtools can be used to configure ldap, migrate everything to nss-pam-ldapd... bran:~# urpmi pam_ldap A requested package cannot be installed: pam_ldap-186-4.mga3.i586 (due to unsatisfied nss_ldap[>= 217]) Continue installation anyway? (Y/n) While some packages may have been installed, there were failures. A requested package cannot be installed: pam_ldap-186-4.mga3.i586 (due to unsatisfied nss_ldap[>= 217]) Continue installation anyway? -- J.A. Magallon \ Winter is coming...
[Mageia-dev] nss-ldap missing ?
Hi...
After my usual ckeck for un-rebuilt packages, I have found this:
bran:~# rpm -qa --nosignature --qf "%{n}-%{v}-%{r}\n" | grep mga1 | sort
libatlas3-sse2-3.8.3-7.mga1
libopencore-amr0-0.1.2-3.mga1
nss_ldap-265-3.mga1
bran:~# rpm -ql nss_ldap
/etc/ldap.conf
/etc/ldap.secret
/lib/libnss_ldap-2.12.1.so
/lib/libnss_ldap.so.2
...
Is nss_ldap in the non-building package list, or is it deprecated and
there is now another method to use ldap for authentication ?
I could rpm -e it (nobody requires it), but I can not install it again
(it is not in any repo).
So I supposed perhaps I should migrate to pam_ldap, but:
bran:~# urpmi pam_ldap
A requested package cannot be installed:
pam_ldap-186-4.mga3.i586 (due to unsatisfied nss_ldap[>= 217])
Continue installation anyway? (Y/n)
While some packages may have been installed, there were failures.
A requested package cannot be installed:
pam_ldap-186-4.mga3.i586 (due to unsatisfied nss_ldap[>= 217])
Continue installation anyway?
Which is the correct method to use ldap ?
TIA
--
J.A. Magallon \ Winter is coming...
[Mageia-dev] Cleaning up init
Hi... After a test with symlinks -r, I discovered I had /etc/rc.d full of dangling symlinks, due to services moved to systemd. Should an update of initscripts clean them (symlinks -rd /etd/rc.d) ? I suppose this will also happen when people updates mga2 to mga3... I also discovered what seem like leftovers from previous conversions in /etc/rc.d/init.d: - dm (native prefdm exists) - irqbalance (package includes both, native and sysv) - linux_logo (same, native+sysv) - lm_sensors (idem) - mpd (idem) - mysqld (idem) - portreserve (idem) - netconsole (package is not even in the repos) - partmon (idem) Are we in time to correct these for Mageia 3 ? And to switch the remins to systemd, like resolvconf or postfix ? TIA -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Questions on dualbooting/grub/bootloader-utils
Hi... I have a curious problem... I have a box with tow drives, the first one has Windows7 and the second Mageia Cauldron. I tweaked the W7 bootloader to add an entry for the second drive with EasyBCD (it gets wrong the hd1,1 part for its grub entry, it shoud be hd1,0, but that is another story...). I can boot Mageia from W7 bootloader, so my linux disk is sdb, and win is sda. Root is right as it should be, /dev/sdb1: direwolf:~# lsscsi [0:0:0:0]cd/dvd HL-DT-ST DVD-RAM GH22NP20 2.00 /dev/sr0 [2:0:0:0]diskATA ST3400833AS 3.AA /dev/sda [3:0:0:0]diskATA ST3250310AS 3.AA /dev/sdb direwolf:~# df Filesystem Type Size Used Avail Use% Mounted on rootfs rootfs244G 6.1G 226G 3% / devtmpfs devtmpfs 1.6G 0 1.6G 0% /dev tmpfs tmpfs 1.6G 87k 1.6G 1% /dev/shm tmpfs tmpfs 1.6G 877k 1.6G 1% /run /dev/sdb1 ext4 244G 6.1G 226G 3% / tmpfs tmpfs 1.6G 0 1.6G 0% /sys/fs/cgroup tmpfs tmpfs 1.6G 13k 1.6G 1% /tmp /dev/sda2 fuseblk 400G 134G 267G 34% /media/win I rewrote GRUBs device.map like: (hd0) /dev/sda (hd1) /dev/sdb and menu.lst to use (hd1,0) as root. But every time I install/update a kernel, GRUBs files get rewritten to invert the disk mapping, so (hd0) is sdb, and menu.lst changed to boot from (hd0,0), and install.sh to write grub on hd0. But that settings do no work on boot, after EasyBCD chainloads to second drive, it is still the _second_ drive...and grub runs to hd0 to look for kernel/initrd... AFAIK, the map hd0->sdb is only usefull for grub installation, once the boot sector is stored it is just something like 'ask bios which is the first drive', so it fails. Why do bootloader-utils (I think?) change device.map, menu.lst settings ? Why forces the boot disk to be always hd0 ? Is there anyway to tell them I dont want to change anything ? TIA PD: is there an easier way to tell 'boot from the second drive' in the Win7 boot menu than using EasyBCD (I think it is doing some things not so well, like numbering first partition in second drive as hd1,1 instead of hd1,0...). -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Missign rebuilds tagged mga1
Hi...
As the mass rebuild seems almost done, I have supposed that everything in
my box not tagged ad mga3 are leftovers from previous installs, as I had
not reinstalled my box since mga1. I checked this:
rpm -qa --nosignature --qf "%{n}-%{v}-%{r}\n" | grep mga1 | sort
(I have another list for mga2 ;) )
But some packages seems to be needed, and skipped in the mass rebuild.
This is the list I get in my main box:
werewolf:~/bin# mga1
jline-0.9.94-4.mga1
lib64atlas3-x86_64-3.8.3-7.mga1
lib64iec16022_0-0.2.4-4.mga1
lib64jbig1-2.0-5.mga1
lib64jbig-devel-2.0-5.mga1
lib64mpeg2dec0-0.5.1-5.mga1
lib64opencore-amr0-0.1.2-3.mga1
net-tools-1.60-34.mga1
perl-Spreadsheet-ParseExcel-0.590.0-1.mga1
python-dateutil-1.5-2.mga1
transfugdrake-1.9.4-1.mga1
Some look pretty important:
jline -> rhino -> jdk7
net-tools -> basesystem
transfugdrake -> userdrake
... and so on.
Mirrors also seem to contain packages dated 2011. I will wait for mirrors to
settle
and look again. If they are dead packages, should not they be killed from
mirrors ?
TIA
--
J.A. Magallon \ Winter is coming...
[Mageia-dev] Mass rebuild and mirrors
Hi... An idea: would you think it could be sane to pause the mass rebuild a couple hours to let mirrors stabilise ? Or is it better to do it once and for all ? Anyone has measured the time needed for that (the resync of mirrors) ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Boot borked cause procfs :((
On 01/08/2013 02:01 PM, Thomas Backlund wrote: Colin Guthrie skrev 8.1.2013 14:28: 'Twas brillig, and JA Magallón at 08/01/13 12:11 did gyre and gimble: Hi... After updating cauldron to latest packages, I rebuilt manually the boot initrd with dracut -f. After that, the system does not boot: mount: /proc: filesystem mounted, but mount(8) failed: No such file or directory Cannot mount proc on /proc! Compile the kernel with CONFIG_PROC_FS! And then a kernel panic... https://bugs.mageia.org/show_bug.cgi?id=8614 Thanks, after glibc/util-linux upgrade, I regenerated initrds and all works fine. Goog job !! I think that I have boxes that boot because I have not rebuilt the initrd, but I don't dare to try one other :). Any ideas o how to debug/fix this ? I heard this is due to new util-linux. My initrd was from 3rd Jan, but util-linux installed on 5th Jan. I've regenerated my initrd so will reboot sometime today to see if it now breaks with the new version of mount. It will. I hope you have a backup initrd to use. > We maybe need to patch mount? I don't think it's really the kernel that's at fault here. I will investigte it tonight.. -- Thomas -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Boot borked cause procfs :((
On 01/08/2013 02:01 PM, Thomas Backlund wrote: Colin Guthrie skrev 8.1.2013 14:28: 'Twas brillig, and JA Magallón at 08/01/13 12:11 did gyre and gimble: Hi... After updating cauldron to latest packages, I rebuilt manually the boot initrd with dracut -f. After that, the system does not boot: mount: /proc: filesystem mounted, but mount(8) failed: No such file or directory Cannot mount proc on /proc! Compile the kernel with CONFIG_PROC_FS! And then a kernel panic... https://bugs.mageia.org/show_bug.cgi?id=8614 Well, i copied the libraries amd mount binary from a beta3 image, rebuilt initrd and the system boots. I think that I have boxes that boot because I have not rebuilt the initrd, but I don't dare to try one other :). Any ideas o how to debug/fix this ? I heard this is due to new util-linux. My initrd was from 3rd Jan, but util-linux installed on 5th Jan. I've regenerated my initrd so will reboot sometime today to see if it now breaks with the new version of mount. It will. I hope you have a backup initrd to use. Now I have ;) > We maybe need to patch mount? I don't think it's really the kernel that's at fault here. I will investigte it tonight.. -- Thomas -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Boot borked cause procfs :((
Hi... After updating cauldron to latest packages, I rebuilt manually the boot initrd with dracut -f. After that, the system does not boot: mount: /proc: filesystem mounted, but mount(8) failed: No such file or directory Cannot mount proc on /proc! Compile the kernel with CONFIG_PROC_FS! And then a kernel panic... I think that I have boxes that boot because I have not rebuilt the initrd, but I don't dare to try one other :). Any ideas o how to debug/fix this ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] NVIDIA CUDA 5 has landed
On 12/14/2012 10:08 PM, Dimitri wrote: Hi, Finally, NVIDIA CUDA Toolkit 5.0.35 has landed in Cauldron. I apologize for the delay, and ask everyone interested to test the packages (especially on x86_64). Many thanks !!! * GCC 4.7 is not supported. However, you can fool nvcc and compile a program with -D__GNUC_MINOR__=6; I used to disable the check in CUDA itself, it is a simple patch, goes attached. So you dont have to tweak every compile line in yor project. BTW, CUDA works fine for me with gcc 4.7. The only problem I know is that cuda-gdb wont work with 4.7 because different formats in binaries (COFF/ELF related ???). The work on the package is still not finished. The package contains an init script (nvidia) that creates device nodes and loads kernel module. This "service" is intended to be started on GUI-less compute nodes. I want to ask for assistance in migrating this script to systemd unit. As far as I know, with latest kernels and NVIDIA drivers the device nodes always get created automatically (never noticed them not being created). So probably the only thing left is to load kernel module. Seems like putting it into modules-load.d/* is a bad idea, because if the module is absent, systemd-modules-load.service will fail. Should we simply make a unit with ExecStart=/usr/sbin/modprobe nvidia-current instead? I also used the script to workaround a problem with running CUDA on a system without X. If there is no X running on the nVidia card, CUDA programs take a lot to launch. So the solution is to launch something like nvidia-smi at boot and let it running. Now you seem to be looking into packaging, I will tell about a problem I have. It can be at the same time very specific, but also very common. I have a box with Intel graphics, they work fine for 2D and desktop, and an nvidia card (a cheap but powerfull GT640), to crunch numbers. This can be a very common setup where someone has a box with intel graphics and an accelerator Tesla card. They want nVidia CUDA but not nVidia GL. So, I want the GL from Mesa for intel graphics, but libcuda from nVidia libraries. Problem: if /usr/lib64/nvidia-current is in the path (via alteratives), libGL also picks the nvidia one, not Mesa's. And that breaks desktop. I have to rm the nvidia GL libs. Also, cuda libraries are independent of GL ones, and should not go into nvidia-current, they could go just fine in /usr/lib[64]. They should not even suggests x11-driver-nvidia, cause for example a server with a Tesla card needs the kernel module but not the x11 driver. Then there is the problem of some utilities you need on a headless CUDA server that are now packaged into the x11-driver (nvidia-smi). In short, I would like a CUDA environment without the CUDA x11 driver and GL libraries. It can work fine. And as a side effect, now that drivers for 8xxx and 9xxx cards are different from the next generations, and both support CUDA, there should not be any cuda part in the driver package... So I would propose these, blame me if this is too complicated. All this can be done and checked in separate steps or package releases: - move libs in nvidia-current-cuda-opencl from /usr/lib[64]/nvidia-current to /usr/lib[64]. None links against libGL, and work on an environment with Mesa's libGL, and with alternatives for gl_conf set to standard. They do not comflict with anybodyelse nor any other package provides an alternative. And they can even be used without an nVidia card with pthreads emulation. (perhaps in the future if AMD or Intel give its own libOpenCL,but then probaly OpenCL libraries would need to be split apart). - move libcuda.so symlink (and all from nvidia-current-devel related to cuda libraries) to nvidia-cuda-toolkit-devel, also in /usr/lib). So you don't need GL devel package and bunch of requires to build CUDA apps... - move nvidia-smi/nvidia-xconfig to its own package (nvidia-cli-utils?), they do not require GL not even X. - perhaps rename nvidia-cuda-tollkit-* to nvidia-cuda-* ;) (I really think the -toolkit part is redundant...) So the target will be to be able to have a CUDA compute environment without the X or GL part of nvidia. Sorry for the long and boring mail, but I had this in mind since looong ago. TIA -- J.A. Magallon \ Winter is coming... --- include/host_config.h.orig 2012-12-14 22:42:46.802874939 +0100 +++ include/host_config.h 2012-12-14 22:43:01.484187297 +0100 @@ -77,11 +77,11 @@ #if defined(__GNUC__) -#if __GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ > 6) - -#error -- unsupported GNU version! gcc 4.7 and up are not supported! - -#endif /* __GNUC__> 4 || (__GNUC__ == 4 && __GNUC_MINOR__ > 6) */ +//#if __GNUC__ > 4 || (__GNUC__ == 4 && __GNUC_MINOR__ > 6) +// +//#error -- unsupported GNU version! gcc 4.7 and up are not supported! +// +//#endif /* __GNUC__> 4 || (__GNUC__ == 4 && __GNUC_MINOR__ > 6) */ #endif /* __GNUC__ */
Re: [Mageia-dev] Video players
On 11/24/2012 09:49 PM, Pierre Jarillon wrote:
Le samedi 24 novembre 2012 17:25:02, JA Magallón a écrit :
If you can post your video test collection somewhere, I can try to
see if totem works for me with them.
Sorry for the delay. I have uploaded a kit (33 files 500MB) of videos on
http://pjarillon.free.fr/docs/demos/
Some types of videos are given with the command file
$ ls | while read f;do file $f;done|awk -F: '{print $2 ":" $1}' | sort
Some formats 19/9 are displayed at 4/3 ratio.
As I said previously VLC works fine in every case.
Sorry for the delay...
In my case, all play in totem except SWF files.
The only problem I saw was some chopping with audio in a couple videos,
the ballmer_dance one and the casseaur.mov.
--
J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] NetworkManager vs initscripts
On 11/28/2012 03:15 PM, Jose Jorge wrote: Le 28/11/2012 14:05, Johnny A. Solbu a écrit : On Wednesday 28. November 2012 12.44, Olivier Blin wrote: You can just set NM_CONTROLLED=yes in the generated ifcfg files. If the interface is configured using NetworkManager, why doesn't NM itself create the file, if it doesn't exist, and add the line, to prevent this? Perhaps a checkbox could be added to make it happen. Because the GUI tools and the installer always create this file. Except hackers, no one removes it, ans MCC adds NM_CONTROLLED=yes if you check the box in the GUI! Well, this is what I have undestood and checked, plz correct me if wrong: - initscripts always want an ifcfg-eth0 (-wlan0) file - if it is not there, it is created by udev - mageia gui tools set the NM_CONTROLLED flag, but still rely on the ifcfg-rh plugin to let NM pull the config from there To get the most 'pure' NM setup I could, I did this: - disabled the ifcfg-rh plugin in NM.conf, just keyfile - configured the network via nm-connection-editor, stored system wide in /etc/NetworkManager/system-connections - creaed a dummy ifcfg-eth0 file with DEVICE=eth0 NM_CONTROLLED=yes lines. No network info in there...just dummy file for initscripts to ignore it, as udev will be trying to create it with DHCP setup if not present... Perhaps GUI toos and installer could do this also, if user wants the interface to be managed by NM. Perhaps it is time to deprecate ifcfg-rh and create native files for NM... NM is ispecially usefull if you connect to multiple wifis and want different routers/DNS/domains depending on your wifi network. Thanks. PD: diggin a bit I found /etc/sysconfig/udev_net, with its BLACKLIST lists, and found also about /etc/sysconfit/network and AUTOMATIC_IFCFG (see /lib/udev/net_create_ifcfg). Places to play... -- J.A. Magallon \ Winter is coming...
[Mageia-dev] NetworkManager vs initscripts
Hi all... I have a curious problem with NM and initscripts mix... I tried to move one of my boxes to a full NM setup, no sysv network scripts. So I set my NetworkManager.conf like this: [main] plugins=keyfile no-auto-default=* [keyfile] hostname=werewolf.xxx.xxx Configured the network with nm-connection-editor (static IP), and got a file in /etc/NetworkManager/system-connections/Ethernet: [802-3-ethernet] duplex=full mac-address= [connection] id=Ethernet uuid=cdc62758-00f4-4728-bf94-e2b0f9fb06d9 type=802-3-ethernet timestamp=1354033074 [ipv6] method=ignore [ipv4] method=manual dns= dns-search=... addresses1=... may-fail=false And finally I deleted ifcfg-eth0 from /etc/sysconfig/network-scripts. Network starts and stops ok, but on reboot, some process inists on creating a ifcfg-eth0 file, with this setup: DEVICE=eth0 BOOTPROTO=dhcp ONBOOT=yes which breaks networking on systems that have a fixed IP, and works by chance if you use DHCP. Who creates that ? How can I kill it ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Video players
On 11/24/2012 05:33 PM, Johnny A. Solbu wrote: On Saturday 24. November 2012 17.25, JA Magallón wrote: And wrt to copyrighted decoders ... Unless the decoder is in the Public Domain, it is copyrighted. The Linux kernel is copyrighted. Even the GNU packages is copyrighted. Don't confuse copyrighted with proprietary. Yeah, sorry, that's what I meant... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Video players
On 11/24/2012 01:28 AM, Pierre Jarillon wrote: For Mageia 3 it is necessary to read all kinds of videos. I have a set of video to test the distro. Totem is out of service but its is however the defaut reader... For some videos, Dragon is not shown in the list of players but works. http://vimeo.com/53980967 from vimeo works with Mageia 1 but not in cauldron. VLC works always. My suggestions are: - When VLC is loaded, it must become the default video viewer. - For each video type, define the viewer (Totem, Dragon) which works. Some points, specially related to Totem that is what I mainly use: - Totem/GStreamer plays all kind of videos I have given it, mainly divx/mp4/h264/mpeg2 with mp3/ac3 audio in avi/mp4/mkv containers. It all depends on your gstreamer being correctly installed. - Videos like http://vimeo.com/53980967 are just FLASH, they do not depend on Totem,VLC o Dragon. It works for me in Cauldron, as YouTube both in Flash or HTML5 formats. - Most desktops DEs have a native player (Totem for Gnome, Dragon for KDE, Parole for XFCE, ...). That should work and be the default out of the box, that is what is most integrated with each DE. Suppose that Totem was the absolutely best player regarding fomats. What would you think KDE users would say if it was made the default for KDE, forcing them to install half Gnome to see videos ? - Many desktops have some kind of 2nd-to-best player. Perhaps VLC is a good option for KDE (it is Qt based anyways), but for Gnome or XFCE (GTK based) the best next option IMHO is Gnome MPlayer (as good as as VLC). So the only problem with Totem is probably urpmi requires/suggests missing to guarantee that is plays all audio/video formats, or at least most used. And I suppose you can have the same problems with Phonon (or whatever is the KDE equivalent to GStreamer). And wrt to copyrighted decoders, the problem is the same for all, you have to use Tainted repos to get a GStreamer/VLC/MPlayer that decodes them. If you can post your video test collection somewhere, I can try to see if totem works for me with them. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release ldetect-lst-0.1.311-1.mga3
On 11/21/2012 11:26 PM, anssi wrote: Name: ldetect-lst Relocations: (not relocatable) Version : 0.1.311 Vendor: Mageia.Org Release : 1.mga3Build Date: Wed Nov 21 23:25:17 2012 Install Date: (not installed) Build Host: ecosse.mageia.org Group : System/Kernel and hardwareSource RPM: (none) Size: 1269863 License: GPLv2+ Signature : (none) Packager: anssi URL : http://svnweb.mageia.org/soft/ldetect-lst/trunk/ Summary : Hardware list for the light detection library Description : The hardware device lists provided by this package are used as lookup table to get hardware autodetection. anssi 0.1.311-1.mga3: + Revision: 320393 - new version 0.1.311 o switch DRIVER2 for GeForce 7xxx and GeForce 8xxx cards to the newly added 304.xx series proprietary driver as support is dropped from nvidia-current This should be 6xxx and 7xxx, as 8xxx and above are supported by the new driver, AFAIK. Is just a typo in the description or also in the lists ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] ANN: shadow-utils & util-linux with new default paths in testing
On 11/20/2012 11:43 PM, Luc Menut wrote: Hi, Following my proposal to fix and unify default paths after UsrMove http://www.mageia.org/pipermail/mageia-dev/2012-November/019931.html I pushed in cauldron core/updates_testing shadows-utils-4.1.5.1-2 and util-linux-2.22.1-7 using these default paths: PATH=/usr/local/bin:/usr/bin for normal user, PATH=/usr/local/sbin:/usr/sbin:/usr/local/bin:/usr/bin for root. Mmmm, this is a discussion I always had with my sysadmin at work... Do you really want that a binary in local can oerride one in system ? Should not local paths be after system ones ? Please test these new packages. Any comments and suggestions are welcome. Regards, Luc -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release totem-3.6.3-2.mga3
On 11/16/2012 11:35 AM, ovitters wrote: Name: totemRelocations: (not relocatable) Version : 3.6.3 Vendor: Mageia.Org Release : 2.mga3Build Date: Fri Nov 16 11:33:14 2012 Install Date: (not installed) Build Host: jonund.mageia.org Group : Video/Players Source RPM: (none) Size: 3160950 License: GPLv2 with exception Signature : (none) Packager: ovitters URL : http://projects.gnome.org/totem/ Summary : Movie player for GNOME Description : Totem is simple movie player for the GNOME desktop. It features a simple playlist, a full-screen mode, seek and volume controls, as well as a pretty complete keyboard navigation. ovitters 3.6.3-2.mga3: + Revision: 318912 - requires gstreamer1(element-scaletempo) (mga#8088) Mmm, something is wrong: pandora:~# rpm -q --provides gstreamer1.0-plugins-bad | grep scale libgstscaletempoplugin.so()(64bit) gstreamer1(element-scaletempo)()(64bit) pandora:~# urpmi totem A requested package cannot be installed: totem-3.6.3-2.mga3.x86_64 (due to unsatisfied gstreamer1(element-scaletempo)) Continue installation anyway? (Y/n) While some packages may have been installed, there were failures. A requested package cannot be installed: totem-3.6.3-2.mga3.x86_64 (due to unsatisfied gstreamer1(element-scaletempo)) Continue installation anyway? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Can't start LibreOffice since last update
On 11/09/2012 01:56 PM, Robert Fox wrote: [rfox@mainfox ~]$ libreoffice /usr/lib64/libreoffice/program/soffice.bin: symbol lookup error: /usr/lib64/libreoffice/program/libvcllo.so: undefined symbol: _ZNK3icu14LEFontInstance10getSubFontEPKDsPiiiR11LEErrorCode Anyone else?? +1 (x86-64) -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Fwd: fallback mode
On 11/06/2012 12:01 AM, Olav Vitters wrote: On Mon, Nov 05, 2012 at 11:39:41PM +0100, Olivier Blin wrote: Olav Vitters writes: On Mon, Nov 05, 2012 at 07:27:26PM +0100, Olivier Blin wrote: Even an Intel Core2 Duo (E8400) from 2008 + Intel Q35 chipset from 2007 can not handle the load with Gnome Shell on a 1920x1080 display, using accelerated GL drivers. Sure it is not using llvmpipe accidentally? What kind of driver does that Q35 use? I have an intel chipset in my old laptop, no clue what, but it is really smooth. Q35 has an Intel GMA3100 chipset, it is using i915. I don't think it is using llvmpipe, since /usr/lib64/gnome-session-check-accelerated-helper succeeds. Is there a better way to check that? System Settings → Details. Which Intel chipset is your old laptop using? Intel® 965GM. I guess that is newer than your laptop. My laptop has a BIOS from 2007, I guess it was made around that time. GNOME shell seems really smooth, same as my main machine though resolution is just 1280×1024. I have connected my Atom netbook to a big monitor, and it handles perfecly the FullHD resolution. Gnome full mode works fine and fast. Its an Atom N450, graphics are recognised as N10 or Intel(R) Pineview GM. All this with the monitor as the _ONLY_ monitor active. If I try to use it on a multi-monitor setup, X gives a message about non-tiled framebuffer and gnome-shell is slow as hell, unusable. Probably that is a hardware limit, or a bug, because it happens even if the external monitor is a projector at 1024x768. I was wating to do more tests to submit a bug -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Notes about live CD
On 10/27/2012 12:40 AM, Wolfgang Bornath wrote: 2012/10/27 JA Magallón : On 10/26/2012 12:54 PM, Olivier Blin wrote: JA Magallón writes: - the installer asks to remove the unused harware support, but the wording for the choice is misleading: [ ] Unsued hardware support What does it really mean if I tick this, I want to keep the unused hw support or to remove it ? The text reads this way: We will remove the following packages, unless you choose otherwise: [ ] Unused hardware support [ ] Unused localization Oops, sorry for the late answer. In the live _CD_, I saw no choice for localization, I think it is because it only includes english ? The live DVD has both marks to check. I will rephrase it as: There is some soft installed by default that is not needed in your current setup. Mark them if you want it to be removed now: [ ] Unused hardware support [ ] Unused localizations -- J.A. Magallon \ Winter is coming... As it is now the user has to read the whole thing (first the leading sentence, then the options, then put both in a meaningful relation) before he can decide what to do. Our goal was always to make it as simple as reasonable for the user to understand what we exactly mean. So why not write in the options what is really done? Some installed packages (hardware and language support) may not be needed on your computer. [ ] Check to remove unused hardware support [ ] Check to remove unused language support Here the option tells exactly and with no room for misunderstanding in one line what it is doing when checked or unchecked. Nice -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Notes about live CD
On 10/26/2012 12:54 PM, Olivier Blin wrote: JA Magallón writes: - the installer asks to remove the unused harware support, but the wording for the choice is misleading: [ ] Unsued hardware support What does it really mean if I tick this, I want to keep the unused hw support or to remove it ? The text reads this way: We will remove the following packages, unless you choose otherwise: [ ] Unused hardware support [ ] Unused localization Oops, sorry for the late answer. In the live _CD_, I saw no choice for localization, I think it is because it only includes english ? The live DVD has both marks to check. I will rephrase it as: There is some soft installed by default that is not needed in your current setup. Mark them if you want it to be removed now: [ ] Unused hardware support [ ] Unused localizations -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Notes about live CD
Hi... I have used the Gnome LiveCD 32bits to do some disk test on an old box, and I have a couple gotchas I would have to share in the list (don't know if they are really bugs). Here I go: - this is an old i965 box. The ATA disk is named like hda. Why has IDE driver preference over SATA ? I would expect it to be sda... - the installer asks to remove the unused harware support, but the wording for the choice is misleading: [ ] Unsued hardware support What does it really mean if I tick this, I want to keep the unused hw support or to remove it ? Perhaps you could rephrase it like this, depending on what does it really mean: [ ] Keep unused harware support or [ ] Remove unused hardware support Thats all for the moment... -- J.A. Magallon \ Software is like sex: \ It's better when it's free
[Mageia-dev] Removing ConsoleKit
Hi... As this page reads http://www.freedesktop.org/wiki/Software/ConsoleKit CK is not maintaned anymore, it is deprecated. We can not remove the packages because of dependencies: werewolf:~# urpmq --whatrequires consolekit consolekit-x11 lib64consolekit0 accountsservice consolekit consolekit-x11 elsa kdm lib64consolekit-devel lib64consolekit0 lxdm polkit slim xinit but in the meantime I have disabled it completely following this: http://forums.fedoraforum.org/showthread.php?t=246130 and disabling it in systemctl. Everything seems to work fine, I have noticed no problem. Will it be possible to build the above packages without CK support and with systemd-enabled options it they are there ? So mga3 will be a CK-free edition ;) TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Problems with gnome-shell and polkit
On 09/24/2012 07:51 PM, Olav Vitters wrote: On Mon, Sep 24, 2012 at 12:42:58AM +0200, JA Magallón wrote: Perhaps polkit-gnome is out of sync ? From the .spec: # install the autostart file manually, since upstream doesn't ship it any more # and decided that it's the job of each DE to start polkit-gnome-authentication-agent-1 # this doesn't seem to apply to gnome-session in GNOME 2.32 # ACK'ed by pterjan # # this file was nicked from the gnome-polkit-0.99 binary rpm, to get all the translations # (ahmad) 02-05-2011 I removed the autostart file in a newly submitted polkit-gnome. Thanks !! BTW, do we need both polkit and policykit installed ? I can remove libpolkit2-0.9-8.mga1,policykit-0.9-8.mga1 without any problem. Should new version obsolete old one ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Problems with gnome-shell and polkit
On 09/24/2012 12:42 AM, JA Magallón wrote:
Hi...
After latest updates GDM works again, but
I get a sad-computer-screen-of-death when I try to log in in Gnome.
The culprit seems to be the gnome-shell / polkit interaction.
With startx from vconsole I get this on .xsession-errors:
Window manager warning: Log level 16: Unable to register authentication agent:
GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: An authentication agent
already exists for the given subject
JS ERROR: !!! Exception was: Polkit.Error:
GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: An authentication agent
already exists for the given subject
JS ERROR: !!! message =
'"GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: An authentication agent
already exists for the given subject"'
JS ERROR: !!! fileName =
'"/usr/share/gnome-shell/js/ui/components/polkitAgent.js"'
JS ERROR: !!! lineNumber = '329'
JS ERROR: !!! stack = '"0
anonymous()@/usr/share/gnome-shell/js/ui/components/polkitAgent.js:329
1 wrapper()@/usr/share/gjs-1.0/lang.js:204
2 anonymous("name" =
""polkitAgent"")@/usr/share/gnome-shell/js/ui/components/__init__.js:56
3 wrapper(""polkitAgent"")@/usr/share/gjs-1.0/lang.js:204
4 anonymous("name" = ""polkitAgent"", 1, [object
Array])@/usr/share/gnome-shell/js/ui/components/__init__.js:22
5 anonymous()@/usr/share/gnome-shell/js/ui/components/__init__.js:21
6 wrapper()@/usr/share/gjs-1.0/lang.js:204
7 anonymous()@/usr/share/gnome-shell/js/ui/components/__init__.js:13
8 wrapper()@/usr/share/gjs-1.0/lang.js:204
9 anonymous()@/usr/share/gjs-1.0/lang.js:145
10 anonymous()@/usr/share/gjs-1.0/lang.js:239
11 start()@/usr/share/gnome-shell/js/ui/main.js:150
12 @:1
"'
If I comment this in /usr/share/gnome-shell/js/ui/components/polkitAgent.js, at
least I
can log in again:
enable: function() {
//this._native.register();
},
disable: function() {
//this._native.unregister();
},
I have this versions of polkit:
lib64polkit1_0-0.107-1.mga3 x86_64
lib64polkit-gir1.0-0.107-1.mga3 x86_64
polkit-0.107-1.mga3 x86_64
polkit-desktop-policy-0.107-1.mga3 noarch
polkit-gnome-0.105-1.mga2 x86_64
Perhaps polkit-gnome is out of sync ?
TIA
From what I looked in Google, they are all the latest versions.
I solved the issue uncomenting those lines and adding this:
AutostartCondition=GNOME3 unless-session gnome
in /etc/xdg/autostart/polkit-gnome-authentication-agent-1.desktop, as
explained here:
http://www.linuxfromscratch.org/blfs/view/svn/gnome/polkit-gnome.html
Is this right for Mageia, or should the tags "GNOME3" and "gnome" changed
to something else ? Is this worth a bug report ?
TIA
--
J.A. Magallon \ Winter is coming...
[Mageia-dev] Problems with gnome-shell and polkit
Hi...
After latest updates GDM works again, but
I get a sad-computer-screen-of-death when I try to log in in Gnome.
The culprit seems to be the gnome-shell / polkit interaction.
With startx from vconsole I get this on .xsession-errors:
Window manager warning: Log level 16: Unable to register authentication agent:
GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: An authentication agent
already exists for the given subject
JS ERROR: !!! Exception was: Polkit.Error:
GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: An authentication agent
already exists for the given subject
JS ERROR: !!! message =
'"GDBus.Error:org.freedesktop.PolicyKit1.Error.Failed: An authentication agent
already exists for the given subject"'
JS ERROR: !!! fileName =
'"/usr/share/gnome-shell/js/ui/components/polkitAgent.js"'
JS ERROR: !!! lineNumber = '329'
JS ERROR: !!! stack = '"0
anonymous()@/usr/share/gnome-shell/js/ui/components/polkitAgent.js:329
1 wrapper()@/usr/share/gjs-1.0/lang.js:204
2 anonymous("name" =
""polkitAgent"")@/usr/share/gnome-shell/js/ui/components/__init__.js:56
3 wrapper(""polkitAgent"")@/usr/share/gjs-1.0/lang.js:204
4 anonymous("name" = ""polkitAgent"", 1, [object
Array])@/usr/share/gnome-shell/js/ui/components/__init__.js:22
5 anonymous()@/usr/share/gnome-shell/js/ui/components/__init__.js:21
6 wrapper()@/usr/share/gjs-1.0/lang.js:204
7 anonymous()@/usr/share/gnome-shell/js/ui/components/__init__.js:13
8 wrapper()@/usr/share/gjs-1.0/lang.js:204
9 anonymous()@/usr/share/gjs-1.0/lang.js:145
10 anonymous()@/usr/share/gjs-1.0/lang.js:239
11 start()@/usr/share/gnome-shell/js/ui/main.js:150
12 @:1
"'
If I comment this in /usr/share/gnome-shell/js/ui/components/polkitAgent.js, at
least I
can log in again:
enable: function() {
//this._native.register();
},
disable: function() {
//this._native.unregister();
},
I have this versions of polkit:
lib64polkit1_0-0.107-1.mga3 x86_64
lib64polkit-gir1.0-0.107-1.mga3 x86_64
polkit-0.107-1.mga3 x86_64
polkit-desktop-policy-0.107-1.mga3 noarch
polkit-gnome-0.105-1.mga2 x86_64
Perhaps polkit-gnome is out of sync ?
TIA
--
J.A. Magallon \ Winter is coming...
[Mageia-dev] Problems with i915 DRI
Hi... I have one netbook where gsm tries to start and I get this in :0-greeter.log: libGL error: failed to load driver: i915 libGL error: Try again with LIBGL_DEBUG=verbose for more details. gnome-session-is-accelerated: No hardware 3D support. gnome-session-check-accelerated: Helper exited with code 256 gnome-session[4008]: WARNING: Session 'gdm-shell' runnable check failed: Child process exited with code 1 The dri driver is there: one:~# ll /usr/lib64/dri total 28112 -rwxr-xr-x 1 root root 387520 2012.09.14 21:11 i915_dri.so* -rwxr-xr-x 1 root root 858376 2012.09.14 21:11 i965_dri.so* -rwxr-xr-x 1 root root 632896 2012.06.13 21:12 i965_drv_video.so* -rwxr-xr-x 1 root root 5532184 2012.09.14 21:11 nouveau_dri.so* -rwxr-xr-x 1 root root 199080 2012.09.14 21:11 nouveau_vieux_dri.so* -rwxr-xr-x 1 root root 355240 2012.09.14 21:11 r200_dri.so* -rwxr-xr-x 1 root root 4882696 2012.09.14 21:11 r300_dri.so* -rwxr-xr-x 1 root root 4853080 2012.09.14 21:11 r600_dri.so* -rwxr-xr-x 1 root root 318136 2012.09.14 21:11 radeon_dri.so* -rwxr-xr-x 1 root root 5831576 2012.09.14 21:11 radeonsi_dri.so* -rwxr-xr-x 1 root root 4919096 2012.09.14 21:11 swrast_dri.so* How can I debug this ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] unity on mageia
On 09/12/2012 10:05 PM, Damian Ivanov wrote: My name isDamian Ivanov. Xiao-Long Chen and me are the maintainers of unity desktop environment for openSUSE and Fedora, Xiao-Long additionally maintains also unity on Arch Linux. We use for building the packages the open build service https://build.opensuse.org/project/monitor?project=GNOME %3AAyatana It actually builds install-able and usable packages for Fedora and openSUSE. This week we'll give an official announcement that unity works on openSUSE and will make a live CD available based on openSUSE with unity. We would also like to see Mageia running unity, but this would not be done by us ( I am mainly working on openSUSE, Xiao-Long is working on Fedora and openSUSE). But I offer my help, if you want to do it also with obs on GNOME:Ayatana. At the moment the structure of a package is as follows: ... There are users of Mageia who would love to see it in Mageia: https://forums.mageia.org/en/viewtopic.php?f=29&t=130 It would't need to be done that much to get it working on Mageia. The spec files for openSUSE and Fedora are quite similar and based on each other. I have no voice nor vote, but I really would prefer Mageia invest efforts if any on integrating Cinnamon better than Unity. But that is just personal stuff... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [RFC] x11-server-1.13
On 09/10/2012 11:12 PM, Thomas Backlund wrote: 10.09.2012 23:46, JA Magallón skrev: On 09/10/2012 07:31 PM, AL13N wrote: Op maandag 10 september 2012 15:02:37 schreef JA Magallón: On 09/08/2012 01:02 PM, Thierry Vignaud wrote: Hi x11-server-1.13 and rebuild drivers are in */update_testing. Tested on one desktop and two netbooks, so working intel, synaptics. Tested also on propietary nvidia driver, and broken. Someting in gdm or gnome-shell bombs in libpthread I wil request an update to the nvidia driver, because of this paragraph in new README: "Fixed a bug that caused pre-release versions of X.Org xserver 1.13 to crash when certain GLX operations were performed, such as when starting Firefox." Perhaps it is related. hey, i crash FF too, if it's on google maps with 3DGL enabled Checked with success. I installed manually the new driver and everything works again. nvidia-current-304.43 pushed to nonfree/release and nonfree/updates_testing Thanks, works perfect. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [RFC] x11-server-1.13
On 09/10/2012 07:31 PM, AL13N wrote: Op maandag 10 september 2012 15:02:37 schreef JA Magallón: On 09/08/2012 01:02 PM, Thierry Vignaud wrote: Hi x11-server-1.13 and rebuild drivers are in */update_testing. Tested on one desktop and two netbooks, so working intel, synaptics. Tested also on propietary nvidia driver, and broken. Someting in gdm or gnome-shell bombs in libpthread I wil request an update to the nvidia driver, because of this paragraph in new README: "Fixed a bug that caused pre-release versions of X.Org xserver 1.13 to crash when certain GLX operations were performed, such as when starting Firefox." Perhaps it is related. hey, i crash FF too, if it's on google maps with 3DGL enabled Checked with success. I installed manually the new driver and everything works again. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [RFC] x11-server-1.13
On 09/08/2012 01:02 PM, Thierry Vignaud wrote: Hi x11-server-1.13 and rebuild drivers are in */update_testing. Tested on one desktop and two netbooks, so working intel, synaptics. Tested also on propietary nvidia driver, and broken. Someting in gdm or gnome-shell bombs in libpthread I wil request an update to the nvidia driver, because of this paragraph in new README: "Fixed a bug that caused pre-release versions of X.Org xserver 1.13 to crash when certain GLX operations were performed, such as when starting Firefox." Perhaps it is related. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/updates_testing x11-driver-video-intel-2.20.6-2.mga3
On 09/06/2012 10:57 PM, tv wrote: Name: x11-driver-video-intel Relocations: (not relocatable) Version : 2.20.6Vendor: Mageia.Org Release : 2.mga3Build Date: Thu Sep 6 22:55:06 2012 Install Date: (not installed) Build Host: jonund.mageia.org Group : System/X11Source RPM: (none) Size: 1629756 License: MIT Signature : (none) Packager: tv URL : http://xorg.freedesktop.org Summary : X.org driver for Intel graphics controllers Description : x11-driver-video-intel is the X.org driver for Intel video chipsets. tv 2.20.6-2.mga3: + Revision: 288933 - rebuild for new xserver Input drivers still prevent my systems to upgrade, perhaps you should also rebuild them for this server... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release totem-3.5.90-2.mga3
On 09/05/2012 10:29 AM, blino wrote: Name: totemRelocations: (not relocatable) Version : 3.5.90Vendor: Mageia.Org Release : 2.mga3Build Date: Wed Sep 5 10:25:26 2012 I get this error message under Firefox: Couldn't load the '/usr/share/totem/mozilla-viewer.ui' interface. Invalid object type `TotemGlowButton' Make sure that Totem is properly installed. No idea about what is it... -- J.A. Magallon \ Winter is coming...
[Mageia-dev] systemd and NAMESPACE
Hi... I have a box (really a netbook) where all the systemd units that need PrivateTmp (rtkit, cups, ...) fail to start with a 226/NAMESPACE error. In other boxes (a couple desktops), all work fine. Any idea ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release clutter-1.11.14-2.mga3
On 09/04/2012 09:01 PM, ovitters wrote: Name: clutter Relocations: (not relocatable) Version : 1.11.14 Vendor: Mageia.Org Release : 2.mga3Build Date: Tue Sep 4 20:59:01 2012 Install Date: (not installed) Build Host: jonund.mageia.org Group : Graphics Source RPM: (none) Size: 4987732 License: LGPLv2+ Signature : (none) Packager: ovitters URL : http://clutter-project.org/ Summary : Software library for fast, visually rich GUIs Description : Clutter is an open source software library for creating fast, visually rich graphical user interfaces. The most obvious example of potential usage is in media center type applications. We hope however it can be used for a lot more. Clutter uses OpenGL (and soon optionally OpenGL ES) for rendering but with an API which hides the underlying GL complexity from the developer. The Clutter API is intended to be easy to use, efficient and flexible. ovitters 1.11.14-2.mga3: + Revision: 288211 - rebuild due to new gobject-introspection, changing API, breaking gnome-shell The i586 version seems to be lost in the ether, has not got to the mirrors... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Video/X/GL/gnome-shell issues?
On 07/27/2012 02:34 AM, Olav Vitters wrote: On Mon, Jul 23, 2012 at 10:23:11PM +0200, Olivier Blin wrote: I'm tempted to blame mesa to some degree, but no real evidence :) There was indeed a refcount fix (c3bc41011f9ffe648b7dd4915c6202b776cd1ab4) in mesa that I did not have in my snapshot. I still have huge issues running GNOME shell in Cauldron. At last !! I thought I was the only one... More from my side ;(... Problems: 1. Cannot skip in mplayer (said to be glibc problem) I think it is some problem im GTK sliders themselves. If I click in the middle of the slider to advance a full page, nothing happens, in several applications. For example, try rithmbox and push in the song time slider: it allows you to move the window (???) instead of skipping. Totem simply blocks. Curiously, in my case GnomeMplayer at least moves the slider, but media stops playing... 2. Sometimes when I use volume up/down keys, the popup window doesn't disappear. After that, the volume keys don't work anymore. I can still switch the volume using the volume icon in the top bar. I can click through the popup window, but cannot see what is underneath it (really annoying). The volume key is handled by gnome-settings-daemon. 3. Switching between apps or workspaces using alt+tab, alt+~, ctrl+alt+up/down makes GNOME shell hang. I'm currently avoiding workspaces all together, because when I switch between workspaces, GNOME shell can easily hang within 5 minutes. 4. If I click on the corner X button from decorator to close the window, many times gnome shell hangs, no visible crash, just freezes. 5. Many many mooore times nautilus just segfaults trying to open a window, or when I open a movie with totem. This was commented on a bug report, but I had not still the time to test the possible solution. 6. GDM behaves weird. On one box it does not allow me to log in, just accepts my password and sits forever showing the dimmed bullet passwd entry... On others, when I log out the system thinks I'm still there (commented also on other post, here for completeness...) Usually I can unfreeze X by stopping and starting "dm" (gdm). Sometimes I have to reboot or shut the machine down completely (usually when the plymouth/gdm background is garbled). If I kill X with ctrl-alt-bs, gdm restarts but still thinks I'm logged (no session menu for my user). I've updated all the packages, running the latest kernel, latest pulseaudio, etc. As driver, I use nouveau driver. I'm guessing glibc is to blame, but not sure. I use nvidia and intel on other boxes, so I think it is not driver related. I'll try and use strace on GNOME shell + gnome-settings-daemon (volume keys) next time one of them hangs again. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] wayland in cauldron (was Re: [changelog] [RPM] cauldron core/release weston-0.95.0-1.mga3)
On 07/25/2012 12:46 PM, Colin Guthrie wrote: 'Twas brillig, and Olivier Blin at 25/07/12 00:00 did gyre and gimble: blino writes: Name: weston Relocations: (not relocatable) Version : 0.95.0Vendor: Mageia.Org Release : 1.mga3Build Date: Tue Jul 24 23:38:30 2012 URL : http://wayland.freedesktop.org/ Summary : The Weston Wayland Compositor FYI, this release allows to test weston (the reference wayland compositor) oustide of X11 by running "weston-launch". This will spawn the compositor in a new tty. Beware that this one is still experimental, and might crash your X11 session. You can also try it inside of a X11 session, which seems to be safer, by running "weston". For now, the package is not much useful, only a sample "weston-terminal" application is provided, but I plan to enable more demo wayland clients in our weston package. I am also looking at enabling the wayland backend in gtk+3, which would enable far more applications to run in wayland (provided that they don't use gdk-x11). It doesn't work for me under nvidia propietary driver: werewolf:~> weston Date: 2012-07-25 CEST [16:48:50.407] weston 0.95.0 http://wayland.freedesktop.org/ Bug reports to: https://bugs.freedesktop.org/enter_bug.cgi?product=weston Build: [16:48:50.407] OS: Linux, 3.4.6-desktop-1.mga3, #1 SMP Fri Jul 20 18:40:37 UTC 2012, x86_64 couldn't open /home/magallon/.config/weston.ini [16:48:50.407] Loading module '/usr/lib64/weston/x11-backend.so' [16:48:50.429] initializing x11 backend couldn't open /home/magallon/.config/weston.ini libEGL warning: DRI2: failed to authenticate [16:48:50.494] EGL version: 1.4 (DRI2) [16:48:50.494] EGL vendor: Mesa Project [16:48:50.494] EGL client APIs: OpenGL OpenGL_ES OpenGL_ES2 [16:48:50.494] EGL extensions: EGL_KHR_surfaceless_gles1 EGL_KHR_surfaceless_gles2 EGL_KHR_surfaceless_opengl [16:48:50.494] GL version: OpenGL ES 2.0 Mesa 8.1-devel [16:48:50.494] GLSL version: OpenGL ES GLSL ES 1.0.16 [16:48:50.494] GL vendor: VMware, Inc. [16:48:50.494] GL renderer: Gallium 0.4 on llvmpipe (LLVM 0x300) [16:48:50.494] GL extensions: GL_EXT_blend_minmax GL_EXT_multi_draw_arrays GL_EXT_texture_compression_dxt1 GL_EXT_texture_format_BGRA GL_OES_compressed_ETC1_RGB8_texture GL_OES_depth24 GL_OES_element_index_uint GL_OES_fbo_render_mipmap GL_OES_mapbuffer GL_OES_rgb8_rgba8 GL_OES_standard_derivatives GL_OES_stencil8 GL_OES_texture_3D GL_OES_texture_npot GL_OES_EGL_image GL_OES_depth_texture GL_OES_packed_depth_stencil GL_EXT_texture_type_2_10_10_10_REV GL_EXT_read_format_bgra GL_NV_fbo_color_attachments GL_OES_EGL_image_external GL_EXT_texture_rg GL_EXT_unpack_subimage GL_NV_draw_buffers GL_NV_read_buffer [16:48:50.519] XCB-XKB not available during build [16:48:50.541] x11 output 1024x640, window id 41943048 [16:48:50.541] Loading module '/usr/lib64/weston/desktop-shell.so' couldn't open /home/magallon/.config/weston.ini [16:48:50.543] libwayland: using socket /run/user/3001/wayland-0 failed to get cairo egl argb device failed to create display: No such file or directory [16:48:50.663] libwayland: disconnect from client 0x1ab8a90 [16:48:50.663] weston-desktop-shell died, respawning... failed to get cairo egl argb device failed to create display: No such file or directory [16:48:50.769] libwayland: disconnect from client 0x1ca5d90 [16:48:50.769] weston-desktop-shell died, respawning... failed to get cairo egl argb device failed to create display: No such file or directory [16:48:50.877] libwayland: disconnect from client 0x1cdd5b0 [16:48:50.878] weston-desktop-shell died, respawning... failed to get cairo egl argb device failed to create display: No such file or directory [16:48:50.987] libwayland: disconnect from client 0x1ab8a90 [16:48:50.987] weston-desktop-shell died, respawning... failed to get cairo egl argb device failed to create display: No such file or directory [16:48:51.099] libwayland: disconnect from client 0x1ca5d90 [16:48:51.099] weston-desktop-shell died, respawning... failed to get cairo egl argb device failed to create display: No such file or directory [16:48:51.205] libwayland: disconnect from client 0x1cdd5b0 [16:48:51.205] weston-desktop-shell died, giving up. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release dconf-0.13.4-4.mga3
On 2012.07.19, at 17:59, Colin Guthrie wrote: > 'Twas brillig, and Colin Guthrie at 19/07/12 16:41 did gyre and gimble: >> 'Twas brillig, and JA Magallón at 19/07/12 16:30 did gyre and gimble: >>> >>> On 2012.07.19, at 06:45, fwang wrote: >>> >>>> Name: dconfRelocations: (not relocatable) >>>> Version : 0.13.4Vendor: Mageia.Org >>>> Release : 4.mga3Build Date: Thu Jul 19 >>>> 06:40:08 2012 >>> >>> There is a problem starting gdm (and I don't know if others) because >>> of gnome-session: >>> >>> gnome-session[2011]: CRITICAL: unable to create file '/run/user/493/dconf': >>> Is a directory. dconf will not work properly. >>> >>> Patch/rebuild needed ? >> >> Yeah I saw this too. If you rmdir the directory, it is just recreated >> automatically. >> >> I know gdm does strange things with dconf to inject it's data in so it >> might even need patching. Worth talking to Ray (halfline) upstream maybe? > > Actually there is a fix from upstream for it in git master. I applied it > here and it seems to work OK so just submitted a new build of gdm. > wow, that's speed ! THX !! -- J.A. Magallon \ Software is like sex: \ It's better when it's free
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release dconf-0.13.4-4.mga3
On 2012.07.19, at 06:45, fwang wrote: > Name: dconfRelocations: (not relocatable) > Version : 0.13.4Vendor: Mageia.Org > Release : 4.mga3Build Date: Thu Jul 19 06:40:08 > 2012 > Install Date: (not installed) Build Host: ecosse.mageia.org > Group : System/Libraries Source RPM: (none) > Size: 243577 License: LGPLv2+ > Signature : (none) > Packager: fwang > URL : http://live.gnome.org/dconf > Summary : Configuration backend for Glib > Description : > This is a configuration backend for Glib's GSettings and part of GNOME 3.0. > > fwang 0.13.4-4.mga3: > + Revision: 272487 > - fix giolibname > - more conflicts should be added There is a problem starting gdm (and I don't know if others) because of gnome-session: gnome-session[2011]: CRITICAL: unable to create file '/run/user/493/dconf': Is a directory. dconf will not work properly. Patch/rebuild needed ? -- J.A. Magallon \ Software is like sex: \ It's better when it's free
Re: [Mageia-dev] gdm doesnt seams to work
On 07/16/2012 12:03 PM, Simple w wrote: Hi, I have updated cauldron and also installed gdm, and runned MCC->Boot->Set up display managerand choose GDM, but when i logged out i got an image saying an error occurred and that i should contact the system administrator. I have more strange behaviour for GDM. I log in, log out, and can not log in again as myself. No session menu in GDM if I choose my name. Other users can log in, but then it happens again for them. More, if I log in and out, and try to restart the box from GDM, it tells me that other users are logged and it needs my password to reboot. It looks like the session is somehow not ended, and GDM thinks I am still logged. It happens in all the boxes I have running cauldron. Any ideas ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Help with libgcc1
On 07/16/2012 01:34 AM, Colin Guthrie wrote: Hi, On my system I see the following: [colin@jimmy gcc]$ rpm -ql libgcc1 /lib/libgcc_s-4.7.1.so.1 /lib/libgcc_s.so.1 /lib64/libgcc_s-4.7.1.so.1 /lib64/libgcc_s.so.1 /usr/lib/libgcc_s.so.1 As you can see this results in a conflict between these two files: /lib/libgcc_s.so.1 /usr/lib/libgcc_s.so.1 They ultimately (via some strange symlinks) point at the same file so it can be dropped, but I cannot, for the life of me see where the symlink: /usr/lib/libgcc_s.so.1 Is created in the spec. I can see a file of that name being specifically moved out the way but I cannot see how it is then recreated with that name again! Can anyone see in the gcc spec where this file is created? ldconfig ? I tried renaming mine, and ldconfig recreated it. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] References to old udev services
On 06/30/2012 05:17 PM, Colin Guthrie wrote: 'Twas brillig, and Colin Guthrie at 30/06/12 11:38 did gyre and gimble: 'Twas brillig, and JA Magallón at 29/06/12 23:38 did gyre and gimble: Hi... I post it here as it applies to various packages. Some systemd files still reference old udev services/targets, even some in systemd itself: werewolf:/lib/systemd/system# grep -r "=udev.*\.service" * fedora-wait-storage.service:After=udev-settle.service fedora-wait-storage.service:Wants=udev.service udev-settle.service mandriva-save-dmesg.service:After=udev-settle.service local-fs.target systemd-udev-settle.service:After=udev-trigger.service werewolf:/lib/systemd/system# grep -r "=udev.*\.target" * libvirtd.service:After=udev.target mandriva-save-dmesg.service:After=udev-settle.service local-fs.target Should not these change to refer to systemd-udev-* ? Yup, I'll fix them up, but I want to rename systemd-udev* to systemd-udevd* as this better fits the other upstream daemon services. I've discussed with upstream so a patch will be submitted upstream too :) These should all be fixed in the next build. One missing :) werewolf:/lib/systemd/system# grep -r "=udev-.*" * systemd-udev-settle.service:After=udev-trigger.service werewolf:/lib/systemd/system# rpm -qf systemd-udev-settle.service systemd-units-185-2.mga3 werewolf:/lib/systemd/system# ls *udev-trig* systemd-udev-trigger.service -- J.A. Magallon \ Winter is coming...
[Mageia-dev] References to old udev services
Hi... I post it here as it applies to various packages. Some systemd files still reference old udev services/targets, even some in systemd itself: werewolf:/lib/systemd/system# grep -r "=udev.*\.service" * fedora-wait-storage.service:After=udev-settle.service fedora-wait-storage.service:Wants=udev.service udev-settle.service mandriva-save-dmesg.service:After=udev-settle.service local-fs.target systemd-udev-settle.service:After=udev-trigger.service werewolf:/lib/systemd/system# grep -r "=udev.*\.target" * libvirtd.service:After=udev.target mandriva-save-dmesg.service:After=udev-settle.service local-fs.target Should not these change to refer to systemd-udev-* ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Broken xulrunner
On 06/29/2012 12:46 PM, Sander Lepik wrote: 29.06.2012 13:14, JA Magallón kirjutas: On 06/28/2012 05:12 PM, David Walser wrote: The QA team has been testing the update for Firefox 10.0.5 and has run into problems with xulrunner. Trying to test it through either perl-Gtk-MozEmbed or gnome-python-gtkmozembed which have both been rebuilt against xulrunner 10.0.5, segfaults are happening. Claire found a reference to an old Mandriva bug that may be relevant: https://qa.mandriva.com/show_bug.cgi?id=30801 See our bugs for the FF update for more: https://bugs.mageia.org/show_bug.cgi?id=6389 https://bugs.mageia.org/show_bug.cgi?id=6548 Can this be fixed, or is all moz/gecko embedding hopelessly broken forever? Not sure it is xulrunner to blame. I am getting segfaults on nautilus and in CUPS's pdftops (the one in /usr/lib64/cups..., not that in /usr/bin), so currently I can't print. I am waiting to test in other boxes to fill a bug report... It looks like bugs in new glibc/gcc or bugs in the apps that surface due to new libs/toolchain... David's message is clearly not about cauldron. Why hijack thread like that?! Oops, sorry... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Broken xulrunner
On 06/28/2012 05:12 PM, David Walser wrote: The QA team has been testing the update for Firefox 10.0.5 and has run into problems with xulrunner. Trying to test it through either perl-Gtk-MozEmbed or gnome-python-gtkmozembed which have both been rebuilt against xulrunner 10.0.5, segfaults are happening. Claire found a reference to an old Mandriva bug that may be relevant: https://qa.mandriva.com/show_bug.cgi?id=30801 See our bugs for the FF update for more: https://bugs.mageia.org/show_bug.cgi?id=6389 https://bugs.mageia.org/show_bug.cgi?id=6548 Can this be fixed, or is all moz/gecko embedding hopelessly broken forever? Not sure it is xulrunner to blame. I am getting segfaults on nautilus and in CUPS's pdftops (the one in /usr/lib64/cups..., not that in /usr/bin), so currently I can't print. I am waiting to test in other boxes to fill a bug report... It looks like bugs in new glibc/gcc or bugs in the apps that surface due to new libs/toolchain... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] ANN: GCC-4.7.1 landing on Cauldron
On 2012.06.27, at 13:58, Thomas Backlund wrote: > Funda Wang skrev 27.6.2012 14:48: >> Will there be rebuilding robot for cauldron so that we could detect >> build failures due to new tool chain? >> > > Not yet. > > In order to avoid a dual rebuild we will wait for GLIBC 2.16 > to land on cauldron too. > > GLIBC 2.16 is scheduled to be released on first week of july. > > The other thing I haven't decided on yet is if we should upgrade > binutils to 2.22.52.0.4 (wich is a beta) to get better support > for the new x32 "arch" or wait for 2.23 to be released... > > Of course if someone want to set up a local rebuild set to test > already, and report breakages already that's ok to... > As always happens with CUDA, new gcc breaks CUDA. I will post a patch in a bug report against CUDA. -- J.A. Magallon \ Software is like sex: \ It's better when it's free
Re: [Mageia-dev] ANN: For the brave. systemd v185 in cauldron updates_testing
On 06/18/2012 11:28 PM, JA Magallón wrote: Hi Sorry for the delay, I've been busy with end-of-term festivals in my children' schools ;)... - keyboard layout under Gnome got reset to English (from Spanish). Changed manually under system settings and fine again. One addtional problem is that GDM starts with english keyboard which is a problem for passwords with non alpha-num chars... :). Interesting. Does this new setting survive a reboot? Does the new mapping work at the system level (i.e. is the layout respected in GDM)? Yep, the setting survives a reboot for the user that changed it, but as I said above it does not change GDM settings. I did not try to do it with system tools, just user preferences. And keymap is OK in virtual consoles. Do you have a /etc/vconsole.conf file? Nop, I have not. I think system settings were in /etc/sysconfig/keyboard. OK, so I don't really understand why it's broken. I think it's maybe a regression in that X is no longer inheriting settings from the console, but I think longer term this is a valid thing (maybe there will not be a console to inherit from in the future?) So, it seems that systemd-localed will write an Xorg snippet into /etc/X11/xorg.conf.d/00-keyboard.conf when the keyboard is set. Thus your setting of it via the Gnome System Settings will have written this file (can you confirm it exists?) and it will remain there for the remaining boots. This file does not exist in my box, I just used user prefs to change the map, not sytem ones. Longer term, we need to make sure the installer runs this daemon for fresh installs. And perhaps we should run it on upgrade automatically too to ensure it's written as needed. I will try to write an /etc/vconsole.conf and see if it works. Perhaps the fix is just to write it from /etc/syscofig/keyboard... This file did not work for GDM. So i checked systemd-localed, it was not started and started it manually, and then copied my settings to system in gnome settings tool. That made the file in xorg.conf.d appear, and GDM was OK. Perhaps udev/dbus do not talk to X server anymore about keyboard layout. My settiongs in xorg.conf were: Section "InputDevice" Identifier "keyboard" Driver "evdev" #Option "XkbLayout" "es" #Option "XkbModel" "pc105" #Option "XkbOptions" "" EndSection File generated by localed is: # Read and parsed by systemd-localed. It's probably wise not to edit this file # manually too freely. Section "InputClass" Identifier "system-keyboard" MatchIsKeyboard "on" Option "XkbLayout" "es" EndSection -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] ANN: For the brave. systemd v185 in cauldron updates_testing
Hi Sorry for the delay, I've been busy with end-of-term festivals in my children' schools ;)... - keyboard layout under Gnome got reset to English (from Spanish). Changed manually under system settings and fine again. One addtional problem is that GDM starts with english keyboard which is a problem for passwords with non alpha-num chars... :). Interesting. Does this new setting survive a reboot? Does the new mapping work at the system level (i.e. is the layout respected in GDM)? Yep, the setting survives a reboot for the user that changed it, but as I said above it does not change GDM settings. I did not try to do it with system tools, just user preferences. And keymap is OK in virtual consoles. Do you have a /etc/vconsole.conf file? Nop, I have not. I think system settings were in /etc/sysconfig/keyboard. OK, so I don't really understand why it's broken. I think it's maybe a regression in that X is no longer inheriting settings from the console, but I think longer term this is a valid thing (maybe there will not be a console to inherit from in the future?) So, it seems that systemd-localed will write an Xorg snippet into /etc/X11/xorg.conf.d/00-keyboard.conf when the keyboard is set. Thus your setting of it via the Gnome System Settings will have written this file (can you confirm it exists?) and it will remain there for the remaining boots. This file does not exist in my box, I just used user prefs to change the map, not sytem ones. Longer term, we need to make sure the installer runs this daemon for fresh installs. And perhaps we should run it on upgrade automatically too to ensure it's written as needed. I will try to write an /etc/vconsole.conf and see if it works. Perhaps the fix is just to write it from /etc/syscofig/keyboard... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] ANN: For the brave. systemd v185 in cauldron updates_testing
On 06/18/2012 01:15 AM, Colin Guthrie wrote: Hi, I've just submitted systemd v185 to updates_testing. There are several important changes worth noting: 1. "systemd-sysvinit" has been merged into the main "systemd" package thus effectively making systemd the only choice for booting. 2. "systemd" package now contains systemd-udevd binary thus replacing the separate udev package (I see no reason to keep the binaries in separate packages). 3. The libudev major number has been bumped so several packages will need rebuilt when this hits main, but so far I've only rebuilt lvm2 for updates_testing. This seems to be enough for basic testing (I have LVM here). 4. Due to the new udevd path, dracut needed updating too. I've updated this and submitted to release (as it's backwards compatible). At a later date I'll update to latest dracut. Please ensure you are capable of restoring your system in the event of things messing up. This will mean booting and manually downgrading packages and reinstalling the older udev package. If all is well, this will go into cauldron at the end of the week. It seems to work ok in my netbook, just two minor problems: - in install: 7/28: systemd # /var/tmp/rpm-tmp.bM0Lmo: line 5: /usr/bin/systemctl: No such file or directory in post-install, systemctl is already moved to /bin - keyboard layout under Gnome got reset to English (from Spanish). Changed manually under system settings and fine again. If I find anything more, I'll tell here... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release nfs-utils-1.2.6-2.mga3
On 06/12/2012 02:29 AM, JA Magallón wrote: On 06/11/2012 10:22 PM, guillomovitch wrote: Name: nfs-utilsRelocations: (not relocatable) Version : 1.2.6 Vendor: Mageia.Org Release : 2.mga3Build Date: Mon Jun 11 22:21:14 2012 Install Date: (not installed) Build Host: ecosse.mageia.org Group : Networking/Other Source RPM: (none) Size: 787418 License: GPL Signature : (none) Packager: guillomovitch URL : http://sourceforge.net/projects/nfs/ Summary : The utilities for Linux NFS server Description : This package provides various programs needed for NFS support on server. guillomovitch 1:1.2.6-2.mga3: + Revision: 259882 - add exports.d directory - use /var/lib/nfs/statd for statd, instead of /var/lib/nfs, as in fedora - add modprobe config file to alias 'nfs4' to 'nfs' - add systemd support - drop sysinit support - merge client and server package, as per redhat setup Some weird things happen with this and systemd... I had nfs server active. After update and reboot, I had no nfs server running. What do I have to add in systemd to get them working ? I got it by: systemctl enable nfs.target (still had no server...) systemctl enable nfs-server.service Are those the correct top level units ? TIA Ah, and after reboot systemd still shows nfs-common.service: werewolf:/etc/systemd# systemctl --all | grep nfs proc-fs-nfsd.mountloaded active mounted RPC Pipe File System nfs-common.serviceerror inactive dead nfs-common.service nfs-idmap.service loaded active running NFSv4 ID-name mapping daemon nfs-mountd.serviceloaded active running NFS Mount Daemon nfs-rquotad.service loaded inactive dead NFS Remote Quota Server nfs-server.serviceloaded active exitedNFS Server nfs.targetloaded active activeNetwork File System Server werewolf:/etc/systemd# systemctl status nfs-common.service nfs-common.service Loaded: error (Reason: No such file or directory) Active: inactive (dead) since Tue, 12 Jun 2012 01:30:45 +0200; 1h 5min ago -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release nfs-utils-1.2.6-2.mga3
On 06/11/2012 10:22 PM, guillomovitch wrote: Name: nfs-utilsRelocations: (not relocatable) Version : 1.2.6 Vendor: Mageia.Org Release : 2.mga3Build Date: Mon Jun 11 22:21:14 2012 Install Date: (not installed) Build Host: ecosse.mageia.org Group : Networking/Other Source RPM: (none) Size: 787418 License: GPL Signature : (none) Packager: guillomovitch URL : http://sourceforge.net/projects/nfs/ Summary : The utilities for Linux NFS server Description : This package provides various programs needed for NFS support on server. guillomovitch 1:1.2.6-2.mga3: + Revision: 259882 - add exports.d directory - use /var/lib/nfs/statd for statd, instead of /var/lib/nfs, as in fedora - add modprobe config file to alias 'nfs4' to 'nfs' - add systemd support - drop sysinit support - merge client and server package, as per redhat setup Some weird things happen with this and systemd... I had nfs server active. After update and reboot, I had no nfs server running. What do I have to add in systemd to get them working ? I got it by: systemctl enable nfs.target (still had no server...) systemctl enable nfs-server.service Are those the correct top level units ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Mageia 3 specifications
On 06/08/2012 11:12 PM, zezinho wrote: Em 06-06-2012 08:22, AL13N escreveu: in any case, imho it's too soon for us to drop drakxtools/mganetapplet/if-cfg* (if ever). let's retalk about this for mga4. +1 I don't mean dropping them, just adaptimg them (except ifcfg-*) It looks much easier to write an independent file and drop it in a folder than to programatically edit ifcfg-*. Specifically: - make drakxtools read/write NM config files. People can use drakx* or the DE tool at their will. - make mganetapplet live and use it for desktops that don't have an specific applet. In Gnome, nm-applet fits better than mga one, and I suppose it happens the same for KDE. Probably there is no specfic applet for others. - if NM supports bonding and VPN, you can almost kill ifcfg-* files. That is the future. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release openssh-6.0p1-1.mga3
On 2012.06.07, at 12:17, Thomas Backlund wrote: > 07.06.2012 13:02, JA Magallón skrev: >> >> On 2012.06.07, at 10:09, guillomovitch wrote: >> >>> Name: openssh Relocations: (not relocatable) >>> Version : 6.0p1 Vendor: Mageia.Org >> >> I have a problem with ssh updates... >> Each time openssh-server is updated, it restarts immediately even in middle >> of an urpmi group update, so you loose the connection to the box and >> urpmi breaks in the middle, so uninstalls are not processed: > > https://wiki.mageia.org/en/Mageia_2_Release_Notes#SSH_daemon > Thanks !! -- J.A. Magallon \ Software is like sex: \ It's better when it's free
Re: [Mageia-dev] [changelog] [RPM] cauldron core/release openssh-6.0p1-1.mga3
On 2012.06.07, at 10:09, guillomovitch wrote: > Name: openssh Relocations: (not relocatable) > Version : 6.0p1 Vendor: Mageia.Org I have a problem with ssh updates... Each time openssh-server is updated, it restarts immediately even in middle of an urpmi group update, so you loose the connection to the box and urpmi breaks in the middle, so uninstalls are not processed: cicely:~# urpmi openssh-server To satisfy dependencies, the following packages are going to be installed: PackageVersion Release Arch (medium "Core Release") openssh6.0p11.mga3x86_64 openssh-askpass6.0p11.mga3x86_64 openssh-askpass-gnome 6.0p11.mga3x86_64 openssh-clients6.0p11.mga3x86_64 openssh-server 6.0p11.mga3x86_64 32KB of additional disk space will be used. 813KB of packages will be retrieved. Proceed with the installation of the 5 packages? (Y/n) http://mirrors.mageia.org/api/mageia.cauldron.x86_64.list: media/core/release/openssh-server-6.0p1-1.mga3.x86_64.rpm http://mirrors.mageia.org/api/mageia.cauldron.x86_64.list: media/core/release/openssh-askpass-6.0p1-1.mga3.x86_64.rpm http://mirrors.mageia.org/api/mageia.cauldron.x86_64.list: media/core/release/openssh-askpass-gnome-6.0p1-1.mga3.x86_64.rpm http://mirrors.mageia.org/api/mageia.cauldron.x86_64.list: media/core/release/openssh-clients-6.0p1-1.mga3.x86_64.rpm http://mirrors.mageia.org/api/mageia.cauldron.x86_64.list: media/core/release/openssh-6.0p1-1.mga3.x86_64.rpm installing openssh-askpass-6.0p1-1.mga3.x86_64.rpm openssh-server-6.0p1-1.mga3.x86_64.rpm openssh-askpass-gnome-6.0p1-1.mga3.x86_64.rpm openssh-6.0p1-1.mga3.x86_64.rpm openssh-clients-6.0p1-1.mga3.x86_64.rpm from /var/cache/urpmi/rpms Preparing... ## 1/5: openssh ## 2/5: openssh-clients ## 3/5: openssh-server ## Connection to cicely closed by remote host. Connection to cicely closed. belly:~> ssh root@cicely Last login: Thu Jun 7 11:54:30 2012 from belly.cps.unizar.es cicely:~# rpm -qa openssh* openssh-5.9p1-5.mga2 openssh-6.0p1-1.mga3 openssh-askpass-common-5.9p1-5.mga2 openssh-server-5.9p1-5.mga2 openssh-askpass-5.9p1-5.mga2 openssh-clients-6.0p1-1.mga3 openssh-clients-5.9p1-5.mga2 openssh-askpass-gnome-5.9p1-5.mga2 openssh-server-6.0p1-1.mga3 Previuosly this was not the case, I don't know if it is a problem in owns openssh %post* or is related to systemd, that realizes the binary change and does something. Restarting sshd under sysvinit does not kill active connections, that is the case on a CentOS box and was in Mageia under sysvinit. This looks like a serious issue...mostly for remote admin. -- J.A. Magallon \ Software is like sex: \ It's better when it's free
Re: [Mageia-dev] Mageia 3 specifications
On 06/06/2012 02:26 AM, David W. Hodgins wrote: On Tue, 05 Jun 2012 17:41:15 -0400, Olav Vitters wrote: But ideally it would be nice if there was only one way to setup a network.. and for me that means NetworkManager :). But drakxtools was more advanced IIRC, so first NetworkManager needs to get more advanced, more documentation, etc. If networkmanager can be installed without having to install half of gnome, and adds things like text only interfaces, then having it replace drakx-net makes sense to me. If a non-gui minimall install can not be done, simply because it requires gnome, then it can't replace drakx-net, in my opinion. This is like MCC and GNOME CC, network-manager is just a CLI tool, you have a Gnome tool to feed it and you can also make drakx-net feed NM with connection description files. Anyways, there is a point with NM that I find also useful, it can configure wifi interfaces at the user level, no need to store them in system hierarchy. drakx-* are more system-oriented. Think in shared laptops over enterprise wifis... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Mageia 3 specifications
On 06/05/2012 11:41 PM, Olav Vitters wrote: On Tue, Jun 05, 2012 at 10:44:31PM +0200, Guillaume Rousse wrote: Le 05/06/2012 16:20, Olav Vitters a écrit : Also, with latest NetworkManager, it seems you can do pretty advanced network configurations (lacks UI). The UI is planned for GNOME 3.6. However, seems that loads of Mageia users have issues with NetworkManager, so not sure what to do. Perhaps safer to wait a year and hope that other distributions solve the bugs? Approximatively 50% of those problems come from the coexistence with drakxtools and sysinit, conflicting for the control of the interface, rather than actual problems in NM code. Waiting for them to magically disapear won't help much... Ah ok, but shouldn't that all be fixed now that the patch is back to make NetworkManager properly ignore stuff that it should ignore? I tink it is _lot_ easier even for drakxtools to write a file into /etc/NetworkManager/system-connections that to mess with ifcfg-* files. And NM is the only sane way to have, in my case, 4 or 5 WiFi configurations for the same interface and let it pick the right one depending on my location. It is just marvelous, and without any ifcfg-* file rewriting on the fly. And I suppose it can also be done, but I never managed to do without NM: let my linux in the far corner of the house boot and connect to wifi without any user logged in... So in my case, a big +1 for NM and kill all sysvinit files for network. I acknowledge that I never tried to do bonding with NM, for example... meaning: only look for USE_NM/NM_CONTROLLED (I forgot which), not demand more than that. But ideally it would be nice if there was only one way to setup a network.. and for me that means NetworkManager :). But drakxtools was more advanced IIRC, so first NetworkManager needs to get more advanced, more documentation, etc. Cool possibilities with just NetworkManager: http://fedoraproject.org/wiki/Features/NMEnterpriseNetworking https://bugzilla.gnome.org/show_bug.cgi?id=677148: (all one bugzilla.gnome.org, seem duplicates, but there is a bug per product, so bug for maybe gnome-control-center, gnome-shell, etc) 677144: network: make sure bridged devices are handled 677145: network: make sure bridged devices are handled 677146: network: make sure bonded devices are handled 677147: network: make sure bonded devices are handled 677148: network: make sure vlans are handled 677149: network: make sure vlans are handled 677150: network: make sure ipoib is handled 677151: network: make sure ipoib is handled I had to lookup what ipoib (IP over infiniband) meant, article is here: http://www.ietf.org/wg/concluded/ipoib.html GNOME shell specific: https://bugzilla.gnome.org/show_bug.cgi?id=677142: multiple nics (NM can handle it, GNOME shell not) 677142: network: make sure multiple nics are handled ok 677143: network: make sure multiple nics are handled ok Note: The support is in NetworkManager. The GNOME bugs are to ensure it is shown/exposed in the UI properly. Not sure if available in stuff like KNetworkManager, but NM is not specific to GNOME. All bugs above are really server like things. I thought drakxtools still was better at setting up Wifi; and I think that is more important to get right than vlan/bridging/etc. -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Cinnamon Desktop
On 05/25/2012 09:10 AM, Olav Vitters wrote: On Thu, May 24, 2012 at 06:52:36PM -0600, fi...@linuxbsdos.com wrote: Is there a binary package for Cinnamon Desktop on Mageia? If not, is there an ongoing attempt to make it available? Same answer as previous, see thread: http://comments.gmane.org/gmane.linux.mageia.devel/10858 I took the wild road, and just installed the two rpms that are on the Fedora 17 repos pointed from the cinnamon site. They even work !!! I had to force the install due to some different names in requires, but binaries seem to work. There are two packages, one for the cinnamon shell itself, and other for a window manager called 'muffin'. I think it even can work with mutter or metacity. So I think the changes are minimal, and probably just building the package for Mageia and giving a session file for that would be enough. Just integration, no changes to anything else I think. It doesn't look bad. Just an intermediate point between Gnome3 and 2. And not so as Unity. -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Reminder: lsb-release
Hi... I have seen mageia-release files in my latest updates and they read alredy as official mageia 2. Just to remind you to update also lsb-release package, although I suppose is on its way... -- J.A. Magallon \ Winter is coming...
[Mageia-dev] GNOME updates
Looking into a sound-juicer pending bug, I noticed that there are a bunch of gnome 3.4 updates (even without going to 3.5). 3.5 is out of scope for mga2, but perhaps we are still in time to get the minor/bugfixes updates. Some I realized: gtk+ 3.4.1 -> 3.4.2 glib 2.32.1 -> 2.32.2 (or 2.33.1) clutter 1.10.2 -> 1.10.4 (or 1.11.2, jump to 1.11 probably dangerous...) dconf 0.12.0 -> 0.12.1 gnome-contacts 3.4.0 -> 3.4.1 What do you think ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] RFT: x11-driver-video-intel-2.19.0-1.mga2
On 05/02/2012 10:31 PM, Sander Lepik wrote: 02.05.2012 13:00, Thomas Backlund kirjutas: And report in this thread success/failure. No problem in older hardware (i965,i915 and NM10). -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Freeze push: nvidia-cuda-toolkit
On 04/22/2012 11:56 PM, Thomas Backlund wrote: 23.04.2012 00:51, Dimitri skrev: Please push nvidia-cuda-toolkit 4.2.9. It adds support for the recently released Kepler (GTX 6XX) architecture. And more important, it has support for current gcc in Mageia. I had to tweak previous versions to get them to work with gcc-4.6 Submitted -- Thomas -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Mounting nfs4 shares
On 04/19/2012 08:16 PM, Anne Wilson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 I have been using lines such as this to mount nfs shares: 192.168.0.40:/Data1 /mnt/borg2_Data1 nfs4 rsize=8192,wsize=8192,timeo=14,intr mount -a Mount is denied because the NTFS volume is already exclusively opened. The volume may be already mounted, or another software may use it which could be identified for example by the help of the 'fuser' command. mount.nfs4: an incorrect mount option was specified Do I need to change these mount lines in some way? Do I need some extra package? (I've already installed nfs-utils-client) Shorewall should be allowing anything from the LAN. What else could I have missed? I'm using nfs4 between Mageia and CentOS boxes in both directions without problems. I use autofs, such like: auto.master: ... /- file:/etc/autofs/auto.global ... auto.global: ... /giga/nada -fstype=nfs4,fsc,soft,intr nada:/ ... cicely:~# df /giga/nada/media Filesystem Type Size Used Avail Use% Mounted on nada:/media/ nfs4 3.6T 2.1T 1.4T 61% /giga/nada/media cicely:~# grep nada /proc/mounts /etc/autofs/auto.global /giga/nada autofs rw,relatime,fd=12,pgrp=3871,timeout=300,minproto=5,maxproto=5,direct 0 0 nada:/ /giga/nada nfs4 rw,relatime,vers=4,rsize=524288,wsize=524288,namlen=255,soft,proto=tcp,timeo=600,retrans=2,sec=sys,clientaddr=155.210.155.200,minorversion=0,fsc,local_lock=none,addr=155.210.152.224 0 0 nada:/media/ /giga/nada/media nfs4 rw,relatime,vers=4,rsize=524288,wsize=524288,namlen=255,soft,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=155.210.155.200,minorversion=0,fsc,local_lock=none,addr=155.210.152.224 0 0 Can you show the output from 'showmount --exports ' ? Also 'rpcinfo -p ' ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Kernel issues with atheros wifi
On 04/17/2012 10:50 PM, Thomas Backlund wrote: 17.04.2012 23:39, JA Magallón skrev: Hi... I've been bitten by this bug: https://bbs.archlinux.org/viewtopic.php?id=139270&p=1 on my netbook. From what I read, ath9k bug is corrected in 3.3.2, and ath5k bug is still present and a couple patches are needed. Are we too late to have a patched 3.3.2 in Cauldron ? ath9k fix is in our kernel-3.3.1-2.mga2 New kernel will land on Cauldron after beta3 is out Thanks, good to know ;) ! -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Kernel issues with atheros wifi
Hi... I've been bitten by this bug: https://bbs.archlinux.org/viewtopic.php?id=139270&p=1 on my netbook. From what I read, ath9k bug is corrected in 3.3.2, and ath5k bug is still present and a couple patches are needed. Are we too late to have a patched 3.3.2 in Cauldron ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Problems with flash 11.2
On 03/31/2012 02:12 AM, JA Magallón wrote: Hi... Is anybody else seeing strange things with flash 11.2 ? In my nvidia boxes it looks like I'm always seeing avatars in pandora... Looks like red and blue channels are switched. Try this: http://www.youtube.com/results?search_query=ntsc+color+bars look at the thumbnail, and then to the video (lookin close to the two rightmost bars...). It doesn't happen on intel graphics. My system is 64bit and nvidia. Still have to try on 32bit with older nvidia card (7600GT). Somebody else ? There is a new version of flash player: 11.2.202.233 (vs 202.228 current in cauldron). I have been using it and: - the color problems are the same with HW-render + SW decode - the crashes seem to be gone with HW-Render + HW-Decode The HW accel really works in nVidia, I have been trying with http://www.youtube.com/watch?v=rE09rUdpB94 and at 1080p, full software puts plugin-container at 100-120 % CPU time, and full hardware leaves it under 10% most of time. So perhaps it is safe to ship an mms.cfg that disables all HW acceleration and put a README for nVidia users on how to do the trick in /etc/adobe/mms.cfg. Or maybe unconditionally enable HW decoding as it does nothing on intel graphics, I think... -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] plymouth gone nuts ?
On 04/03/2012 02:43 AM, Colin Guthrie wrote: 'Twas brillig, and JA Magallón at 03/04/12 01:13 did gyre and gimble: On 04/03/2012 02:07 AM, JA Magallón wrote: Hi... With latest updates, I noticed that GDM again started in tty2, and tty1 is still stuck on bootscreen/plymouth dots. I get a message on tty1 about "wait-for-playmouth-quit failed, see systemctl status...". And GDM starts on tty2. Something weird is happpening: cicely:~# systemctl --all --full | grep plym systemd-ask-password-plymouth.path loaded active waiting Forward Password Requests to Plymouth Directory Watch plymouth-quit-wait.service loaded failed failed Wait for Plymouth Boot Screen to Quit plymouth-quit.service loaded inactive dead Terminate Plymouth Boot Screen plymouth-read-write.service loaded inactive dead Tell Plymouth To Write Out Runtime Data plymouth-start.service loaded inactive dead Show Plymouth Boot Screen systemd-ask-password-plymouth.service loaded inactive dead Forward Password Requests to Plymouth cicely:~# ps -ef | grep plym root 161 1 0 01:56 ? 00:00:00 /bin/plymouthd --attach-to-session --pid-file /run/plymouth/pid root 29109 8671 0 02:05 pts/1 00:00:00 grep --color plym Any idea ? I have noticed this, plz correct me if I'm wrong: werewolf:/lib/systemd/system# grep After display-manager.service After=livesys-late.service systemd-user-sessions.service After=getty@tty1.service plymouth-quit.service werewolf:/lib/systemd/system# grep After plymouth-quit.service After=rc-local.service plymouth-start.service werewolf:/lib/systemd/system# grep After plymouth-quit-wait.service After=rc-local.service plymouth-start.service Should not all the chain be: display-manager.service -> after plymouth-quit-wait.service plymouth-quit-wait.service -> after plymouth-quit.service No. plymouth-quit-wait will simply hang until plymouth quits of it's own accrord. If you actively quit plymouth first via plymouth-quit then there is literally no point in running plymouth-quit-wait as it'll just exit immediately. Ah, thanks. I jus thought that 'plymouth quit' was asynchronous, and the -wait script just made sure to wait before launching DM... Also u're forgetting about the Conflicts= directives... display-manager.service:Conflicts=getty@tty1.service plymouth-quit.service display-manager specifically conflicts with plymouth-quit.service (as it internally quits plymouth or allows a smooth transition if appropriate). So the ordering is correct as it stands. Now the question is what is preventing plymouth from being quit? As I said, display-manager should do it for you (see the /etc/X11/prefdm script), or simply let GDM do it itself. I'll see what I can dig up. Col -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] plymouth gone nuts ?
On 04/03/2012 02:07 AM, JA Magallón wrote: Hi... With latest updates, I noticed that GDM again started in tty2, and tty1 is still stuck on bootscreen/plymouth dots. I get a message on tty1 about "wait-for-playmouth-quit failed, see systemctl status...". And GDM starts on tty2. Something weird is happpening: cicely:~# systemctl --all --full | grep plym systemd-ask-password-plymouth.path loaded active waiting Forward Password Requests to Plymouth Directory Watch plymouth-quit-wait.service loaded failed failed Wait for Plymouth Boot Screen to Quit plymouth-quit.service loaded inactive dead Terminate Plymouth Boot Screen plymouth-read-write.service loaded inactive dead Tell Plymouth To Write Out Runtime Data plymouth-start.service loaded inactive dead Show Plymouth Boot Screen systemd-ask-password-plymouth.service loaded inactive dead Forward Password Requests to Plymouth cicely:~# ps -ef | grep plym root 161 1 0 01:56 ? 00:00:00 /bin/plymouthd --attach-to-session --pid-file /run/plymouth/pid root 29109 8671 0 02:05 pts/1 00:00:00 grep --color plym Any idea ? I have noticed this, plz correct me if I'm wrong: werewolf:/lib/systemd/system# grep After display-manager.service After=livesys-late.service systemd-user-sessions.service After=getty@tty1.service plymouth-quit.service werewolf:/lib/systemd/system# grep After plymouth-quit.service After=rc-local.service plymouth-start.service werewolf:/lib/systemd/system# grep After plymouth-quit-wait.service After=rc-local.service plymouth-start.service Should not all the chain be: display-manager.service -> after plymouth-quit-wait.service plymouth-quit-wait.service -> after plymouth-quit.service Trying now... -- J.A. Magallon \ Winter is coming...
[Mageia-dev] plymouth gone nuts ?
Hi... With latest updates, I noticed that GDM again started in tty2, and tty1 is still stuck on bootscreen/plymouth dots. I get a message on tty1 about "wait-for-playmouth-quit failed, see systemctl status...". And GDM starts on tty2. Something weird is happpening: cicely:~# systemctl --all --full | grep plym systemd-ask-password-plymouth.path loaded active waiting Forward Password Requests to Plymouth Directory Watch plymouth-quit-wait.service loaded failed failedWait for Plymouth Boot Screen to Quit plymouth-quit.service loaded inactive dead Terminate Plymouth Boot Screen plymouth-read-write.service loaded inactive dead Tell Plymouth To Write Out Runtime Data plymouth-start.service loaded inactive dead Show Plymouth Boot Screen systemd-ask-password-plymouth.service loaded inactive dead Forward Password Requests to Plymouth cicely:~# ps -ef | grep plym root 161 1 0 01:56 ?00:00:00 /bin/plymouthd --attach-to-session --pid-file /run/plymouth/pid root 29109 8671 0 02:05 pts/100:00:00 grep --color plym Any idea ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Problems with flash 11.2
On 03/31/2012 08:43 AM, Dimitrios Glentadakis wrote: Στις 31/03/2012 08:38:19 γράψατε: Στις 31/03/2012 02:38:24 JA Magallón γράψατε: And, btw, to get hw video decoding I had to write a file in /etc/adobe/mms.cfg with EnableLinuxHWVideoDecode=1 Thanks, this solved the problem for me i had to reverse to disable hw acceleration because the embedded youtube videos in konqueror are invisible (in youtube it is ok) Yep, and flash also behaves mch more unstable, I'm having hangs I has never seen before. Crappy flash... BTW, how can I check if html5 has hardware support ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Problems with flash 11.2
On 03/31/2012 02:12 AM, JA Magallón wrote: Hi... Is anybody else seeing strange things with flash 11.2 ? In my nvidia boxes it looks like I'm always seeing avatars in pandora... Looks like red and blue channels are switched. Try this: http://www.youtube.com/results?search_query=ntsc+color+bars look at the thumbnail, and then to the video (lookin close to the two rightmost bars...). It doesn't happen on intel graphics. My system is 64bit and nvidia. Still have to try on 32bit with older nvidia card (7600GT). Somebody else ? Googling a bit more and with some experiments, I found the problem. Channels are swapped if you use accelerated video rendering but software decoding. If both are soft, or both accelerated there is no problem. And, btw, to get hw video decoding I had to write a file in /etc/adobe/mms.cfg with EnableLinuxHWVideoDecode=1 There are some more settings here http://wiki.debian.org/FlashPlayer and a fully commented file here http://sources.gentoo.org/cgi-bin/viewvc.cgi/gentoo-x86/www-plugins/adobe-flash/files/mms.cfg I don't know how useful are the rest, but this one really enables HW in my nvidia card. Perhaps it should be a good idea to add it to the rpm, after downloading binaries. If it does not cause trouble in other chipsets. Will try on intel... -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Problems with flash 11.2
Hi... Is anybody else seeing strange things with flash 11.2 ? In my nvidia boxes it looks like I'm always seeing avatars in pandora... Looks like red and blue channels are switched. Try this: http://www.youtube.com/results?search_query=ntsc+color+bars look at the thumbnail, and then to the video (lookin close to the two rightmost bars...). It doesn't happen on intel graphics. My system is 64bit and nvidia. Still have to try on 32bit with older nvidia card (7600GT). Somebody else ? -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] GNOME 3.3.92 submission
On 03/22/2012 12:11 AM, Olav Vitters wrote:
On Wed, Mar 21, 2012 at 04:57:47PM +0100, Olav Vitters wrote:
On Tue, Mar 20, 2012 at 07:21:44PM +0100, Olav Vitters wrote:
On Mon, Mar 19, 2012 at 10:27:49AM +0100, Olav Vitters wrote:
I'll try and ensure that at least the SPEC files are already updated
(via my that is triggered via ftp-release-list). However, I do expect
various submission failures, etc.
Updated list of packages which are hosted on 'ftp.gnome.org':
Submitting now in the order specified by "jhbuild list" (to avoid any
dependency problems). I'm using mgaqueue.sh. Never used that before, so
let's see :)
All modules have been submitted. A few failed, if someone could look at
those? The easy failures I tried to fix. I'll also try and look at them
tomorrow, but likely need help.
As we are in the way to gnome-3.4, should not task-gnome-{minimal}
renumbered to 3.4.0 ?
Also enhanced my mga-gnome script to:
1. Allow limiting the modules (to just GNOME Core+Apps)
2. Allow sorting the 'still to be built' according to predefined order
GNOME has "jhbuild", this has a list subcommand which gives you the
order in which packages need to be built. That output can be used by
mga-gnome to limit it to GNOME Core&Apps + sort it according to
jhbuild.
meaning:
mga-gnome -l ~/.packages packages --sort ~/.packages --diff --version
-l limits it to just the upstream names from a file
--sort sorts it according to where in the file the upstream name
appears
--
J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Too late for this ? (cairo 1.12)
On 03/26/2012 12:35 AM, Funda Wang wrote: I'm afraid yes, cairo is a base library for all gtk2/3 apps. We cannot do such a big version bump now. Well, I understand. Anyways, as src.rpm was pretty simple, I rebuilt it locally and works pretty fine. I think things got a bit smoother, but anyways it can be self suggestion. Just for the record, it built without any problem. I just replaced source tarballs and changed versions in spec, no lib major bump nor anything, so it replaced old one. And I found that there was a plf flag in the spec (to enable font aa), but we have no tainted build, I think. I have tainted repos enabled but urpmi just installs the standard version. 2012/3/26 JA Magallón: http://cairographics.org/news/cairo-1.12.0/ -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Too late for this ? (cairo 1.12)
http://cairographics.org/news/cairo-1.12.0/ -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Latest kernels vs IDE/libata
Hi all... I have a queston about an issue that should be decided by the 'thinkin minds' in the distro, that probably have have wider sight than me... I have some boxes with ide disks, that long tine ago were driven by the libata subsystem, as the pure PATA susbsytem is being slowly deprecated. Perhaps the kernel driver loading priorities have changed, perhaps before IDE was not builtin into kernel. But now, to get my IDE drove again by libata, I have to blacklist generic-ide and company. So I had to put a file ide.conf into modules.d like: blacklist ide-generic blacklist ide-pci-generic blacklist piix so ata_piix drives my ICH5 again. I think it is desirable, libata works much better than old IDE driver, all your drives are sdX, instead of mixing sdX for SATA and hdX for PATA, etc... Would it be desirable to have an ide-blacklist.conf file just to go migrating all working-ok-with-libata controllers to libata, and only leave pure PATA drivers for non supported oldies ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Can't install nvidia-cuda update
On 03/12/2012 09:30 PM, Dimitri Jakov wrote:
Should be fixed now. Please update and report your experience. Thanks!
Dimitri
In the mirrors there is only new nvidia-cuda-toolkit(-devel) packages,
but still require something not detected:
werewolf:~/in# rpm -U nvidia-cuda-toolkit-4.1.28-3.mga2.nonfree.x86_64.rpm
error: Failed dependencies:
libcuda.so.1()(64bit) is needed by
nvidia-cuda-toolkit-4.1.28-3.mga2.nonfree.x86_64
nvidia-cuda-toolkit = 4.1.28-1.mga2.nonfree is needed by (installed)
nvidia-cuda-toolkit-devel-4.1.28-1.mga2.nonfree.x86_64
werewolf:~/in# rpm -qf /usr/lib64/nvidia-current/libcuda.so.1
nvidia-current-cuda-opencl-295.20-2.mga2.nonfree
werewolf:~/in# rpm -q --provides
nvidia-current-cuda-opencl-295.20-2.mga2.nonfree
nvidia-current-cuda = 295.20-2.mga2.nonfree
nvidia-current-cuda-opencl = 295.20-2.mga2.nonfree
nvidia-current-cuda-opencl(x86-64) = 295.20-2.mga2.nonfree
The latter does not provide libcuda.so.1()(64bit). Should it be automatic
or not, it should show anyways in --provides ?
Perhaps the library location in /usr/lib{64}/nvidia-current makes rpm skip
the automatic provides for libs ?
TIA
--
J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] ANN: version freeze and what it means for Mageia 1 updates
On 03/12/2012 09:06 AM, Samuel Verschelde wrote: Hi, Just a quick reminder that the version and release of a package in cauldron must ALWAYS be higher (or equal) to that in Mageia 1. Otherwise we cannot I would explicitely add to this to check also updates, backports _and_ testing repos... guarantee upgrade from Mageia 1 to Mageia 2. Since we are in version freeze we must check this with more accuracy. Packagers, let's keep this (obvious) rule in mind (I'm not sure it's enforced in the buildsystem yet, but I may be wrong). QA Team, before validating an update, please check the version in cauldron. BTW : do we have a check.mageia.org page to spot packages for which this is not the case, if there are any? Maybe this reminder is not useful at all, but I think it's always better to send it rather than not. Best regards Samuel "Gives lessons and does nothing" Verschelde -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Can't install nvidia-cuda update
It looks like some package does not provide libcuda (the library itself is there): Some requested packages cannot be installed: nvidia-cuda-toolkit-4.1.28-1.mga2.nonfree.x86_64 (due to unsatisfied libcuda.so.1()(64bit)) nvidia-cuda-toolkit-devel-4.1.28-1.mga2.nonfree.x86_64 (due to unsatisfied libcuda.so.1) Continue installation anyway? (Y/n) werewolf:~# rpm -qf /usr/lib64/nvidia-current/libcuda.so.1 nvidia-current-cuda-opencl-295.20-2.mga2.nonfree werewolf:~# rpm -q --provides nvidia-current-cuda-opencl nvidia-current-cuda = 295.20-2.mga2.nonfree nvidia-current-cuda-opencl = 295.20-2.mga2.nonfree nvidia-current-cuda-opencl(x86-64) = 295.20-2.mga2.nonfree -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] agetty vs. mingetty
On 03/08/2012 03:17 PM, Colin Guthrie wrote: Hi, Since switching to systemd, the mingetty we used to default to in inittab has been replaced by agetty (the default choice upstream). The question is should we switch it back to mingetty? Lookin other things I also looked at agetty manual, and found a couple useful things: o Optionally displays an alternative issue file instead of /etc/issue. o Optionally does not ask for a login name. o Optionally invokes a non-standard login program instead of /bin/login. This looks ideal for the case where prefdm fails and you want to show a big barner about it on tty1, or even drop the user direcly into drakx text mode without any login. Perhaps it can be done in prefdm if launching DISPLAYMANAGER fails. So, if space or memory is not an issue, I will go for agetty in both systemd and inittab. An argument against mingetty is that the current support in systemd will automatically work over serial consoles. With embedded systems and such more common these days, this might actually be a common situation (for very small values of "common" of course!). From the man page of mingetty: mingetty is a minimal getty for use on virtual consoles. Unlike agetty(8), mingetty is not suitable for serial lines. I recommend using mgetty(8) for this purpose. So if we switch to mingetty in getty@.service we'll be sacrificing serial console support. Have people got strong opinions on this? Can anyone remember why we ended up defaulting to mingetty in the first place? Col PS I'm aware of some "strange" behaviour on the current gettys. When on a getty you can use the cursor keys and move the cursor around. Keys such as backspace don't seem to work if you mistype your username or password (I'm pretty sure this used to work, but maybe I'm misremembering?). This is slightly improved with mingetty - in that the cursor keys don't move the text entry area, but they do print non-printable characters that still can't be deleted. I'll try and make this nicer, but it's probably not the highest priority problem on my list :p -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] deprecate text installer
On 03/07/2012 09:34 PM, Maarten Vanraes wrote: considering https://bugs.mageia.org/show_bug.cgi?id=4724 and that it's likely not working in cauldron for months, and also considering it's beyond our perl guru jq and our drakx-guru tv, and considering it looks like google didn't help... and considering there's 4 or 5 text-installer bugs i wanted to test out... perhaps we should consider in fact that text installer does not work... after all, if we get this fixed, there's still the other text-installer bugs... maybe it's premature for me, but considering this is version freeze... we're getting closer to the end... WDYT? This made me think of systems managed through IPMI or ALOM (yup, that later is only a Sun thing, but to get the idea...), and I don't think this interfaces are capabe of VGA graphics. That is not a big deal, but it forces you to attach a monitor to the VGA to install. BTW, will the text installer ever work through IMPI ? Is it even possible that kinkd of install ? -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Two versions of udisks
Hi... The latest update in Cauldron pulled udisks-1.0.4, and I had also installed udisks2. I sounded strange to have both installed, but perhaps they complemented... I launched both monitors and both detect USB thumb insertion, and SD card insertions. And old udisks is only required by: one:~# rpm -q --whatrequires udisks perl-Hal-Cdroms-0.04-2.mga2 one:~# rpm -q --whatrequires udisks2 gnome-disk-utility-3.3.92-1.mga2 gvfs-1.11.4-1.mga2 one:~# rpm -q --whatrequires perl-Hal-Cdroms no package requires perl-Hal-Cdroms I removed perl-Hal-Cdroms and udisks and everything seems to work. What is needed the perl module for ? Can it be updated to require udisks2 ? TIA -- J.A. Magallon \ Winter is coming...
[Mageia-dev] udisks moved to usr ?
Hi... Just a notice, binaries from udisks have moved from /lib to /usr/lib (from 1.90 to 1.92). Sure they are not needed before /usr gets mounted ? TIA -- J.A. Magallon \ Winter is coming...
Re: [Mageia-dev] Update blocked on i586
On 02/25/2012 08:29 AM, Jani Välimaa wrote: Should be fixed now. Thanks, it worked. -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Update blocked on i586
Hi.. I have updated Cauldron on several boxes, all x86-64 went ok and now I even got rid of libcogl7, everything updated. But on i586 update is sill blocked for some packages, as it looks like gnome-themes-standard-3.3.90.1 for i586 got lost in the ether: Some requested packages cannot be installed: adwaita-cursor-theme-3.3.90.1-1.mga2.noarch (in order to keep gnome-themes-standard-3.3.5.1-1.mga2.i586) adwaita-gtk2-theme-3.3.90.1-1.mga2.i586 (in order to keep gnome-themes-standard-3.3.5.1-1.mga2.i586) adwaita-gtk3-theme-3.3.90.1-1.mga2.i586 (in order to keep gnome-themes-standard-3.3.5.1-1.mga2.i586) Continue installation anyway? (Y/n) BS error ? Mirroring ? can you please resubmit ? TIA -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Firefox craches on libstdc++
hi... after latest update to 10.0.2, firefox chrashes here: werewolf:~> firefox pure virtual method called terminate called without an active exception Aborted (core dumped) #2 0x77775aed in __gnu_cxx::__verbose_terminate_handler() () from /usr/lib64/libstdc++.so.6 #3 0x77773c96 in ?? () from /usr/lib64/libstdc++.so.6 #4 0x77773cc3 in std::terminate() () from /usr/lib64/libstdc++.so.6 #5 0x777746df in __cxa_pure_virtual () from /usr/lib64/libstdc++.so.6 #6 0x75e6a60c in js::ProxyHandler::has(JSContext*, JSObject*, long, bool*) () from /usr/lib64/firefox-10.0.2/libxul.so #7 0x75e6df43 in ?? () from /usr/lib64/firefox-10.0.2/libxul.so #8 0x75e6dfdf in ?? () from /usr/lib64/firefox-10.0.2/libxul.so #9 0x75e42b54 in ?? () from /usr/lib64/firefox-10.0.2/libxul.so #10 0x75e4aa67 in ?? () from /usr/lib64/firefox-10.0.2/libxul.so #11 0x75e206ec in ?? () from /usr/lib64/firefox-10.0.2/libxul.so #12 0x75e314df in ?? () from /usr/lib64/firefox-10.0.2/libxul.so #13 0x75e316c7 in ?? () from /usr/lib64/firefox-10.0.2/libxul.so #14 0x75dbb4cd in ?? () from /usr/lib64/firefox-10.0.2/libxul.so #15 0x75dbb584 in JS_EvaluateUCScriptForPrincipals () Missing a virtual method loos like something needs a rebuild, but what ? Anybody else sees this ? -- J.A. Magallon \ Winter is coming...
[Mageia-dev] Handling of tty1 [was Can't switch to virtual consoles]
On 02/14/2012 07:39 PM, Thierry Vignaud wrote: > On 14 February 2012 19:01, Juan Luis Baptiste wrote: >>> I have noticed that now I can't switch to a virtual console >>> (ctrl+alt+F#), is this intentional, or a bug ? I have searched >>> bugzilla but didn't found anything. >>> That reminds me of another issue... I have some sever boxes booting in text mode, by making default.target point to multi-user.target. I have no getty spawned for tty1, so they look strange, boot messages remain in screen. Why is getty for tty1 missing ? annwn:/lib/systemd/system# ps -ef | grep getty root 2525 1 0 Feb13 tty3 00:00:00 /sbin/agetty tty3 38400 root 2530 1 0 Feb13 tty2 00:00:00 /sbin/agetty tty2 38400 root 2534 1 0 Feb13 tty5 00:00:00 /sbin/agetty tty5 38400 root 2539 1 0 Feb13 tty6 00:00:00 /sbin/agetty tty6 38400 root 2542 1 0 Feb13 tty4 00:00:00 /sbin/agetty tty4 38400 root 32761 32556 0 23:00 pts/000:00:00 grep --color getty annwn:/lib/systemd/system# ps -ef | grep dm root 2470 1 0 Feb13 ?00:00:00 rpc.idmapd root 32763 32556 0 23:00 pts/000:00:00 grep --color dm TIA! -- J.A. Magallon \ Winter is coming...
