Re: [Mageia-dev] M3 beta - where to report problems?
On Mon, 08 Apr 2013 17:43:38 -0400, Thomas Backlund wrote: Colin Guthrie skrev 9.4.2013 00:31: 'Twas brillig, and Thomas Backlund at 08/04/13 13:20 did gyre and gimble: It's actually easier than that. Add "systemd.unit=multi-user.target" on kernel / grub command line and ith will boot to runlevel 3 Or simply add a "3" onto the command line as has been the case for many years :) Of course I should probably discourage that seeing as "3" is pretty meaningless in absolute terms these days! :) Are you sure ? IIRC there has been reports of people doing it and still eneded up at graphical.target ... I tested with an install from Mageia 3 beta 4, and can confirm it still works. Regards, Dave Hodgins
Re: [Mageia-dev] M3 beta - where to report problems?
On Mon, 08 Apr 2013 12:52:06 -0400, Anne Wilson wrote: Rebooting, nothing has changed. I get grub, lots of disk activity led flickering, but no output to the screen, with or without Esc. I can boot into Safe moade, which lands me in Rescue Mode - but I haven't a clue what to do next. Try editing the kernel parameters in grub. Remove "splash quiet", if present. Add " 3 xdriver=vesa". The 3 gets translated by systemd to systemd.unit=multi-user.target, is shorter to type, and easier to remember. Regards, Dave Hodgins
Re: [Mageia-dev] [Mageia-sysadm] setup package not installed until quite late on...
On Mon, 01 Apr 2013 16:56:40 -0400, Thierry Vignaud wrote: setup's scriptlets were archeological remanants of mdv2007. I've killed them for now. If needed, they can be restored either by spliting setup-basesystem (https://bugs.mageia.org/show_bug.cgi?id=9326#c19) or by converting them to lua as suggested by you. As the posttrans scriptlet no longer runs pwconv 2>/dev/null >/dev/null || : grpconv 2>/dev/null >/dev/null || : what is supposed to create /etc/shadow and /etc/gshadow? Regards, Dave Hodgins
[Mageia-dev] List of packages that use dconf in scriptlets.
Like rtkit, updating gdm during an upgrade is also blocking urpmi in a postinstall scriptlet. This time it's a dconf command. https://bugs.mageia.org/show_bug.cgi?id=9535 A list of packages using dbus-send or dconf in the scriptlets would be help helpful. Regards, Dave Hodgins
[Mageia-dev] List of packages that use dbus-send in scriptlets.
Is there any way to get a list of all packages that use dbus-send, in the installation scriptlets? As per https://bugs.mageia.org/show_bug.cgi?id=9534, which is for the rtkit package, it uses dbus-send in it's postinstall scriptlet, which doesn't work during upgrade, blocking urpmi until the dbus-send is manually killed. A list of packages using dbus-send in the scriptlets would help ensuring upgrading with those packages is tested. Regards, Dave Hodgins
Re: [Mageia-dev] [changelog] [RPM] cauldron nonfree/release chromium-browser-stable-25.0.1364.173-4.mga3.nonfree
On Mon, 25 Mar 2013 02:09:49 -0400, Thierry Vignaud wrote: On 24 March 2013 21:48, dmorgan wrote: dmorgan 25.0.1364.173-4.mga3: + Revision: 405124 + rebuild (emptylog) what for? There should never be an empty changelog... As per Bug 9529 - Changelogs are not available for new packages. I think there is something wrong with the build system, as all recent packages I checked have no change logs. Regards, Dave Hodgins
Re: [Mageia-dev] Fw: Wifi
On Sat, 23 Mar 2013 12:52:15 -0400, Sander Lepik wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 23.03.2013 04:09, Trish Fraser kirjutas: Hi all, This seems to be an ongoing issue - can someone please help me with an answer for this user? And the magical bug number is? :) No bug number yet. The forwarded message was sent to the contact address, after a failed attempt to get wifi working, for a person brand new to Mageia. Jason has agreed to join the qa team, so we'll be helping him learn to use bugzilla, and provide useful information. Regards, Dave Hodgins
[Mageia-dev] Help needed with ldap server.and gdm.
I've been working on getting an ldap server working in Mageia 2, and have a strange error, that I'm not sure how to debug. I have two users. The user dave existed before adding the ldap server, and is defined in both /etc/passwd (group and shadow too), as well as in the ldap database. The user qatest is only defined in the ldap database. I've been using the instructions from http://linsec.ca/Using_OpenLDAP_for_User_Authentication#Configuring_PAM_to_use_LDAP In the auth section of /etc/pam.d/system-auth, if I use authsufficient/lib/security/pam_pwdb.so likeauth nullok both users are listed by gdm, but only the qatest user can login, and the "su -" command fails with "unknown pam error". The user dave can login ok in run level 3, and startx, but also cannot use the "su -" command, or login via gdm If I use the line authsufficient/lib/security/pam_unix.so likeauth nullok both users can login with run level 3, startx, and use "su -", but the qatest user is not listed by gdm, so can't login with a normal run level 5 startup, while the user dave can. So it looks like I should be using pam_unix, but need something to get gdm to list the ldap only user. Any suggestions? Thanks, Dave Hodgins
Re: [Mageia-dev] Fw: Wifi
On Fri, 22 Mar 2013 22:09:56 -0400, Trish Fraser wrote: Hi all, This seems to be an ongoing issue - can someone please help me with an answer for this user? I've already responded to the op, basically stating that more info is needed, and suggesting Jason consider joining the qa team, so we have more hardware available for testing. Our messages crossed paths, in that I sent my response in the same pop3 connection where I downloaded Trish's message that the message was being forwarded here. Regards, Dave Hodgins
Re: [Mageia-dev] 404 when trying to download Mageia-3-beta3-Boot-nonfree-x86_64-CD.iso
On Mon, 11 Mar 2013 22:03:27 -0400, Glen Ogilvie wrote: Hi, I would like to test the installer, to see if bug: #9032 is resolved, However, can't download the network installer, it is giving me a 404. http://www.mageia.org/en/downloads/get/?q=Mageia-3-beta3-Boot-nonfree-x86_64-CD.iso Also, I don't see any standard full DVD images available, which would be good to test as well. I've closed bug 9032, after seeing your message here, and then doing a test install using lvm logical volumes. I expect the network installers will be pushed to the mirrors, when the dvd iso images are pushed. Interested in joining the qa team? Regards, Dave Hodgins
Re: [Mageia-dev] System doen't boot with LVM
On Mon, 18 Feb 2013 06:19:00 -0500, Olivier Thauvin wrote: It is impossible to use an already existing encrypted partition: it said 'cryptsetup failed' and nothing more. For this part, see https://bugs.mageia.org/show_bug.cgi?id=8769 - diskdrake no longer provides the "use" button Regards, Dave Hodgins
Re: [Mageia-dev] [RFC] rsyslog vs journalctl
On Mon, 11 Feb 2013 20:16:38 -0500, Johnny A. Solbu wrote: On Monday 11. February 2013 23.42, David W. Hodgins wrote: make sure there is a line with "127.0.0.1 beria.inria.fr beria" in /etc/hosts, then run "dracut -f". Why run dracut? My mistake. I was thinking journalctl was being started during the initrd processing, but looking at the hooks, it only creates the socket, with journalctl not actually starting until after the root pivot. Regards, Dave Hodgins
Re: [Mageia-dev] [RFC] rsyslog vs journalctl
On Mon, 11 Feb 2013 13:59:46 -0500, Guillaume Rousse wrote: Whereas my host name, set from HOSTNAME variable in /etc/sysconfig/network, is beria.inria.fr. I have to restart my syslog daemon to get it correct. So I guess just ensuring the hostname is either set prior the start of the logging process, or refreshing this last one thereafter, would avoid the issue. In addition to having HOSTNAME=beria.inria.fr in /etc/sysconfig/network put "beria.inria.fr" (without the quotes or HOSTNAME=) in /etc/hostname, and make sure there is a line with "127.0.0.1 beria.inria.fr beria" in /etc/hosts, then run "dracut -f". Regards, Dave Hodgins
[Mageia-dev] Getting abrt/libreport working.
Regarding https://bugs.mageia.org/show_bug.cgi?id=8937 is anyone able to get abrt working on Mageia 2? If so, any idea what else needs to be installed or configured? Regards, Dave Hodgins
Re: [Mageia-dev] Freeze push: opera 12.14
On Tue, 05 Feb 2013 18:30:41 -0500, FundaWang wrote: Hello, Could opera 12.14 be pushed into cauldron? It addresses a re-occuring crash, If the version in Mageia 2 Nonfree Updates Testing is ready for qa testing, a bug report assigned to qa is needed. Regards, Dave Hodgins
Re: [Mageia-dev] x11-driver-video-qxl/qemu/libvirt (was Re: Please remove qemu, and qemu-img from Mageia 3.)
On Mon, 04 Feb 2013 18:09:52 -0500, Christiaan Welvaart wrote: On Sun, 3 Feb 2013, David W. Hodgins wrote: During testing of updates to qemu, and x11-driver-video-qxl, it has become very clear that no-one could possibly be using these packages, as they are so slow, as to be useless. Qemu (or rather each qemu-system-*) is an emulator at its core, so slowness is not a huge surprise. You probably wanted to run it in (fast) virtualization mode - use qemu-kvm for that (and the kvm module for your cpu must be loaded to get virtualization working). But running qemu directly is tiresome, try virt-manager - it works about as well as virtualbox even if I need to run it as root. I use virt-manager/qemu-kvm to build+test mga2 updates for iceape, only graphics (spice+vmvga) is slow (but better than vnc+cirrus). Thanks, I'll take a look at virt-manager. Regards, Dave Hodgins
Re: [Mageia-dev] Please remove qemu, and qemu-img from Mageia 3.
On Sun, 03 Feb 2013 04:16:54 -0500, Pascal Terjan wrote: On Sun, Feb 3, 2013 at 9:45 AM, David W. Hodgins wrote: During testing of updates to qemu, and x11-driver-video-qxl, it has become very clear that no-one could possibly be using these packages, as they are so slow, as to be useless. Try loading kvm module and selecting the fast drivers for network/disk/... lsmod|grep kvm kvm_amd55516 0 kvm 413942 1 kvm_amd Whether it's loaded or not doesn't seem to make any difference. If I add the -enable-kvm option, it fails with ... KVM not supported for this target No accelerator found! Checking under htop shows it's clearly cpu bound, using 2 of my 4 cores. The command I used to run qemu was ... qemu-system-x86_64 -cdrom "$Iso" -hda mageia"$Arch".qcow2 -boot d -net nic -net user,net=192.168.10.0/16,host=192.168.10.3 -m 4096 -vga qxl with the Iso and Arch variables set appropriately. Suggestions for faster disk/network options? Given that other people do find it useful, it's obvious there's something wrong either with the options I'm using (or not using), or my hardware or bios settings. Hopefully it's not the hardware. The cpu is an AMD FX(tm)-4170 Quad-Core Processor, with the following flags shown in /proc/cpuinfo ... flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc extd_apicid aperfmperf pni pclmulqdq monitor ssse3 cx16 sse4_1 sse4_2 popcnt aes xsave avx lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs xop skinit wdt lwp fma4 nodeid_msr topoext perfctr core arat cpb hw_pstate npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold Regards, Dave Hodgins
[Mageia-dev] Please remove qemu, and qemu-img from Mageia 3.
During testing of updates to qemu, and x11-driver-video-qxl, it has become very clear that no-one could possibly be using these packages, as they are so slow, as to be useless. Over 6 hours to do a net install of a kde x86-64 system, on hardware where that would normally be under 20 minutes. There are many problems with xorg crashing, mouse pointer not being where the pointer is shown, etc. These bugs are described in https://bugs.mageia.org/show_bug.cgi?id=8938 Please drop these useless packages, so we don't have to test or install security updates for them. Thanks, Dave Hodgins
Re: [Mageia-dev] Questions on dualbooting/grub/bootloader-utils
On Fri, 18 Jan 2013 17:36:07 -0500, Barry Jackson wrote: On 18/01/13 21:34, David W. Hodgins wrote: When I select the second drive, as far as grub is concerned, it is hd0, while as far as the kernel is concerned, it's sdb. Regards, Dave Hodgins Try swapping your SATA data cables over. The sda / sdb thing is determined by the detection order i.e. SATA channel IIANM. I should have added, that I'm doing this intentionally. I currently have grub2 on sda, while qa testing the Mageia 3 beta 2 iso images, with grub legacy on sdb for my Mageia 2 installation, just in case there are problems with Mageia 3 or grub2 detecting or booting my Mageia 2 installation. Regards, Dave Hodgins
Re: [Mageia-dev] Questions on dualbooting/grub/bootloader-utils
On Fri, 18 Jan 2013 07:06:00 -0500, JA Magallón wrote: Why do bootloader-utils (I think?) change device.map, menu.lst settings ? Why forces the boot disk to be always hd0 ? The bootloader-utils assume the sdb will be hd0 at boot time, which is the case on my system. Is there anyway to tell them I dont want to change anything ? Have you tried making the file read only? No idea if that will work. 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...). I use f8 on my system, during boot, to get to a boot menu, then select the second drive (my bios remembers the last selection, and keeps it, on reboots). When I select the second drive, as far as grub is concerned, it is hd0, while as far as the kernel is concerned, it's sdb. Regards, Dave Hodgins
Re: [Mageia-dev] [338616] Use official tarball
On Fri, 04 Jan 2013 07:18:05 -0500, Thomas Backlund wrote:
root-odjjhxpcy38dnm+yrof...@public.gmane.org skrev 4.1.2013 12:29:
-# Original source
http://www.bouncycastle.org/download/bcprov-%{archivever}.tar.gz
-# is modified to
-# bcprov-%{archivever}-FEDORA.tar.gz with patented algorithms removed.
-# Specifically: IDEA algorithms got removed.
-Source0: bcprov-%{archivever}-FEDORA.tar.gz
+# Use original sources from here on out.
+Source0:
http://www.bouncycastle.org/download/bcprov-%{archivever}.tar.gz
Ummm, does this mean you pushed patented into core/release,
or has the patent expired ??
Expired. https://bugs.mageia.org/show_bug.cgi?id=6910
Regards, Dave Hodgins
Re: [Mageia-dev] fstab issue - can mount but not umount as user, ehh, what?
On Fri, 28 Dec 2012 11:13:10 -0500, Johnny A. Solbu wrote: When I plug in the usbkey, I can mount it as a reguklar user, but I cannot unmount it. The error message says that only root can do that. The fstab entry is this: == UUID=0893-1889 /media/usbbrikke vfat nosuid,noexec,dmask=022,fmask=133,uid=500,noauto,user,iocharset=utf8 0 0 == Change the user option to users. https://bugs.mageia.org/show_bug.cgi?id=3533 Regards, Dave Hodgins
Re: [Mageia-dev] Installing Beta 1
On Sun, 23 Dec 2012 18:04:38 -0500, Thomas Spuhler wrote: I installed (and tried to) Beta 1 on a vbox I created a brand new HDD and installed Beta2 from the DVD (64bit) Everything went smooth I did not install the optional upgrades I rebooted the system and did not get the login screen, but was able to login in a shell I did a startx, got a flash of the X windows, but it die immediately. I couldn't find anything suspicious in the log. I then started all over and added the optional upgrades. Everything went fine this time and I have a shiny system. Only one little oddy, it installed the desktop-kernel AND the server-kenrel What did you select during the install? I just did a kde (using default package group selections) install using the 64 bit beta 1 dvd, without installing updates, and it worked ok. Regards, Dave Hodgins
Re: [Mageia-dev] DBus and automount problem with KDE
On Sat, 22 Dec 2012 00:53:02 -0500, Curtis Hildebrand wrote: I need some help debuging a problem with automount and DBus in KDE. For example, a USB key is recognized in the log files, but won't mount or show in Dolphin. All was fine till a few days ago. Gnome works as expected. I suspect it's related to another problem I'm having with DBus. Here are some symptoms: ]$ dbus-monitor --system Failed to open connection to system bus: The maximum number of active connections for UID 502 has been reached See https://bugs.mageia.org/show_bug.cgi?id=8393 Regards, Dave Hodgins
Re: [Mageia-dev] Utter frustration
On Thu, 29 Nov 2012 17:35:01 -0500, Frank Griffin wrote:
I don't think anything's going to get done in that timeframe, but
CTRL-ALT-{R,S,E,I,U,B} will get you through the weekend.
Shouldn't that be ctrl+alt+sysrq-r,s,e,i,s,u,b?
Regards, Dave Hodgins
Re: [Mageia-dev] forkbomb protection
On Wed, 28 Nov 2012 13:00:05 -0500, Johnny A. Solbu wrote: On Wednesday 28. November 2012 17.53, David Walser wrote: Their pam package has a /etc/security/limits.d/90-nproc.conf file that has: *softnproc1024 As the last comment on the bug says, it's a bit confusing that it's in limits.d/ and not the limits.conf file itself, His point is that any limits set in «/etc/security/limits.d/» overrides the «limits.conf» file. and in fact I'm not sure what is responsible for processing limits.d/* as limits.conf says nothing about it (Fedora's is the exact same as ours). We should add some comments in «/etc/security/limits.conf» about it. Anyway, one way or another it would be nice to have this limit set by default on Mageia, IMHO. WDYT? I think we should have this. This is also being discussed in the usenet newsgroup alt.os.linux.mageia. I've confirmed the forkbomb will kill my Mageia 2 x86-64 system, with the default value for nprocs of 127910. Interestingly, it doesn't kill the system if I run the forkbomb right after rebooting, only if it's been in use for a while. This is a quad core with 16GB of ram. I've added a line to /etc/security/limits.conf on my system, with * hardnproc 1 The forkbomb no longer has impact, except for the need to kill all of the user's bash processes. # ps -A|wc -l 218 I think 1 should be more than adequate, yet low enough to stop the bomb from killing the system. Regards, Dave Hodgins
Re: [Mageia-dev] USB speaker systems
On Wed, 21 Nov 2012 19:39:15 -0500, Richard Couture wrote: I was looking at the Bose Companion 5 speaker system and noticed that it intefaces with the computer via USB. Am I going to run into problems with Mageia if I buy it? According to https://en.wikipedia.org/wiki/Bose_computer_speakers#Companion_5 it works with linux kernels that support USB audio including Pulse Audio, so it should be ok. Haven't tried it myself though. Regards, Dave Hodgins
Re: [Mageia-dev] ANN: shadow-utils & util-linux with new default paths in testing
On Tue, 20 Nov 2012 19:35:37 -0500, JA Magallón wrote: 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 ? Yes. The purpose in my understanding, is that /usr/local binaries are intended to override /usr binaries. Same with stuff in /opt. Regards, Dave Hodgins
Re: [Mageia-dev] Anyone know where /usr/bin/shutdown comes from?
On Fri, 07 Sep 2012 11:51:05 -0400, Colin Guthrie wrote: Not owned by any package so presumably it's from a %post or similar. Anyone know which package or a way to easily search for it? See grep -i shutdown /usr/share/msec/plugins/* Regards, Dave Hodgins
[Mageia-dev] Package without signature - libqthelp4
The following package has bad signature: /var/cache/urpmi/rpms/libqthelp4-4.8.2-3.mga3.i586.rpm: Missing signature (OK ((none))) I thought this wasn't supposed to be able to happen. Regards, Dave Hodgins
Re: [Mageia-dev] libpostfix issue.
On Mon, 16 Jul 2012 20:18:45 -0400, Luis Daniel Lucio Quiroz wrote: Le lundi 16 juillet 2012 09:32:13 Colin Guthrie a écrit : ldconfig: /usr/lib64/libpostfix-xsasl.so.1 is not an ELF file - it has the wrong magic bytes at the start. how tofix this"? Looks to me like it's a postfix problem. file /usr/lib/libpostfix-xsasl.so.1 /usr/lib/libpostfix-xsasl.so.1: current ar archive ar -t /usr/lib/libpostfix-xsasl.so.1 xsasl_server.o xsasl_cyrus_server.o xsasl_cyrus_log.o xsasl_cyrus_security.o xsasl_client.o xsasl_cyrus_client.o xsasl_dovecot_server.o Not sure about how to fix it. Regards, Dave Hodgins
Re: [Mageia-dev] krb5 paths issue
On Wed, 04 Jul 2012 18:27:38 -0400, David Walser wrote: An issue has been raised while QA testing an update for krb5 for Mageia 1 and Mageia 2 about apparently incorrect paths in init scripts: https://bugs.mageia.org/show_bug.cgi?id=6469 Would the correct thing to do be to change /var/kerberos/krb5kdc to /etc/kerberos/krb5kdc as proposed in this patch to kadmin.init?: https://bugs.mageia.org/attachment.cgi?id=2476&action=diff I also noticed a line in kprop.init that says: [ -f /var/kerberos/krb5kdc/kpropd.acl ] || exit 6 Should that be changed as well? Yes. Ideally, only the config files should be in /etc, and the database should be in /var. In order to get the security update out quickly, the scripts should be changed to /etc, to match the current config files. I think a new bug report should be opened, requesting changing the various scripts to use something like eval $(sed 's/ //g' /etc/kerberos/krb5kdc/kdc.conf|grep ^database_name) to get the location of the database (with similar code for the other files) and have new installs put the database in /var. Also, in krb5 in Cauldron, I similarly noticed the following... kadmin.service: ConditionPathExists=!/var/kerberos/krb5kdc/kpropd.acl kprop.service: ConditionPathExists=/var/kerberos/krb5kdc/kpropd.acl Should those be changed as well? Yes. Regards, Dave Hodgins
Re: [Mageia-dev] bug 2317 revisited: --update option should behave like --search-media
On Fri, 22 Jun 2012 14:11:58 -0400, AL13N wrote: ok, i guess when people said, supported, i immediately assumed full support. that kind of misled me. My understanding, is that backports will have minimal testing. We ensure the backport will install, on a system that currently only has release + updates, and that the basic functions of the package work, where it's possible for us to test. We cannot possibly test every possible combination of selected backports, and will not attempt to. What I don't want to see, which I did with Mandriva backports, is cases where unsigned rpm packages were in the repositories, or installation required the use of --allow-nodeps or --allow-force, due to file conflicts with release or updates packages. Those problems were rare, but often enough, that I wouldn't let the end users I support install backports themselves. Regards, Dave Hodgins
Re: [Mageia-dev] bug 2317 revisited: --update option should behave like --search-media
On Thu, 21 Jun 2012 17:17:50 -0400, zezinho wrote: I vote for Mrs.B workaround : I agree. Simply treating release media that's enabled, as if it had the update flag seems to the easiest way to safely fix 2317. Two people have tested simply enabling Release media as update media and not really noticed any drop in performance. I don't know what the In my case, I compared the time between mgaapplet "Computing new updates", and the message "Checking... Updates are available" (with updates actually being available), was 18 seconds when Release was not marked as an update repository, and it was 18 seconds when it was marked as an update repository. As mgaapplet is only using around 30 KB of ram, I don't see memory being an issue. Regards, Dave Hodgins
Re: [Mageia-dev] Proposed Feature:Backports_update_applet
On Fri, 15 Jun 2012 01:07:20 -0400, andre999 wrote: If the fact it is a backport is ignored, then every backport would be, by definition, an update. Even packages newly imported to Mageia. It depends on whether the user has set the backport repsository as an update repository, or not. Backports are considered separately because they are much more at risk to not function properly, since they weren't tested with the rest of the release, being added afterwards. So we have to be much more careful Thats why qa will be testing backports, to ensure they install cleanly, and the main features work. As a qa member, I will not validate a backport that either requires --allow-force, or that doesn't at least run. Regards, Dave Hodgins
Re: [Mageia-dev] Backports policy clarification (and discussion)
On Tue, 12 Jun 2012 14:10:02 -0400, Sander Lepik wrote: There is no such repo as "Core Backports Updates". If the package is validated it will be pushed into "Core Backports". Ah. Yes. Should have looked more carefully first. Sorry for the noise. Regards, Dave Hodgins
Re: [Mageia-dev] Backports policy clarification (and discussion)
On Fri, 08 Jun 2012 04:38:06 -0400, Samuel Verschelde wrote: I re-read the backports policy, and there's a part I think needs to be pointed out before people start to backport packages. I have a different question. Will repositories such as "Core Backports" ever actually be used? If we follow the procedures used for normal updates, a backport will first be submitted to "Core Backports Testing", then assigned to qa. Once validated, the backport srpm will be pushed to "Core Backports Updates". When would "Core Backports" ever be used? Regards, Dave Hodgins
Re: [Mageia-dev] [Mageia-artwork] Mageia 2 beta 3 available for tests
On Mon, 11 Jun 2012 15:47:56 -0400, Juan Luis Baptiste wrote: On Wed, Apr 18, 2012 at 4:43 PM, tumbeliina wrote: ~~~ That's because that's an old link for beta release, Mageia 2 final was released on May 22. You can get the final version from: Note the date. Seems old messages are just getting approved by moderators now. Regards, Dave Hodgins
Re: [Mageia-dev] Backports policy clarification (and discussion)
On Fri, 08 Jun 2012 10:22:41 -0400, Samuel Verschelde wrote: I think you missed my point. If Mageia 1 "backports" has higher versions than Mageia 2 "release" (not backports), upgrade can fail because currently our tools do not take backports from the target release (mageia 2) into account when upgrading a distro. In the upgrade from Mandriva 2010.2 to Mageia 1, it was made clear, that upgrading from a system with 2010.2 Backports was not supported. It may work, but was not recommended. I think we should keep the same policy for the upgrade from Mageia 1 to 2. I.E. Don't use backports if you are planning on later doing an upgrade, rather then a clean install. That way, Mageia 1 users who want firefox 13 can get it, without us having to replace the Mageia 2 iso images with an upgraded installer, that will keep backports enabled for 2, if it was enabled for 1. Regards, Dave Hodgins
[Mageia-dev] Suggestion to obsolete ncpfs for Mageia 3.
As per https://bugs.mageia.org/show_bug.cgi?id=6153#c4 I think the package ncpfs should be obsoleted in Mageia 3, unless we have a user step forward, who has access to a Novell Netware server, and who is willing to help with future qa testing. Regards, Dave Hodgins
Re: [Mageia-dev] Mageia 3 specifications
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. Regards, Dave Hodgins
Re: [Mageia-dev] Mageia 3 specifications
On Tue, 05 Jun 2012 05:02:35 -0400, Anne Nicolas wrote: Le 30/05/2012 23:43, Anne Nicolas a écrit : Here is the process: - until 15/06: make proposals for Mageia 3 specifications (features) following our policy - 16 to 18/06: choose features that will be added for official specifications of Mageia 3 - publish and communicate on it Mageia features policy: https://wiki.mageia.org/en/Features_policy Template for feature proposal: https://wiki.mageia.org/en/Feature_Template Please read carefully this policy and use the template. Once it's done, feel free to mail -dev if you think you need to discuss your feature and find people so that you can propose resources to work on it. ping all. Very few proposals for now. Please take some time to fill it, it's important for coming release Is there an index page that makes it easy to see what features have already been submitted? Regards, Dave Hodgins
Re: [Mageia-dev] My proposals for Mageia 3
On Mon, 04 Jun 2012 11:39:08 -0400, Marcello wrote: Hi, i've written some proposals for Mageia 3. Is there anyone who wants to adopt them? https://wiki.mageia.org/en/CheckboxesToChooseDE The installer already has these, so I don't understand what you want changed. You can select kde or gnome, or custom, where these additional choices are present. https://wiki.mageia.org/en/CreateHibridISO Bad title. All of the iso images are already hybrid iso images, in that they can be booted from cd/dvd or usb drive, etc. I think you are referring to creating a live dvd, rather then the current live cd images. Note that there are still a lot of people with limited download speed, and/or download caps, beyond which they have to pay extra. I do agree that a live dvd would be more useful, and be less work to create/test than the current live cds. https://wiki.mageia.org/en/EditionBasedScreenshots The doc team does update the documentation for each release already. https://wiki.mageia.org/en/InstallMissingSoftware But what packages should that include? Ask 10 people, you'll get at least 10 different answers. https://wiki.mageia.org/en/NewbieHow-To That I agree would be useful, but will require a lot of review to figure out what is actually needed for a new user. Regards, Dave Hodgins
Re: [Mageia-dev] Updating usb_modeswitch in mga2
On Sun, 03 Jun 2012 14:45:52 -0400, Colin Guthrie wrote: Hi, In order to fix https://bugs.mageia.org/show_bug.cgi?id=4853 we would need a newer usb_modeswitch (and it's -data supplement). As it's an update for a bugfix, where a patch is not suitable, I see no problem pushing it as an update. Regards, Dave Hodgins
[Mageia-dev] Unsigned packages in Core Release fore Mageia 2.
As per https://bugs.mageia.org/show_bug.cgi?id=4925 it seems some packages never got installed by qa members during testing. Can someone with access to a full mirror check all packages to see how many others there are? Regards, Dave Hodgins
Re: [Mageia-dev] [soft-commits] [4584] (systemd_unit_exists) factorize
On Wed, 23 May 2012 14:38:09 -0400, Colin Guthrie wrote: NAK. No it's not redundant it's very important. The symlinks *will* be broken when run via the installer due to the $::prefix as they will point to "non-existent" paths in /lib/systemd/ (not to /mnt/lib/systemd...) so I treat any symlink that exists (even if it points to an invalid unit) as "valid" here. Wouldn't it have been better to use relative symlinks then? Regards, Dave Hodgins
[Mageia-dev] Great job everyone
I've completed qa testing of the latest i586 iso images. Four clean installs, two upgrade from Mageia 1 tests, all completed without any packaging errors. No release blocker bugs encountered. Thanks to everyone for a job well done. Regards, Dave Hodgins
Re: [Mageia-dev] [soft-commits] [4545] Add more missing packages (#5947, #5924, #5951)
On Thu, 17 May 2012 16:27:38 -0400, Thierry Vignaud wrote: Why includ uclibc https://bugs.mageia.org/show_bug.cgi?id=5924 It's required by libuClibc-zlib1, which is on the dvd. Regards, Dave Hodgins
Re: [Mageia-dev] Proposal to drop the package wicd
On Sat, 12 May 2012 08:40:15 -0400, zezinho wrote: Em 12-05-2012 05:39, David W. Hodgins escreveu: I think the package should be made obsolete in both Cauldron, and Mageia 1. Opinions? I agree. I've opened https://bugs.mageia.org/show_bug.cgi?id=5926 requesting the wicd package be dropped/obsoleted. Regards, Dave Hodgins
Re: [Mageia-dev] Proposal to drop the package wicd
On Sat, 12 May 2012 08:40:15 -0400, zezinho wrote: Em 12-05-2012 05:39, David W. Hodgins escreveu: I think the package should be made obsolete in both Cauldron, and Mageia 1. Opinions? I agree. If no-one objects, I'll post a request to the sysadmin team in a few hours. Regards, Dave Hodgins
[Mageia-dev] Proposal to drop the package wicd
As discussed in https://bugs.mageia.org/show_bug.cgi?id=5608 while testing a security update for the wicd package, I found that the updated version, the Core release version, and the Cauldron version are all unable to run, due to incompatibilities with python 2.7. As the package conflicts with the drake network tools, dhcpd, and networkmanager, (but installing the package doesn't auto disable them), I don't think we need yet another way to for people to try and configure their network connections. I think the package should be made obsolete in both Cauldron, and Mageia 1. Opinions? Thanks, Dave Hodgins
Re: [Mageia-dev] Calligra 2.4.? in Mga 2
On Tue, 08 May 2012 22:37:49 -0400, Funda Wang wrote: Should be 2.4.1, I've pushed it into core/updates_testing I think that should be pushed into cauldron release, to avoid problems updating from Mageia 1. Regards, Dave Hodgins
Re: [Mageia-dev] freeze push: request Gimp 2.8 regarding single window interface!
On Mon, 07 May 2012 06:22:30 -0400, Olav Vitters wrote: We already have gimp 2.7.x, so this is nothing new (already an issue in Mageia 2). A discussion on the alt.os.linux.mageia usenet newsgroup shows that 2.8 fixes https://bugzilla.gnome.org/show_bug.cgi?id=675684https://bugzilla.gnome.org/show_bug.cgi?id=675684 Regards, Dave Hodgins
Re: [Mageia-dev] Mageia cypher encryption proposal
On Sat, 05 May 2012 17:03:21 -0400, Simple W wrote: But with be better if users could choose from a cypher list, like it happens for example in debian installer, and not have a single one to handle encrypted partitions. That's a much bigger change to diskdrake. You can open an enhancement bug report requesting that, but it'll be up to the diskdrake maintainer, if he has the time, and wants to do it. Regards, Dave Hodgins
Re: [Mageia-dev] Mageia cypher encryption proposal
On Sat, 05 May 2012 05:12:38 -0400, Simple W wrote: Like things are now im forced (after isntall Mageia 2") to edit /usr/lib/DrakX/ds/dmcrypt.pm and replace aes-xts-benbi with aes-xtl-plain and recreate the encrypted partitions so that i can access them when for example running windows. I'll submit a bug report with the proposed patch, as an update for rpmdrake, after Mageia 2 is out. While it's a very minor change, as I found out from the prior change, minor changes can have unexpected effects. The prior change worked fine on an installed system (I've been using it since 2008), but caused unexpected problems in the installer, as the xts module had to be explicitly loaded. Regards, Dave Hodgins
Re: [Mageia-dev] no sound
On Sat, 05 May 2012 12:05:50 -0400, Colin Guthrie wrote: Hopefully that'll get sound, but the correct fix is for me to fix this in PulseAudio... I should have opened a bug report about this, as this affects my system as well. I'm so used to the problem, that I automatically run pavucontrol and change the config from Analog Stereo Output to any of the Analog Surround Output options. From lspicdrake ... snd_via82xx : VIA Technologies, Inc.|VT8233/A/8235/8237 AC97 Audio Controller [MULTIMEDIA_AUDIO] (rev: 60) Regards, Dave Hodgins
Re: [Mageia-dev] Freeze Push: e2fsprogs
On Thu, 03 May 2012 10:21:40 -0400, Colin Guthrie wrote: Aside from this, the fact that e2fsck uses the current time as one of it's checks is just daft in the first place. If I have to replace the battery in my motherboard I don't want to have to get a bogus fsck warning. The change should stand in it's own right as being a sensible one. # grep future /var/log/dmesg [7.201526] dracut: 91-root: Superblock last write time is in the future. As this bug affects everyone in the western hemisphere, who has the rtc set to localtime, I agree with pushing the proposed fix. Regards, Dave Hodgins
Re: [Mageia-dev] Mageia 2 LiveCD contents...
On Sun, 29 Apr 2012 14:47:59 -0400, Thomas Backlund wrote: So atleast every cd with size > 700M need to loose some stuff... So please help out with suggestions... Looking at the list, the first one I would consider removing is inkscape. I'ts large, a leaf package, and doesn't look like something that a lot of people would use. Regards, Dave Hodgins
Re: [Mageia-dev] Handling single user/rescue/failsafe mode
On Thu, 26 Apr 2012 05:38:07 -0400, Colin Guthrie wrote: It seems that in mga1 single user mode just gave a shell without requiring root password. In Mandriva and Mageia 1, it's controlled my the msec setting of ENABLE_SULOGIN. When enabled by msec-gui, it does ask for the password in run level 1. That should be the same in Mageia 2, in my opinion. Regards, Dave Hodgins
Re: [Mageia-dev] [Mageia-discuss] many systemd temporary directories
On Tue, 24 Apr 2012 16:51:09 -0400, Colin Guthrie wrote: 'Twas brillig, and Florian Hubold at 24/04/12 21:27 did gyre and gimble: Am 24.04.2012 20:51, schrieb Juergen Harms: I just saw that systemd creates (and does not purge) lots of directories like /tmp/systemd-namespace-/ - I now have nearly 50 of them (one partial explanation might be that I did quite some re-booting these last days). Nothing serious, they are small - but nevertheless ... Is this normal, shouldnt there be some automatic purging mechanism? Juergen Maybe we should default to /tmp being on tmpfs so that this gets purged on every reboot? This would also be more in accordance with FHS imho. This is one of my goals for mga3 (I have it in my notes for the technical specs discussion that will come), but certainly not for mga2 as this is quite a big change and some apps may have to be patched to use /var/tmp instead etc. if they store large things in there. As long as it's still optional. With only 2GB of ram, I prefer to keep /tmp on disk, since I use it for things like dvd creation etc. Regards, Dave Hodgins
[Mageia-dev] LSB init scripts question.
On my Mageia 1 system, I noticed that xinetd wasn't getting started. After comparing the lsb headers, I modified the script as per the following patch, and confirmed that on reboot, it's now getting started. $ cat xinetd.patch --- /etc/init.d/xinetd.original 2011-01-11 23:35:28.0 -0500 +++ /etc/init.d/xinetd 2012-04-17 23:38:19.507878097 -0400 @@ -7,7 +7,7 @@ # Provides: xinetd # Required-Start: $network $syslog # Required-Stop: $network $syslog -# Default-Start: 345 +# Default-Start: 3 4 5 # Short-Description: A powerful replacement for inetd # Description: This startup script launches xinetd daemon ### END INIT INFO I was going to file a bug report, but decided to check other files. I found that udev-post and crond also are missing the spaces. udev-post does not get run at boot, as confirmed by /var/lock/subsys/udev-post not existing, however crond is getting started. This doesn't make sense to me. If the spaces are required, how is crond still getting started, while xinetd and udev-post are not? Regards, Dave Hodgins
Re: [Mageia-dev] Inverted colors in the flashplayer
On Tue, 10 Apr 2012 19:44:38 -0400, Pierre Jarillon wrote: Le mercredi 11 avril 2012 00:38:51, Pascal Terjan a écrit : > flash-player-plugin-11.2.202.228-1.mga1.nonfree > > With Mageia 1 and flashplayer 11.1.102.63 colors are correct. > > IMHO, it should be better to come back to the previous release of > flashplayer or to ask for a better player to Adobe. See http://comments.gmane.org/gmane.linux.mageia.devel/13644 Thanks, but is there anything automated for a beginner? Only geeks are able to understand the workaround and create this with root privileges: " to get hw video decoding I had to write a file in " /etc/adobe/mms.cfg with " EnableLinuxHWVideoDecode=1 It works for me (Intel - Nvidia - x86_64) I'm very surprised that it works. If you look at the html source, it's embedded video, not actually flash. Anyway, the colours look fine here. Regards, Dave Hodgins
[Mageia-dev] Latest upgrade testing
I think developers may be interested in the latest qa testing of upgrading from Mageia 1 to Caulron. Check out https://forums.mageia.org/en/viewtopic.php?f=4&t=2187 Regards Dave Hodgins
Re: [Mageia-dev] usb printers
On Mon, 02 Apr 2012 03:51:44 -0400, zezinho wrote: Em 01-04-2012 10:01, zezinho escreveu: I tried to use an USB printer, and while it was detected -> task-printing installed, it was not automaticaly added, as it is in MGA1. Trying to add it manually, only a LPT (parallel) printer is shown in MCC. Before I report a bug, has anyone else tried to use an USB printer? I found out that usblp blocks cups USB detection in Arch Wiki : https://wiki.archlinux.org/index.php/CUPS#USB_printers Blacklisting it worked for me. I no one steps out, I will add this blacklist file to cups package. Note that it is needed for some printers https://bugs.mageia.org/show_bug.cgi?id=2240 https://bugs.mageia.org/show_bug.cgi?id=2264 so if blacklisting by default is enabled, that should at least be pointed out in a README.urpmi file, so users of those printers will know how to un-blacklist it. Regards, Dave Hodgins
Re: [Mageia-dev] Report on my update from 1
On Sun, 01 Apr 2012 16:26:41 -0400, Colin Guthrie wrote: 'Twas brillig, and Thierry Vignaud at 01/04/12 20:01 did gyre and gimble: On 1 April 2012 02:01, Pascal Terjan wrote: - While dracut was running, it displayed plenty of libkmod: index_mm_open: major version check fail: 65537 instead of 2953311319 This is because kernel's module*bin are still generated by module-init-tools, not by kmod yet. Is this something we can ignore? I don't think so. It's likely to generate a lot of bug reports, and isn't good for the image of Mageia. FWIW, I've been running the kmod versions of things for a while. It's very late in the cycle to consider flipping over to them rather than module-init-tools, but I have made sure it's just a simple matter of flipping a --with argument in kmod package if we do. I think it should be treated as a bug that needs to be fixed. The sooner it gets done, the better the chance to find any problems and get them fixed before release freeze. Regards, Dave Hodgins
Re: [Mageia-dev] unable to mount encrypted partitions created with drakdisk
On Wed, 28 Mar 2012 21:40:36 -0400, simple w8 wrote: You have chosen thekeysize 512 thats the one is not supported under the FreeOTFE project... Isnt possible to change the keysize to a value that can be supported under Windows? Yes. Test what works on your own system, then open a bug report asking for the change. Just don't use cbc, as it's too insecure. Regards, Dave Hodgins
Re: [Mageia-dev] unable to mount encrypted partitions created with drakdisk
On Wed, 28 Mar 2012 19:00:04 -0400, simple w8 wrote: Something is not right, Free-OTFE says is does support xts but its not being able to mount the partitions encrypted in Mageia, what can be the cause? We're using --cipher aes-xts-benbi --key-size 512. As per http://www.ody.ca/~dwhodgins/Luks-Howto.html#Changelog this was suggested to me several years ago, and is what I've been using since then. You may see the FreeOTFE supported cypher modes here: http://www.freeotfe.org/docs/Main/Linux_volumes.htm I wasn't aware luks encrypted volumes could be opened by anything else, or I would have considered that when suggesting the cipher change. I don't think using the benbi initial vector generation algorithm (64 bit) instead of the plain (32 bit) algorithm makes much of a difference, in terms of security. It was just the cbc mode that I was concerned with. Can you modify /usr/lib/libDrakX/fs/dmcrypt.pm to use aes-xts-plain (line 68), create an encrypted volume, and test it with freeotfe? If that works, open a bug report and request the change. Regards, Dave Hodgins
[Mageia-dev] How should packages with version in name be handled?
I've started qa testing upgrading from Mageia 1 to Cauldron, to ensure going from Mageia 1 to 2 will go smoothly. I'm running into a some cases like https://bugs.mageia.org/show_bug.cgi?id=5103 where Mageia 1 has libchamplain-gtk0.8_1 and Cauldron has libchamplain-gtk0.12_0 As the packages contain version numbers in the package name, urpmi is not treating it as a package update, which results in messages such as The following packages have to be removed for others to be upgraded: libchamplain-gtk0.8_1-0.8.3-1.mga1.i586 (due to missing libclutter-gtk-0.10.so.0) libchamplain0.8_1-0.8.3-1.mga1.i586 (due to missing libclutter-glx-1.0.so.0) Am I correct in thinking that adding an obsoletes of libchamplain-gtk0.8_1 to libchamplain-gtk0.12_0 would get urpmi to remove it without generating the above message? If so, should I be opening bug reports for each of these cases, or should it be left as is, and simply tell the users to go ahead and answer yes to the prompt? Regards, Dave Hodgins
Re: [Mageia-dev] unable to mount encrypted partitions created with drakdisk
On Wed, 21 Mar 2012 03:41:00 -0400, simple w8 wrote: But i think it would be better to have in diskdrake some option allowing the user to choose the cypher, since the default cypher used in cryptsetup and in other apps that support luks, is cbc, and this way diskdrake is putting it incompatible with remaning apps that support luks. Agreed. Go ahead and write a patch! :-) Sorry, but my perl knowledge is very limited. Looking through the code, I can usually figure out what it's doing, I was able to create the patch that simply added the --cipher option to the existing code in diskdrake, but altering the dialog to add selection of which cipher to use, would probably take me a very long time. You should probably open an enhancement bug report, requesting the change, that can then be assigned to the diskdrake maintainer. Btw, I had no idea that luks encryption was compatible with truecrypt. Thanks for that info. Regards, Dave Hodgins
Re: [Mageia-dev] unable to mount encrypted partitions created with drakdisk
On Tue, 20 Mar 2012 20:57:47 -0400, simple w8 wrote: Hi, I use sometimes to boo Windows, and if i need some file i use to mount my /home encrypted partition using FreeOTFE (or TrueCrypt) to access the files i need, but i saw today that the /home partition that was created in mageia is not possible to mount with FreeOTFE or TrueCrypt. But the /home partition created with mandriva drakdisk can be mounted with those 2 applications, so seams that Mageia drakdisk isnt doing something right. That's likely a result of https://bugs.mageia.org/show_bug.cgi?id=3092 With the default settings, cryptsetup uses cbc mode encryption, which makes it much more likely that it can be cracked. See http://clemens.endorphin.org/nmihde/nmihde-A4-os.pdf for an explanation of how this is done. In Mageia's diskdrake, the setting is overridden to use xts mode, instead of cbc mode. If you prefer to use the less secure method, in order to keep it compatible with the windows applications, you'll have to re-encrypt it manually with cryptsetup luksFormat /dev/sd??. Don't forget to backup the data first, as this will erase that device. Regards, Dave Hodgins
Re: [Mageia-dev] [Freeze] please let in xonotic
On Sun, 11 Mar 2012 08:55:22 -0400, Manuel Hiebel wrote: Le dimanche 11 mars 2012 à 14:47 +0200, Sander Lepik a écrit : > According to sophie it's not a new package: > http://sophie.zarb.org/rpms/810fca5bd939c8e4a6858a5711c5200f > Yeah, i take some of my words back. But there is another problem - 0.6.0 seems to be in mdv2010.2. So upgrade path from Mandriva is broken. Do we support upgrade from mandriva backport ? I'am not sure. (also maybe we should not look anymore what they are doing... and remember that the desktop support is eol: http://www.mandriva.com/en/support/lifecycle/ ) I meant to raise this topic a few days ago, but got busy with other things. My understanding was that Magea 1 supported updating from Mandriva 2010.2 + Contrib + PLF-Free. There were several updates submitted to qa a few days ago, that are only available in backports in Mdv 2010.2. If we start accepting those under the exceptions rule, that allows packages that were present in Mdv 2010.2 but missing in Mageia 1, I'm very concerned that we could run into a lot more bug 2317 problems, which would be very difficult to avoid. Either we'll have to rewrite our dependency checking scripts to figure out which packages need to be linked from Core Release to Core Updates (not simple), or link everything in Core Release to Core Updates, or change Core Release to be tagged as an update medium, which would have a massive impact on cpu usage for mgaapplet. See https://bugs.mageia.org/show_bug.cgi?id=4814 for example. Opinions? Regards, Dave Hodgins
Re: [Mageia-dev] Mageia support for GMA 3600 (Cedar Trail Atom)
On Thu, 08 Mar 2012 04:52:46 -0500, wrote: Thanks but neither nokmsboot nor i915.modeset=0 make much of a difference. With nokmsboot I get the mageia splash screen for a second Probably best to open a bug report then. My guess is that it needs a firmware file, but I have no idea which one. Regards, Dave Hodgins
Re: [Mageia-dev] Mageia support for GMA 3600 (Cedar Trail Atom)
On Wed, 07 Mar 2012 21:00:38 -0500, wrote: As far as I understand there is frame buffer support in the 3.3 kernel for the GMA 3600 so in theory it should work. Oops. Sorry, that second one should have been i915.modeset=0 Regards, Dave Hodgins
Re: [Mageia-dev] Mageia support for GMA 3600 (Cedar Trail Atom)
On Wed, 07 Mar 2012 21:00:38 -0500, wrote: On Thu, 8 Mar 2012 tux99-...@uridium.org wrote: Hi, I just tried Mageia 2 beta 1 on a Intel DN2800MT board hoping that it would have framebuffer support for the GMA 3600, but the attached monitor goes into standby (no signal) as soon as the frame buffer gets activated during boot. Try disabling kms by adding either nokmsboot or modeset.i915=0 to the kernel options. Regards, Dave Hodgins
[Mageia-dev] Problems with systemd and /media.
Since systemd puts /media on tmpfs, specifically in /lib/systemd/system/media.mount, it's causing problems on upgrade of a system with fstab entries that have mount points in /media. https://bugs.mageia.org/show_bug.cgi?id=3550 At present, we have an errata entry regarding this https://wiki.mageia.org/en/Mageia_2_Errata#Base_system Currently the bug is assigned to the installer, with the idea that those fstab entries should be automatically removed. I've since been thinking, it might be better if either systemd, or dracut could parse /etc/fstab, and automatically recreate any mountpoints under /media. Opinions? Regards, Dave Hodgins
Re: [Mageia-dev] New dracut - please test
On Sat, 25 Feb 2012 05:59:39 -0500, Colin Guthrie wrote: Can you test that generating a new initrd after booting via dracut (and thus getting a hostonly one) does actually activate your usr partition? It's working now with a hostonly initramfs. Thanks! Still can't access the dvd due to the pata_acpi loading before pata_via. Regards, Dave Hodgins
Re: [Mageia-dev] Order of nics when multiple network interfaces detected.
On Fri, 24 Feb 2012 06:46:07 -0500, Colin Guthrie wrote: Not really no, but perhaps you're now using NetworkManager and it's somehow not updating the persistent rules? It's udev. The file /lib/udev/write_net_rules is missing. Regards, Dave Hodgins
Re: [Mageia-dev] New dracut - please test
On Fri, 24 Feb 2012 06:42:01 -0500, Colin Guthrie wrote: The other big change here is to automatically generate a much bigger initramfs when doing an upgrade from mga1. This will include a lot more stuff (e.g. lvm, raid etc) that may or may not be needed on a given setup, but until you boot with dracut you cannot generate an initramfs that will be able to detect only what is needed for boot. Looking at the current version of the init script, it's clear what the problem is ... check_finished && break udevsettle check_finished && break The above statement will always be true on a single core system, so the following code never gets executed. if [ -f $hookdir/initqueue/work ]; then rm $hookdir/initqueue/work fi for job in $hookdir/initqueue/*.sh; do [ -e "$job" ] || break job=$job . $job check_finished && break 2 done Regards, Dave Hodgins
Re: [Mageia-dev] New dracut - please test
On Fri, 24 Feb 2012 06:42:01 -0500, Colin Guthrie wrote: The other big change here is to automatically generate a much bigger initramfs when doing an upgrade from mga1. This will include a lot more I just tried it running dracut in a chroot from mga1, and got the bigger initramfs. It still fails to mount /usr, unless I break at pre-pivot and manually run lvm vgchange -a y I think that should be done before looking for the root filesystem, or swap, etc. I also still get the pata_acpi loading before pata_via, which makes my hard drives hd? instead of sd?, and stops /dev/sr0 from being created. Regards, Dave Hodgins
[Mageia-dev] Order of nics when multiple network interfaces detected.
In Mageia 1, which nic is assigned eth0 or eth1 etc, is controlled by /etc/udev/rules.d/70-persistent-net.rules. In beta 1, /etc/udev/rules.d/70-persistent-net.rules is not being created, although I can't figure out why as /usr/lib/libDrakX/network/connection/ethernet.pm has not been changed. I've also noticed that dracut doesn't include /etc/udev/rules.d/70-persistent-net.rules in the initrd, if it does exist. Is there some new method systemd is supposed to be using to control the order of nic assignments? As per https://bugs.mageia.org/show_bug.cgi?id=4352 it isn't working. I can't see what's change to prevent the rule from being created, so I don't know which rpm package the bug report should be for. Can someone who knows this part take a look? Thanks, Dave Hodgins
Re: [Mageia-dev] Unetbooting and extlinx - strange issue
On Sun, 19 Feb 2012 05:42:46 -0500, Robert Fox wrote: After installing unetbootin on my updated Cauldron box: When I start unetbootin - I get an error message stating that extlinx is not found - but oddly enough, it is installed . . . chmod a+x /usr/sbin/extlinux The executable has wrong file permissions. Reported as https://bugs.mageia.org/show_bug.cgi?id=4627 Regards, Dave Hodgins
Re: [Mageia-dev] Problem with rpm -q --whatrequires
On Tue, 21 Feb 2012 04:27:17 -0500, Colin Guthrie wrote: Is this because the kernel requires it as Requires(pre) and thus it doesn't show up? Should it show up? I don't know if it should show up for rpm or not, but it works with urpmq ... [root@hodgins /]# rpm -q --whatrequires dracut no package requires dracut [root@hodgins /]# urpmq --whatrequires dracut dracut kernel-desktop-3.2.6-3.mga2 kernel-desktop586-3.2.6-3.mga2 kernel-linus-3.2.6-1.mga2 kernel-linus-3.2.7-1.mga2 kernel-netbook-3.2.6-3.mga2 kernel-rt-3.2.6-0.rt13.1.mga2 kernel-server-3.2.6-3.mga2 kernel-tmb-desktop-3.2.6-1.mga2 kernel-tmb-desktop586-3.2.6-1.mga2 kernel-tmb-laptop-3.2.6-1.mga2 kernel-tmb-server-3.2.6-1.mga2 kernel-vserver-3.2.6-1.mga2 kernel-vserver-3.2.7-1.mga2 Regards, Dave Hodgins
Re: [Mageia-dev] About dm
On Thu, 16 Feb 2012 17:06:02 -0500, Luc Menut wrote: btw, each DE should requires its preferred DM. I disagree with that. A suggests it's own dm with a require for some dm is all that's needed. If I install multiple desktops, I don't need or want multiple desktop managers. Regards, Dave Hodgins
Re: [Mageia-dev] Mounting a CDROM problem
On Thu, 16 Feb 2012 04:20:32 -0500, Robert Fox wrote: responded: The requested operation has failed.: Error mounting: mount exited with exit code 1: helper failed with: mount: mount point /media/cdrom does not exist Do you have an entry in /etc/fstab for /media/cdrom? If so, try removing it. https://wiki.mageia.org/en/Mageia_2_Errata#Base_system Regards, Dave Hodgins
Re: [Mageia-dev] New Dracut: Please test
On Wed, 15 Feb 2012 19:10:09 -0500, Colin Guthrie wrote: Can you try with dracut-015-3.mga2 when it arrives? See https://bugs.mageia.org/show_bug.cgi?id=4537 Regards, Dave Hodgins
Re: [Mageia-dev] New Dracut: Please test
On Wed, 15 Feb 2012 19:10:09 -0500, Colin Guthrie wrote: OK, that was a very strange quirk it all boiled down to a string comparison function :s Can you try with dracut-015-3.mga2 when it arrives? Still doesn't work. If you cannot boot, you can just pass rd.lvm.lv=foo/bar on the kernel command line (where foo is the vgname and bar is the lvname)... tho' this might or might not work too well as lvm might not be on the initrd The lvm command and the dm_mod kernel module are not in the initramfs. I'm using chroot from Mageia 1, to access the system. Regards, Dave Hodgins
Re: [Mageia-dev] New Dracut: Please test
On Tue, 14 Feb 2012 10:55:17 -0500, Colin Guthrie wrote: Can everyone please test the new dracut please? Especially those of you With dracut-015-2, / on a partition, /usr on a lvm logical volume, the initramfs no longer has the lvm2 command, or the dm_mod kernel module making it impossible to mount /usr. The file 50mount-usr.sh is present in hooks/pre-pivot, but the file 30parse-lvm.sh is no longer in hooks/cmdline. Regards, Dave Hodgins
Re: [Mageia-dev] New Dracut: Please test
On Tue, 14 Feb 2012 10:55:17 -0500, Colin Guthrie wrote: Can everyone please test the new dracut please? Especially those of you I'll test it shortly. I think there is a slight problem when dracut gets updated at the same time as the kernel, udev, or anything else that is going to get installed in the initramfs. Rather then triggering the running of dracut when the kernel gets installed, I think it would be better to have something that runs at the end of urpmi or MageiaUpdate, that check to see if dracut or anything in the existing initramfs has been updated, and if so, then run dracut. What do you think? Regards, Dave Hodgins
Re: [Mageia-dev] Issues with dracut
On Tue, 17 Jan 2012 07:22:30 -0500, Colin Guthrie wrote: Are things working OK for you now with dracut or is it still busted? Just to clarify why I think the problem is happening on single core systems. On a multi-core system, the bash and udevd processes will be running on different cores. When the script executes the udev settle command, it continues to execute, so the loop checking to see if udev is done finds it isn't, so it then looks for/runs the initqueue jobs. On a single core system, the bash script waits for the settle command to finish, so then finds it's done, and exits without even trying to run the initqueue jobs. The patch in my prior message is effectively changing the script from "udev done or jobs done" to "udev done and jobs done". Regards, Dave Hodgins
Re: [Mageia-dev] Issues with dracut
On Tue, 17 Jan 2012 07:22:30 -0500, Colin Guthrie wrote: Are things working OK for you now with dracut or is it still busted? I've finally figured out why it wasn't working for my system. Two separate problems. First, /dev/$volumegroup/$logicalvolume symlinks are no longer created. My fstab had ... /dev/91/usr /usr ext4 defaults,relatime,user_xattr 1 2 which used to work, but /dev/91/usr is not created when the lvm scan is run. /dev/mapper/91-usr is created, but not /dev/91/usr. Second, on my single core system, the loop to process the things found by udev only ran while :; do check_finished && break at which point it ended the loop, without trying to run the initqueue jobs. I've attached a patch to fix the init script. Would you prefer I open a new bug report for this? Regards, Dave Hodgins init.patch Description: Binary data
Re: [Mageia-dev] Too many tmp directories
On Sat, 21 Jan 2012 12:35:39 -0500, Balcaen John wrote: while searching for some strange KDE behaviour (see bug 107 about those icons), I found at least three different tmp directories: /tmp/ /var/tmp/ ~/tmp/ (one for every user) http://en.wikipedia.org/wiki/Filesystem_Hierarchy_Standard#Directory_structure /tmp - files may not survive reboot. /var/tmp - files may survive reboot ~/tmp- files may survive reboot, also subject to things like quotas, for regular users. Regards, Dave Hodgins
Re: [Mageia-dev] [changelog] [RPM] 1 core/updates_testing firefox-9.0-0.1.mga1
On Thu, 29 Dec 2011 12:20:36 -0500, Manuel Hiebel wrote: (and iirc QA don't check it anymore, as you can see in the updates of firefox 6 or later) We do test them when they become available, but won't hold back a security update for firefox, waiting for the extensions. Regards, Dave Hodgins
Re: [Mageia-dev] [bugs] [Bug 2111] transcode, A linux video stream processing utility
On Sat, 24 Dec 2011 06:51:29 -0500, Oliver Burger wrote: And I do know, many plf packages are in tainted (that's why we do have tainted), the question remains, if a package being in plf2010.1 qualifies for pushing to updates. My understanding is that packages in 2010.2 plf free can go to tainted updates, but plf nonfree cannot.
Re: [Mageia-dev] Issues with dracut
On Sun, 18 Dec 2011 17:22:46 -0500, Colin Guthrie wrote: Can you run the attached? It works. It identifies /usr as being on 252:9, which is on 8:12, which has the fs_type LVM2_member. Regards, Dave Hodgins
Re: [Mageia-dev] Issues with dracut
On Sat, 17 Dec 2011 22:00:51 -0500, Colin Guthrie wrote: The shell script you'll need to tweak is /usr/lib/dracut/modules.d/90lvm/module-setup.sh. That is the problem. In that module, it finds /usr on dev/block/252:9, and the call to get_fs_type correctly returns the file system type as ext4, but the module-setup.sh is looking for LVM2_member which will never be returned for a mounted file system. I don't see how it could be working on your system. Regards, Dave Hodgins
Re: [Mageia-dev] mariadb
On Sat, 17 Dec 2011 19:12:21 -0500, Maarten Vanraes wrote: oh and were any mysql libraries still loaded? ie: programs linked to mysql, after upgrade? See https://bugs.mageia.org/show_bug.cgi?id=3804 Regards, Dave Hodgins
Re: [Mageia-dev] mariadb
On Sat, 17 Dec 2011 19:12:21 -0500, Maarten Vanraes wrote: that is a lot of memory, do you have any mysqld logs? and also, for the record, you have mariadb installed? This is from akonadi under kde4. I'll look to see if I can find any logs from that. can you check if you have the latest, if there is still a part of mysql installed? Yes, it's mariadb-5.5.18-0.bzr3169.20111216.3.mga2. what kind of databases do you have, and which ones were created by mariadb? what kind of load do you have? what app is querying mariadb? do you have a slow-queries mysql log file? how big are the recent mysql-binlogs? perhaps you can make a bug report, this would be easier to follow up on. oh and were any mysql libraries still loaded? ie: programs linked to mysql, after upgrade? I'll open a bug report, and attach any log files I can find. As I only have a few seconds after kde starts, before the system starts swapping everything out, I can really only get debug output by putting commands in autostart, or by looking at files from my mageia 1 install. Regards, Dave Hodgins
Re: [Mageia-dev] Issues with dracut
On Sat, 17 Dec 2011 15:21:14 -0500, Colin Guthrie wrote: Dracut should install all dynamic libs needed by apps. Which app needs this library specifically? Do you have /usr on a logical volume? I've figured out now, that /lib/dracut/hooks/pre-pivot/50mount-usr.sh is failing, as lvm has not yet been activated. Later, I'm dropped to a shell, after the switch to the new root, and since /usr hasn't been mounted, any commands that try to load /usr/lib/libncursesw.so.5 such as lvm, grep, or less, will fail. When I've copied /usr/lib/libncursesw.so.5* to /lib and run ldconfig, it still fails in 50mount-usr, but succeeds later, after the root switch. The dm-mod module loading, and the "lvm vgchange -ay --sysinit" command should be run before looking for the root file system, as it too could be on a logical volume. Regards, Dave Hodgins
Re: [Mageia-dev] Issues with dracut
On Sat, 17 Dec 2011 15:21:14 -0500, Colin Guthrie wrote: Dracut should install all dynamic libs needed by apps. Which app needs this library specifically? When I tried to boot after running dracut with the default config, I was dropped to a shell, with the non-lvm root mounted. I then tried running "/sbin/lvm vgchange -a y --ignorelockingfailure", and it failed due to the missing libncursesw.so.5 module. Now that I think about it, a bit more, I think I had already been chrooted into the root, so since /usr hadn't been mounted, of course the module could not be accessed. Something else must have gone wrong before the chroot, that I didn't see. So I've got the necessary lib... I wonder if this is a side effect of me having lib64ncursesw-devel-5.9-2.mga2 installed (i.e. the dev package)? Perhaps it's someone the key to including the right lib? Perhaps you can test by installing the 32 bit devel package and seeing if the lib is then included automatically? I'll test that, and report back. Regards, Dave Hodgins
Re: [Mageia-dev] mariadb
On Sat, 17 Dec 2011 11:10:29 -0500, Maarten Vanraes wrote: I mentioned that this library should be versioned, and that it's not at this time, he agreed, and said that since it's ABI compatible with mysqlclient, that he should not use 0 major, but the 18 major (same as mysqlclient) the code is still 100% exactly the same, it just went from unversioned in upstream to versioned. since we had versioned manually mysqld to major 0, i provided the compatibility symlinks for it. I've now figured out why my cauldron install is locking up. I managed to get htop running before it locked. It showed swap usage climbing. I got it to sort by vmsize. It showed mysqldb using over 2.5GB MV, and 1.5GB ram, on my 2GB ram system. Regards, Dave Hodgins
