Re: [Cooker] KDE 3.2 beta1 - KDM Session Type Selection
Michael Lothian wrote: I'm suffering the same problem. I just installed KDE 3.2 beta 1 from Cooker (kdebase-kdm 3.1.93 -23mdk) and I have found a problem with KDM session type selection. It just doesn't work. It always starts IceWM even if I select something else. Is this a known problem? Is there any fix available for it? It goes further. If you select GNOME during the first-time wizard, then logout and login, you get a default of KDE. It is after that that everything goes to IceWM.
Re: [Cooker] KDE 3.2 beta1 - KDM Session Type Selection
[EMAIL PROTECTED] wrote: For me it works fine if I choose Default. Well, it depends on what you want to get for Default. If I go through the drakfw wizard and choose GNOME, that's what I expect to get the next time. But that aside, the problem is that no matter what WM you choose explicitly from the (I assume) MDKKDM submenu, all you get is ICEWm. Maybe this doesn't happen if all you ever try to use is KDE.
[Cooker] No KDE app menu entries in GNOME ?
I've just done a new installation from a Monday (Nov. 10) refresh of cooker, and I'm finding that the launcher in GNOME no longer contains any entries for KDE apps, such as Kate, KCalc, etc. I'm sure these were there in the 9.2 RCs, since I always add them to my panel. Have these disappeared by design, or is this something that needs to be corrected ?
[Cooker] Orphan files ?
I'm looking for some guidance on trying to find the package which owns a file. Several times now when I've had a problem with a particular executable or one of the initialization script files, I've tried to find the owning package for purposes of filing a bug report by using rpmdrake's seach on filename feature, only to get no hits from the search. The latest example is /usr/bin/update-menus. Should this technique work, or should I be doing something else ?
Re: [Cooker] Orphan files ?
James Sparenberg wrote: urpmf /usr/bin/update-menus will give you the package that includes that file. Michael Altizer wrote: It seems as though this should work. Using urpmf --files update-menus returns the following results: menu:/usr/bin/update-menus menu:/usr/share/man/man1/update-menus.1.bz2 Hmm. It appears that I have a /usr/bin/update-menus without having the menus package installed. I've also got the man page. However If I do what you suggested, I get: [EMAIL PROTECTED] ftg]# urpmf --files update-menus majesty-demo:/usr/local/games/maj_demo/.manifest/scripts/update-menus.sh mindrover-demo:/usr/games/mindrover_demo/.manifest/scripts/update-menus.sh [EMAIL PROTECTED] ftg]# urpmf /usr/bin/update-menus [EMAIL PROTECTED] ftg]# urpmi menu Everything already installed [EMAIL PROTECTED] ftg]# In addition, if I look in my cooker tree, I have menu-2.1.5-124mdk.i586.rpm but neither rpmdrake install nor rpmdrake remove lists the package when I select list all packages, alphabetical. A urpmi.update -a made absolutely no difference in these results. Is there some lower-level query I can do to discern why urpmi thinks the package is installed and I can see that the files are present, but rpmdrake thinks it is neither here nor there ?
Re: [Cooker] Orphan files ?
James Sparenberg wrote: An alternative. rpm -q --whatprovides /usr/bin/update-menus [EMAIL PROTECTED] ftg]# rpm -q --whatprovides /usr/bin/update-menus menu-2.1.5-124mdk
Re: [Cooker] Orphan files ?
Leon Brooks wrote: $ rpm -qf /usr/bin/update-menus menu-2.1.5-123mdk [EMAIL PROTECTED] ftg]# rpm -qf /usr/bin/update-menus menu-2.1.5-124mdk
Re: [Cooker] Orphan files ?
Leon Brooks wrote: Try this: rm -vf /var/lib/rpm/__db.00* rpm --rebuilddb I'll be happy to, but one of the reasons to post this was to see if anybody wanted a snapshot of the existing files that seem to cause urpmi to recognize that the package is installed but neither urpmf nor rpmdrake to recognize the files included in it, or in the case of rpmdrake, that the package even exists (since it isn't recognized as being either installed or not). If somebody can tell me which files to copy off to the side for inclusion with a bug report, I'll be happy to move on. If the responsible parties aren't interested, then please just say so, and I'll abandon trying to preserve evidence
Re: [Cooker] Test - please ignore
I have just done a Reply and not a Reply All so you can see that something is getting through. -- Regards Frank Big or small, a challenge requires the same commitment to resolve. Registered Linux User # 324213 Quel Qun wrote: I got two successive delivery failure :(
Re: [Cooker] [9.2] misleading : Press Y within 5 seconds File system check
FACORAT Fabrice wrote: I suffer from this too. But normally this has been corrected. If you launch filesystem check, the system should reboot automatically after. There are two interpretations for misleading here. The original complaint was that the message gave the impression you had to fsck an ext3 system. But the message is misleading even for ext2, since it doesn't accurately convey what happens if you do or do not press Y. As far as I can tell, if you don't press Y when the message appears during the first boot after an unclean shutdown, it runs e2fsck and reboots anyway. On the reboot, you'll get the same message. But this time, if you don't press Y, it will find that the partition is now clean and continue the boot. If you do, I think it redoes the e2fsck and reboots again (this is from memory, I'm not sure of it). I'm guessing that the old code drove the message solely from whether the partition was marked unclean or not, so that if it ran fsck and rebooted, you wouldn't get the message at all during the reboot. It looks like the new code separates the state of the partition from how you respond to the message, and will only consider the state of the partition if you don't press Y. If this is the case, then on the reboot a better message would be Your system appears to have shut down uncleanly, and it has not yet successfully rebooted although the root filesystem has been checked and now appears clean. Press Y within 5 seconds to force another check/reboot cycle, otherwise this boot will continue. Is there any point to even showing the message on the reboot if the root partition now shows as clean ?
Re: [Cooker] Boot-up speed
Buchan Milne wrote: Only load services when they're needed (or after gui is up) I think that's what WinXP does. It's not that simple. For me to be able to log in (on my desktop in a LAN), I need at least NFS (ie portmap, nfslock) and autofs up and running. For a disconnected LDAP setup (see the article on mandrakesecure.net), I need LDAP up and running too (otherwise is is not possible to login, or show user lists if using user lists on the dm). I'm guessing here, but when booting Mandrake on a reasonably fast machine (2.4GHz), the major delays I see are the depmod and activities which are looking for things which aren't there, such as NFS mounts from other machines which aren't up, or LAN adapters which aren't in use (e.g. a laptop NIC which is only used in-home or in-office). When these things ARE there, startup proceeds quickly. The problem may be the timeouts involved in looking for things you're not going to find. This is *definitely* a problem in shutdown. If, for example, my desktop mounts an NFS directory from my laptop, and I then shut the laptop down, shutting down the desktop will retry umount for the mount 3 times while shutting down NFS, and another 3 times during the final unmounting file systems phase. Each retry takes something like 90-120 seconds. Granted, on the particular issue of NFS, it's probably possible to tweak things to make this better, but that doesn't hold in every case. My laptop, when used in-house, often has its on-board NIC plugged into my internal LAN. If it doesn't (when on the road), there is a delay of at least a minute trying to bring up eth0. Part of the solution may simply be to revisit these timeouts, and adopt the philosophy that in today's networks, if you don't find it in a few seconds, it's probably not there, and you should either abandon the attempt or else background it. Another part of the solution would be a way to let particular components block on the availability of others, so that one component's initialization can proceed asynchronously, and a second component only blocks from the time it decides it requires the service until the time it is ready. A partial solution would be to create a few parallel initialization streams each of which serially runs components which depend upon one another.
Re: [Cooker] Boot-up speed
Jos wrote: I did some testing myself, working on something similar, though I still don't know what people prefer: that I hack init, or replace /etc/rc.d/rc by a binary... Is the problem that the necessary initializations take too long, or that they take too long when run serially under a single thread ?
[Cooker] No kernels on uninett ?
I resync'd cooker with ftp.uninett.no about an hour ago, and found that all of my .15 kernels were deleted with no replacements. Sure enough, all of the cooker kernel files (except for the docs) are gone from the mirror. I've continued to check over the past hour, so this isn't a case of an update in-flight.
Re: [Cooker] 9.2 cannot produce a floppy boot disk.
Randy Welch wrote: Only problem is that it will only find the first installed linux on your disk. My test installs are always in later partitions. The rescue image tends to find the first one it sees. (so on my system it would see my 9.1 install instead of the 9.2... I almost never tell rescue to mount partitions for me. Just go to console, and mount the correct partition on /mnt directly.
Re: [Cooker] 9.2 cannot produce a floppy boot disk.
Ron Stodden wrote: Buchan Milne wrote: Rescue is available for network and hd installs. Have you tried it? I have used it a few times on a network install. That´s news to me. Pray tell us how rescue is initiated from an hd or network install, and where that is so unmissably documented that I must have missed it ... AFAIK, all of the boot paths (floppy images, ISOs) give you the F1/Enter option when they start.
[Cooker] Isolinux broken
In current cooker (and for several days now), booting isolinux to do a hard drive install fails on my notebook. You go through specifying the drive location, get second stage and probing serial ports, several lines of stuff that flies by too fast to read, and the a dead black screen. A hard drive install using hd.img from the same cooker tree doesn't have this problem.
[Cooker] Incomplete 22.12 kernel move
I updated cooker this morning from uninett.no several times until I got no activity, and have been updating with no activity for several hours now, so I believe I have everything available. The rsync update deleted the .10 kernel and downloaded the .12 kernels, but when I copy hd.img to a floppy and try an install, I get There was an error installing packages: kernel-2.4.22.10mdk-1-1mdk.i586 Somebody is still looking for the old one.
[Cooker] MakeCD creating 9 ISOs ?
Apologies if I somehow missed this thread, but how come ./MakeCD -t /data -a /data/cooker/i586 which used to produce 4 650MB ISO images is now producing 9 ? Is mkcd including all of contrib now ?
Re: [Cooker] MakeCD creating 9 ISOs ?
Greg Meyer wrote: use -c switch to control number of iso's. Yes, but without knowing how the content has changed from the days when the same command produced 3 full CDs and one partially full CD, it's hard to pick the correct number.
Re: [Cooker] resigning pending
Warly wrote: For rc2, but there should not be many packages unsigned. You might want to spell that re-signing to avoid scaring the hell out of us
Re: [Cooker] Expert install from isoliunx ?
Buchan Milne wrote: Hmmm, you should have given us this info before ... since you can't use both isolinux and lilo (they are both bootloaders, isolinux is for booting from CDs), I assume you are bootstrapping your install from lilo? If so, you need to provide the parameters isolinux would normally provide to the kernel (look in the syslinux.cfg) via the append line of your lilo.conf Thanks, I added expert to the append line, and that did the trick.
Re: [Cooker] about shorewall update
J.A. Magallon wrote: I think shorewall does something strange on update. I had shorewall installed on a box, but not active. After an update with urpmi, the updated package auto activated itself Shorewall has become a major pain since sometime in the 8.2 - 9.0 timeframe when somebody decided to make the default setting pass absolutely nothing. Before that, I used to install and enable it just to see whether my firewall was doing a good job. Since then, I install it, but disable it, because I'm not willing to do the work to re-research which ports ought to be open and which oughtn't. It's interesting that when you use the MDK configuration tool for Shorewall and select pass everything, MDK disables Shorewall completely. You can see this by selecting Shorewall during install and then selecting pass everything during Summary configuration. When you get back to Summary, you'll see Firewall disabled, and the drakservices menu will show Shorewall with the Start on Boot checkbox unchecked. Frankly, if I want a firewall, then I want it configured with some intelligent choices which I can use as a base from which to work. I don't want to have to redo the research the authors did just to figure out what can be safely re-enabled. Just FYI, I've always used Standard security level, both before and after Shorewall became unusable out-of-the-box.
Re: [Cooker] about shorewall update
Adam Williamson wrote: Block everything is just about the most sensible default I can think of, given that there's no port that all or even most users will want open. I guess it depends upon what you're doing. In my case, the default settings blocked traffic to my gateway, both in and out, and effectively shut down the network.
Re: [Cooker] Expert install from isoliunx ?
Guillaume Cottenceau wrote: Frank Griffin [EMAIL PROTECTED] writes: Sorry if this is a dumb question, but is there any way to do an expert install from an isolinux boot ? I can't seem to find anyplace to request it F1 then linux expert? That's how I'm used to doing it from floppy image boots, but I tried twice and couldn't find anyplace in the isolinux sequence which mentioned the possibility of hitting F1. Where do I do that ?
Re: [Cooker] Expert install from isoliunx ?
Guillaume Cottenceau wrote: That's how I'm used to doing it from floppy image boots, but I tried twice and couldn't find anyplace in the isolinux sequence which mentioned the possibility of hitting F1. Where do I do that ? In the bootloader stage. Very first graphical image of mandrake. When you computer boots. I've tried leaning on F1 at every possible place, with no effect. Can I put something in the LILO entry for booting isolinux ?
Re: [Cooker] urpmi --auto-select asks too many questions
Paul Dorman wrote: On Thursday 04 September 2003 10:19, Liam Quin wrote: However, if I'm asked, say, package has bad signature, continue anyway? I might be asked that question 10 times. The problem is that I want to leave a urpmi --auto-select running, not having to check up on it every few minutes. You can add the missing keys yourself with two commands: There are also several packages currently which have either invalid signatures (not the entire directory as before the new keys were brought into urpmi) or no signatures at all. You can bypass signature prompts with the --no-verify-rpm switch.
Re: [Cooker] mirrors ? cannot rsync with ftp.sunet.se
Adam Williamson wrote: On Mon, 2003-09-01 at 13:14, Dave Cotton wrote: On Mon, 2003-09-01 at 12:31, Adam Williamson wrote: Well, it was a Sunday yesterday... And so should have given mirrors a chance to catch up if they could. I mean, there's not necessarily going to be any new packages on a Sunday. That wasn't the pattern. ftp.sunet.se had been updating fine on Friday and Saturday. On Sunday, about 200MB of packages disappeared with no replacements, and they're only coming back now as I write this. This is not normal mirror activity. I just wish someone would explain what is going on.
[Cooker] Expert install from isoliunx ?
Sorry if this is a dumb question, but is there any way to do an expert install from an isolinux boot ? I can't seem to find anyplace to request it
Re: [Cooker] Whole Day Shot Trying HD Network Installation
Felix Miata wrote: Eventually I got set up for NFS install, and that won't even get out of text mode. Each time, right after entering server IP and directory it just sits there, and on tty3, last line I see is 'preparing nfsmount for [IP]:[dir]', and no evidence that SCSI ever loaded. Has SCSI support been deleted from the boot floppies? I can't remember when the previous versions asked for SCSI support confirmation. The network installs (actually, all installs) support SCSI. I recently did one to an aic79xx SCSI system, and that driver was only included during this cycle. It should probe your system and put up popups saying which driver it's trying. It sounds like your NFS setup is somehow wrong; check the syslog of the system which owns the exported drive. Is ftp or CD install the only way that works for rc1? I really don't like burning CD's for betas, and I've never figured out how to select a ftp or http source for installation. Where do I point the installer for ftp or http? The easiest way for all these installs, as explained to me previously here, is to boot the isolinux/alt0 image. Then you can pick the install you want (HD, NFS, HTTP, etc.) and get prompted for the needed info. Put the following entry in lilo.conf, where /data/cooker is the root of your cooker tree: image=/data/cooker/i586/isolinux/alt0/vmlinuz label=isolinux root=/dev/ram3 initrd=/data/cooker/i586/isolinux/alt0/all.rdz append=ramdisk_size=32000 read-only You don't need to have a cooker tree; you can just retrieve vmlinuz and all.rdz from i586/isolinux/alt0 at a cooker site, put them somewhere, and point lilo to them. You need to rerun LILO each time these get refreshed or moved.
Re: [Cooker] Re: Main mirror BROKEN AGAIN!!!!!
David Walser wrote: Did sunet ever fix their timestamps? Last I checked (it wasn't recently) all of sunet's timestamps were off by an hour. I don't know. When I switched to it, it downloaded quite a bit, and that may be why.
Re: [Cooker] Main mirror BROKEN AGAIN!!!!!
Austin wrote: Is uninett.no the only server with rsync ? No, ftp.sunet.se uses it as well, and at the moment appears to be in much better shape than uninett. My sync hasn't finished yet, so I don't know how good much better actually is...
Re: [Cooker] More mirror madness
Adam Williamson wrote: I see about five people posting to the list moaning about the mirrors, so I don't post to the list moaning about the mirrors. Why would I? I think the point is that what you *don't* see is a single post saying this is what's going on, and this is when we think it will be fixed. Or even this is what's going on (draw your own conclusions). So the rest of us sit here and repeatedly hit the servers every half-hour or so, hoping to scrape up the RPMs that bugzilla tells us have fixed the bugs we've reported or that are needed by the other half of the RPMs that actually made it through but have dependencies on those that didn't. The alternative is to wait for the next set of ISOs to be released, and figure that maybe any mirror which has them has actually stabilized. In which case, why bother to try to maintain local cooker mirrors and test all the stuff that's already happened by the time an ISO is cut ? Paul Dorman has a valid point. I work in a development environment where development (and not IT management) is the priority. But if I pulled something like this, they wouldn't find enough of me left to fire, and it wouldn't be management with blood dripping down its chin. I just wish I understood enough about the architecture behind the layer of the primary mirror to be able to offer constructive criticism.
Re: [Cooker] resigning some packages
Warly wrote: Some mirroring traffic may be generated. I guess that explains why about 400MB of RPMs disappeared in the last hour from uninett with no replacements ! :-)
[Cooker] Missing gmp, klips packages in contrib
[EMAIL PROTECTED] i586]# urpmi super-freeswan installing /data/contrib/i586/super-freeswan-1.99_kb2c-1mdk.i586.rpm Installation failed: gmp is needed by super-freeswan-1.99_kb2c-1mdk klips = 1.99_kb2c is needed by super-freeswan-1.99_kb2c-1mdk klips-alg = 1.99_kb2c is needed by super-freeswan-1.99_kb2c-1mdk klips-natt = 1.99_kb2c is needed by super-freeswan-1.99_kb2c-1mdk [EMAIL PROTECTED] i586]# There are no gmp* or klips* packages in either cooker or contrib...
Re: [Cooker] Missing gmp, klips packages in contrib
Luca Berra wrote: the subject should have read: broken super-freeswan package in contrib. super-freeswan isn't broken unless it doesn't really need those packages. If it needs them and they aren't there, then the problem is that they aren't there. Are there guidelines/rules for coding subject lines here ? If so, where, please ?
Re: [Cooker] Software submission for the Mandrake distribution
Pierre Jarillon wrote: I agree too. But I have another reason : such a software on the same computer is a fantastic SECURITY HOLE ! It allows to write on protected files. I remember such a software for windows95 which allows to acces to ext2. Please don't spread this software ... and use reiserfs or better the new reiser4. Sorry, but I disagree. Until it broke, I used a driver InstalledFileSystem on OS/2 to access ext2 partitions and exchange files. I still use such an IFS to access FAT32 partitions from OS/2. Back when OS/2 was my primary OS, this made Linux MUCH more usable for me (especially before the HPFS write support was stable). Nobody in their right mind sets up a multi-boot machine if it is really going to be multi-user on more than one of the OS's *AND* if users are going to be able to reboot it. Anybody who can reboot a multiboot machine is pretty much guaranteed to be the sysadmin for most (if not all) of the system images. For this reason, I'm not too concerned about cross-filesystem access ignoring permissions. I would echo the opinion of another poster that if the author has time to spare and the necessary expertise, it would be real nice to get the NTFS writable code working. However, I would not disparage or denigrate the work that he has done so far. For a Windows user looking to incorporate Linux, the ability to work initially in Windows using familiar tools will be perceived as a distinct advantage, even if he will have to rely heavily on dos2unix on the Linux side for text files.
[Cooker] Today's urpmi woes
Autoconf 2.13-18 isn't being recognized as 2.13: The following packages have to be removed for others to be upgraded: kdevelop-2.1.5-10mdk.i586 (due to unsatisfied autoconf == 2.13) libkdevelop2-2.1.5-10mdk.i586 (due to unsatisfied kdevelop == 3:2.1.5-10mdk) (y/N) but... [EMAIL PROTECTED] ftg]$ cd /data/cooker/i586/Mandrake/RPMS [EMAIL PROTECTED] RPMS]$ ls autoconf* autoconf-2.13-18mdk.noarch.rpm autoconf2.5-2.57-5mdk.noarch.rpm and [EMAIL PROTECTED] man]# rpm -q -f /usr/bin/autoconf autoconf-2.13-17mdk Apparently I have -17 installed, but -18 isn't seen as satisfying KDevelop's requirement.
[Cooker] Old cooker packages not being deleted ?
This could be my imagination, but it seems like starting either this morning or last night, rsync --delete updates from ftp.uninett.no aren't deleting any files in my local tree. Within 24 hours, my cooker RPMS directory (as reported by rsync) has gone from about 2.5 GB to 2.9GB, and I haven't been seeing any messages about older versions being deleted (and some of the updates involved most of KDE, which should have bumped a version number). Am I dreaming or is something amiss ?
Re: [Cooker] Gnome splash takes a LONG time to disappear
Robert Fox wrote: Fresh Cooker install - tried Gnome and it works but the initial splash screen stays up for a long time (like two minutes) then disappears. What causes this? I've seen this as well. It's OK the first time (after the firsttime wizard), but on subsequent logins it not only hangs around but covers anything else on the screen. I had a similar problem with post-9.1 2.2 GNOME (circa May 18) where logging in and out of different userids would eventually cause a hang when you requested logout. Well, not a true hang, but the logout dialog box would take about 5 minutes to come up. In both cases, the blockage seems to break if you initiate enough other GNOME-related activity. In the older logout case, opening a GNOME terminal would trigger it. I haven't noticed a pattern in the new case. It's like something is holding a lock and then decides to release it, or maybe a socket listener isn't listening closely enough...
[Cooker] Error in cooker tree links
When I updated cooker this morning, I noticed that two new links were created in cooker/i586/Mandrake: RPMS3 and RPMS4. However, they were created as ../../contrib/jpackage.org/generic/RPMS/ and ../../contrib/jpackage.org/mandrake/RPMS/ whereas the RPMS2 link which points to contrib is created as ../../../contrib/i586// Looks to me as though the new links should have an additional ../ prepended to them, and may need a i586/ added at the end. Also, the double slash at the ends of the RPMS2 link looks like an error (even though it doesn't seem to cause problems). The mirror involved was ftp.uninett.no.
[Cooker] rpmdrake vs. urpmi weirdness
I just tried to install the MandrakeClub Commercial FlashPlayer package on today's cooker with rpmdrake, and got the error: FlashPlayer-6.0-3mdk.i586 (due to unsatisfied rpmlib(PayloadFilesHavePrefix) = 4.0-1 I then went to command line and tried: urpmi --no-verify-rpm --allow-force FlashPlayer and it installed cleanly without a single complaint. I use --no-verify-rpm because of the current key incompatibilities in Cooker, and I know that urpmi has complained about unsatisfied conditions (see an earlier bug report today about printer-utils) even with this switch. So is rpmdrake imagining unsatisfied dependencies or is urpmi ignoring them ? Or am I just doing something dumb without realizing it ?
Re: [Cooker] How to dump current rpm db for install ?
Ben Reser wrote:
Not really. However you could do:
rpm -qa --qf '%{NAME}\n' package.list
Then run urpmi for each package name after the install:
for package in `cat package.list`; do
urpmi $package
done
You'll have some that won't exist anymore (libraries for example with
changing major names). But it ought to work for anything else.
Can't say I've done this myself...
I gave this a try importing rpm databases from various cooker systems
to my current one, and it seems to work fine. I did write some local
code to compare the package.list to a reference one resulting from a
vanilla install with all boxes checked and no individual packages
selected, because invoking urpmi for all of the already-installed guys
ate up considerable time, even on a fast machine.
Thanks !
Re: [Cooker] rpmdrake vs. urpmi weirdness
Charles A Edwards wrote: Had you not used --allow-force urpmi would have given you the same error as did rpmdrake. That's not what I would expect. From the man page: --allow-force Allow urpmi to ask user to continue installation using no depen- dencies checking or forced installation due to error. By default urpmi exit immediately in such case. Previously, I've had it display the errors and then prompt me as to whether I want to continue.
[Cooker] hd.img without floppy ?
Just an idle, uninformed question, but is there any way to tell a Mandrake install boot (whether from CD or floppy with maybe old hd.img on it) to reboot-in-place using the hd.img from a local disk cooker mirror ? The object being not to have to create new hd.img boot floppies all of the time...
Re: [Cooker] 9.2 MEDIAS - MEDIA
Pierre Jarillon wrote: IMHO, won't be fixed... for another reason: In latin, media is a plural. But we don't speak latin anymore. This word is imported in several langages. Then the word follows usual rules of its new langage. Sorry, but media is considered plural in English as well, with the singular being medium. Even if there are languages which have imported media as a singular, the complaint concerned only the English translation, and thus is valid (whether or not anyone intends to fix it).
Re: [Cooker] MDKKDM fails dependencies/ GDM mis-installed
Frank Griffin wrote: In this morning's cooker, MDKKDM fails install with the following error: mdkkdm-9.1-28mdk.i586 due to missing kdebase-kdm-config-file-3.1.3-4mdk.i586 An expert install installs GDM without indicating that MDKKDM had an error, and on reboot the resulting GDM initializes with an error popup saying Session directory is missing. The only indication of the MDKKDM error is if you open MCC and try to switch the DM to MDKKDM, at which point you're told that it isn't installed. If you run rpmdrake to install it, you'll get the dependency error. Looks like this was fixed in the time it took me to reboot and refresh cooker...
Re: [Cooker] rpmdrake vs. urpmi weirdness
Guillaume Cottenceau wrote: Is it reproducable with another package? Here I can see no problems.. I did a fresh install this morning, went to a root command line, added the Club Commercial site as a media, and did urpmi FlashPlayer This got the same error that I reported for rpmdrake. When I added --allow-force, it worked as before. So, the problem appears to be in the FlashPlayer RPM, which wouldn't be so annoying if swfdec actually worked on most of the swf's out there :-)
Re: [Cooker] 9.2 MEDIAS - MEDIA
Robert L martin wrote: And to bring it to point since RED HAT wouldn't dare attempt to sell a distro with some french word misspelled i think Mandrake should extend the same curtesy. I'm sure you've got a point here, but I obviously missed whatever it referred to... (and as requested in 4/4 time no accents and all ows same (sort of slur into the N on browN) Well, no, actually, they seem to pronounce most *ow* words as *aye*, as in hye nye bryene cye (with the ye pronounced as the English word high). Just a kind of inside joke; I host Northern Irish kids for several weeks in the summer (they just left), and I was feeling nostalgic...
Re: [Cooker] Flash plugin
Adam Williamson wrote: You can't do that without nuking most of your installation, as it uninstalls the other version of libstdc++, which tons of packages depend on. Actually, I was able to do it on current cooker, but not my stable cooker from May 18. On current cooker, you can force it on, and libstdc++5 doesn't get uninstalled. On the older cooker it did (and everything broke). Reinstalling libstdc++5 again left libstdc++2 in place, fixed the problems, and left a working Flash plugin. Someday I may actually understand this stuff.
[Cooker] Packaging errors as of today
The following occur in today's cooker: [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm squid Some package requested cannot be installed: squid-2.5.STABLE3-2mdk.i586 (due to unsatisfied perl(Authen::Smb)) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm mon Some package requested cannot be installed: mon-0.99.2-3mdk.i586 (due to unsatisfied perl(Filesys::DiskSpace)) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm SnortSnarf Some package requested cannot be installed: SnortSnarf-020516.1-2mdk.noarch (due to unsatisfied perl(BasicSorters)) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm naat-backend Some package requested cannot be installed: naat-backend-0.8-30mdk.i586 (due to unsatisfied perl(rpmtools)) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm gnome-db Some package requested cannot be installed: gnome-db-0.2.96-8mdk.i586 (due to unsatisfied perl-base[== 2:5.8.0]) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm latex2html Some package requested cannot be installed: latex2html-2002-4mdk.noarch (due to unsatisfied perl(Win32)) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm MySQL-Max The following packages have to be removed for others to be upgraded: MySQL-4.0.14-1mdk.i586 (due to conflicts with MySQL-Max[ 4.0.11]) (y/N) [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm perl-Magick Some package requested cannot be installed: perl-Magick-5.5.7.10-1mdk.i586 (due to unsatisfied perl-base[= 5.801]) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm libkdeutils1 Some package requested cannot be installed: libkdeutils1-3.1.3-6mdk.i586 (due to unsatisfied kdeutils[*][== 3.1.3-6mdk]) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm libkdegraphics0 Some package requested cannot be installed: libkdegraphics0-3.1.3-8mdk.i586 (due to unsatisfied kdegraphics[*][== 1:3.1.3-8mdk]) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]#
Re: [Cooker] 9.2 MEDIAS - MEDIA
Adam Williamson wrote: If you were really trying to Anglify it you'd say mediums But that's already taken. Medium has several meanings in English. One is that of an underlying transport for information, in which case the plural is media. That's the one we're talking about here. Medium can also mean a person with psychic abilities who serves as an underlying transport between dead and living people. In that case, the plural is mediums. The difference here presumably comes from the latin medium/media being a neuter-gender noun (i.e. dealing with inanimate objects), and this usage actually dealing with people. In the case of the last paragraph, the medium is an object. In the case of this paragraph, it's a person, and since the latin term didn't refer to a person, the later English rules for pluralization were used for the new usage. And then again, it can be an adjective, as in I'll have a medium (-sized) coffee. Now, if some kind Northern Irish person would care to discourse on the pronunciation of How now, brown cow, we can put this thread to bed...
Re: [Cooker] hd.img without floppy ?
Keld Jørn Simonsen wrote: Could all of that be put into a tarball with every update, or maybe in a rpm? It is easier and faster to just do everything on the harddisk, in stead of writing floppies or even burning cd's. And in the case of floppies, a had disk install can be from a bigger initrd disk than what can be on a floppy. Pixel had an even better solution, which is to use isolinux/alt0 all.rdz and vmlinuz. Just take the LILO example from the Wiki page for using hd.img, substitute all.rdz for hd.rdz and vmlinuz for vmlinuz-hdimg (using their actual locations in your cooker tree), and append a /sbin/lilo to your cooker update script (in case the update changes one of these files). Booting this gives you a panel from which you can choose HD install.
Re: [Cooker] rpmdrake vs. urpmi weirdness
Guillaume Cottenceau wrote: So, the problem appears to be in the FlashPlayer RPM, Maybe you could report that to the author of the RPM then? Thanks! Happy to, but the maximum info display in rpmdrake Changelog just lists * Thu Dec 12 2002 Mandrake Linux Team http://www.mandrakeexpert.com 6.0-3mdk as the maintainer. Can you tell me to whom to send this request ?
Re: [Cooker] Old cooker packages not being deleted ?
Rolf Pedersen wrote: I thought rsync was stating this was due to RPMS3 and RPMS4 on the mirrors: http://marc.theaimsgroup.com/?l=mandrake-cookerm=106027621928253w=2 also reported here: http://marc.theaimsgroup.com/?l=mandrake-cookerm=106033173610904w=2 and here: http://marc.theaimsgroup.com/?l=mandrake-cookerm=106018467231079w=2 I saw the comments, but I guess it didn't register. I'm not seeing any messages from rsync though about I/O errors or skipping deleting. My only difference from your case is that I'm using -avl (not -avL) and ftp.uninett.no.
Re: [Cooker] rpmdrake vs. urpmi weirdness
Frank Griffin wrote: I'll try it on a fresh cooker install. The one I did it on was up to date for today, but the install was from a few days ago and I had been keeping it current with urpmi. Well. I will as soon as the install gets fixed so that X (and therefore rpmdrake) can be run
Re: [Cooker] Packaging errors as of today
One more... [EMAIL PROTECTED] ftg]# urpmi --no-verify-rpm super-freeswan Some package requested cannot be installed: super-freeswan-1.99_kb2c-1mdk.i586 (due to unsatisfied klips-alg[= 1.99_kb2c]) (Y/n) Everything already installed [EMAIL PROTECTED] ftg]# urpmi klips-alg no package named klips-alg [EMAIL PROTECTED] ftg]#
Re: [Cooker] Old cooker packages not being deleted ?
Rolf Pedersen wrote: Well, I added --exclude= options for RPMS3 and RPMS4 and my command (which had worked for two years) seems to be working again. Maybe whoever added these links could just fix them ? I mean, it's only been about four days since I first reported this (and others verified it).
Re: [Cooker] rpmdrake vs. urpmi weirdness
Guillaume Cottenceau wrote: Is it reproducable with another package? Here I can see no problems.. I'll try it on a fresh cooker install. The one I did it on was up to date for today, but the install was from a few days ago and I had been keeping it current with urpmi.
Re: [Cooker] hdlist.cz asnd synthesis
Charles A Edwards wrote: Neither the hdlist or the synthesis has been updated on any of the mirrors since 2003.08.09. 6:00 PM for Main somewhat latter for the other mirrors. I was going to reply that I saw one today, but upon investigating it seems that a synthesis.hdlist.cz appeared today in RPMS rather than base ??
Re: [Cooker] hd.img without floppy ?
Thanks everyone, the loopback solution on the Wiki page seems like just the thing I was looking for.
[Cooker] How to dump current rpm db for install ?
The current install gives you a way to save your package selections to a floppy which can then be used in later installs to avoid having to manually re-select everything. Is there any way to generate such a floppy from a running system ? In other words, once I get a particular cooker stocked with various packages added over time (many post-install), can I create a floppy that can be fed to another install which will select all of the packages in the first system, whether they were selected during installation or after installation ?
[Cooker] MDKKDM fails dependencies/ GDM mis-installed
In this morning's cooker, MDKKDM fails install with the following error: mdkkdm-9.1-28mdk.i586 due to missing kdebase-kdm-config-file-3.1.3-4mdk.i586 An expert install installs GDM without indicating that MDKKDM had an error, and on reboot the resulting GDM initializes with an error popup saying Session directory is missing. The only indication of the MDKKDM error is if you open MCC and try to switch the DM to MDKKDM, at which point you're told that it isn't installed. If you run rpmdrake to install it, you'll get the dependency error.
[Cooker] User Management suggestion/question
I try to keep all user data on non-release-sensitive partitions (e.g. /data/user/xxx) and then have scripts to create links within /home/user to the real files. The idea is that I can create new systems or cooker systems at will, run the scripts, mount the partitions (or copies thereof), and test. Something I find frustrating from release to release is the creation of userids. If I do a true install, there doesn't seem to be any way to point MDK to another system partition and say go copy the usernames, UIDs, groupnames, and GIDs from here in order to seed the new system's directory. Moreover, even if I re-enter the IDs in the same order, the UIDs change slightly from release to release. This happens because several of the user management GUI tools automatically assign the UID or GID and do not allow overrides. Thus, if any 500 automatically cretaed UIDs change from release to release, you can't easily force IDs you define to have the same UIDs they had in the previous system, e.g. the addition of admin in 9.1 (and its removal now in Cooker). Is there a sensible way to import user/group data from an existing install ? If not, should there be ? If there can't be, can we at least establish a UID and GID range that belongs to the sysadmin and doesn't get messed with ?
[Cooker] Something today broke install printer/X
Something moved to cooker today has seriously broken the install process. I've been updating continuously throughout the day, and in the morning (US/EST - GMT-5:00) installs were fine. After about 3:00PM, printer configuration during install wouldn't recognize any printers, and the X configuration during install failed because it couldn't find the X modules. All of this was noticed through hd.img installs done throughout the day with continually updated hd.img boot diskettes. The failed installs were both expert and non-expert (all of the installs I've done over the past few days were expert, so I tried both).
[Cooker] Today breakage w/splash and shutdown
Sometime during today a cooker update caused the following behavior change: 1) vga=788 no longer has a splashscreen and boots in character mode. It had a valid splash this morning. 2) selecting reboot in a GNOME logout actually reboots rather than going back to the DM (yay !!) 3) for some reason, (2) switches you to vc 1 (as it should), but all of the shutdown messages are coming out in text mode on vc 7 (where DM was) This is using MDKKDM and GNOME.
Re: [Cooker] New Bootsplash
Warly wrote: Yes, I think that I will likely add one static image for the very first part of the boot with some message in the image such as Booting, please wait and switch to the translated version whenever the / is available. I'd also vote for retaining the progress bar even after F2 is pressed. Also, the verbose display reacts to keystrokes by producing garbage (e.g. ]]A]]B) in the display even though (one hopes) the keystrokes will never take effect.
Re: [Cooker] urpmi --auto-select - how to get around it?
Kim Schulz wrote: [snip] Use --keep not an option for my version of urpmi. and when I try to upgrade urpmi, then I get alot of other deps (especially alot of kde and mozilla stuff). I had similar problems on another thread, and the suggestion was to use --allow-nodeps --allow-force I had to respond to some scary prompts, but urpmi urpmi eventually installed the latest urpmi. BTW, I'm happy to see that as of yesterday an hd.img install with all categories but games checked completes correctly with only 3 (I think) package install errors. So I can revert to testing installed systems and have less dependence on incrementally upgraded systems.
[Cooker] Bugzilla New Bug error today
This morning, I bring up Bugzilla's main page, log in, select Enter New Bug, and get an Internal Server Error. Got the same bug trying to get to New Bug from other pages as well. Error occurred at about 09:17 AM EST (GMT - 5:00).
[Cooker] New Bootsplash
I like the look of the new bootsplash, but there's a slight glitch. If you use a SCSI disk driver which takes a while to initialize (both aic7xxx and aic79xx fall into this category, taking 15-30 seconds), you're left looking at a blank bootsplash with what looks like an empty textfield at the bottom right of the screen. At the least, there ought to be a loading driver xx, please wait message somewhere so that the user doesn't think the boot is hung. With the old one, at least you had the Mandrake logo and the textarea. The new one just looks incompletely drawn, contributing to the appearance of a hang. Of course, the proposed message would have been an improvement to the old one too.
Re: [Cooker] urpmi --auto-select - how to get around it?
Kim Schulz wrote: just tried this and it wanted to install around 300 rpm's but ended up with this: Installation failed: arts = 3001:1.1.3-1mdk is needed by libarts-devel-1.1.3-1mdk arts = 3001:1.1.3-1mdk is needed by libarts-1.1.3-1mdk libxml2.so is needed by (installed) libcroco0-devel-0.1.0-2mdk kdelibs = 3000:3.1.2-20mdk is needed by (installed) kdelibs-devel-3.1.2-20mdk libstdc++.so is needed by (installed) octave-2.1.50-1mdk kdegraphics = 1:3.1.2-6mdk is needed by (installed) kdegraphics-devel-3.1.2-6mdk libgcc_s.so is needed by (installed) octave-2.1.50-1mdk libreadline.so.4.2 is needed by (installed) ghc-5.00.2-2mdk libm.so is needed by (installed) libcroco0-devel-0.1.0-2mdk libm.so is needed by (installed) libgtop1-devel-1.0.13-5mdk libm.so is needed by (installed) octave-2.1.50-1mdk libglib-1.2.so is needed by (installed) libgtop1-devel-1.0.13-5mdk libz.so is needed by (installed) libcroco0-devel-0.1.0-2mdk perl-base = 5.8.0 is needed by (installed) gnome-db-0.2.96-6mdk perl-base = 2:5.8.0 is needed by (installed) gaim-0.64-3mdk perl-base = 2:5.8.0 is needed by (installed) gaim-0.64-3mdk Try installation without checking dependencies? (y/N) what now I'm not sure what you're upgrading from. I did it from base 9.1, and I'd say that it involved at least 300 packages. And, like I said, you get some scary prompts. I don't recall the components you list being in them, but the feeling I got from the list was that the packages were OK, it was just the dependencies that were incorrect. So I just answered YES to everything.
[Cooker] MakeCD broken again
As of last night's cooker, cd /data/cooker/i586/misc ./MakeCD -t /data -a /data/cooker/i586 ~/tmp/MakeCDOut 2~/tmp/MakeCDErr gives ..//Mandrake/mdkinst/usr/bin/perl: error while loading shared libraries: libperl.so: cannot open shared object file: No such file or directory It also happens after refreshing this morning...
Re: [Cooker] dm - prefer kdm or gdm
Gary L. Greene wrote: The problem is Frédéric, is that when use the shut down option in GNOME when I have GDM as the login manager, it won't shut the system down, whereas when I used GNOME with KDM, it would. What it does now is it only drops me back at GDM, then I have to select the shutdown from the system options. According to fcrozat, this is supposed to be fixed. See bug 3584.
Re: [Cooker] dm - prefer kdm or gdm
Frederic Crozat wrote: Yes, I saw that yesterday on my test system (it is not related to gdm).. I'll investigate monday. You can fill a bug against gnome-session so I won't forget about it.. Actually, I've seen this happen with mdkkdm as well lately. I was waiting to get a stable cooker (with an installable KDE :-) ) to report it, though. The symptom is the same: you choose either shutdown or reboot from GNOME, and mdkkdm comes back up with the login prompt. I pointed out bug 3584 because the visual symptom is the same if you don't actually CTRL-ALT-F1 to the console screen and notice that shutdown is in progress, and a shutdown usually takes longer than it take people (me, anyway) to decide that my shutdown request didn't work and redo it from mdkkdm or gdm.
Re: [Cooker] troubles with cooker mirror
parag shah wrote: Since early morning 30th july i am unable to successully update cooker urpmi source i.e, sunsite.uio.no. I'm using that site with no problem, but I'm using the alias recommended by the Welcome message, i.e. ftp.uninett.no.
Re: [Cooker] problems with main mirror - ftp.uninett.no
Robert Fox wrote: Just recently when I Rsync with this mirror - it's deleting files locally but not downloading new ones - is there something wrong with the sync? That happens to me occasionally, but I've always assumed that I just caught it in the middle of an rsync update from Mandrake, since the deletes appear to happen first.
Re: [Cooker] Building Cooker
Per Lindström wrote: Try to run urpmi.update to update the package lists for specified medias. urpmi.update -a updates package lists for all medias except those on removable devices. I tried a urpmi.update -a. Either that or intervening updates to cooker did make a change in what I saw (both urpmi urpmi and urpmi --auto-select now get to the install phase), but the upgrades still fail. Here are the errors for urpmi urpmi: Installation failed: librpm-4.0.4.so is needed by gnorpm-0.96-13mdk librpmdb-4.0.4.so is needed by gnorpm-0.96-13mdk librpmio-4.0.4.so is needed by gnorpm-0.96-13mdk ...and here are the errors for urpmi --auto-select Installation failed: perl(Bit::Vector) is needed by perl-Date-Calc-5.3-2mdk perl(Authen::Smb::Smb) is needed by squid-2.5.STABLE3-1mdk librpm-4.0.4.so is needed by gnorpm-0.96-13mdk librpmdb-4.0.4.so is needed by gnorpm-0.96-13mdk librpmio-4.0.4.so is needed by gnorpm-0.96-13mdk xemacs == 21.4.12 is needed by xemacs-tramp-20020411-7mdk libMrm.so.2 is needed by lesstif-clients-0.93.41-3mdk devel(libnoatunarts) is needed by kdemultimedia-devel-3.1.2-10mdk perl-BSD-Resource is needed by mod_perl-common-1.3.27_1.27-9mdk perl(BSD::Resource) is needed by mod_perl-common-1.3.27_1.27-9mdk libcap.so.1 is needed by zsh-4.1.1-2mdk libAiksaurus.so.0 is needed by lyx-1.3.2-2mdk perl(LWP::Parallel::UserAgent) is needed by squidGuard-1.2.0-6mdk
Re: [Cooker] Building Cooker
Buchan Milne wrote: (snip) Thanks very much for taking the time to write this. You have drastically improved my urpmi expertise (not particularly difficult, considering my level, but thanks anyway !).
[Cooker] Building Cooker
Yeah, sorry, it's me again. I've waited awhile through a large flood of cooker updates, and I am still unable to install a cooker system. The ISOs produced by mkcd still loop in available packages, but I gather from the TODO page that Warly knows about that. If I install 9.1, define my local cooker directory images as a urpmi source, and try urpmi --auto-select, I get errors I will include below (they're long and I wouldn't want you to lose interest). Someone had suggested previously that urpmi itself needed to be updated, so I tried urpmi urpmi. That also got errors I'll include below. I had no problems at all through the 9.0 and 9.1 alpha/beta cycles, until mkcd broke sometime after June 3. I've already tried completely re-downloading my cooker directories, and it hasn't helped. Could those of you who are managing to build cooker please share your secrets ? I'd like to be able to test again Output from urpmi urpmi: * The following packages have to be removed for others to be upgraded: mkcd-3.4.7-1mdk (due to missing perl-RPM) perl-RPM-0.40-4mdk (due to missing librpmio-4.0.4.so, due to missing librpmdb-4.0.4.so, due to missing librpm-4.0.4.so) do you agree ? (Y/n) To satisfy dependencies, the following packages are going to be installed (108 MB): drakconf-9.2-0.5mdk.i586 drakxtools-9.2-0.12mdk.i586 drakxtools-http-9.2-0.12mdk.i586 drakxtools-newt-9.2-0.12mdk.i586 elfutils-0.80-1mdk.i586 ethereal-0.9.13-1mdk.i586 glibc-2.3.2-5mdk.i586 glibc-devel-2.3.2-5mdk.i586 glibc-static-devel-2.3.2-5mdk.i586 gnorpm-0.96-13mdk.i586 gurpmi-4.4-8mdk.noarch harddrake-9.2-0.12mdk.i586 harddrake-ui-9.2-0.12mdk.i586 libbzip2_1-devel-1.0.2-16mdk.i586 libelfutils1-0.80-1mdk.i586 libelfutils1-devel-0.80-1mdk.i586 libnet-snmp50-5.0.8-5mdk.i586 libuser1-0.51.7-2mdk.i586 locales-2.3.2-3mdk.i586 locales-en-2.3.2-3mdk.i586 locales-pt-2.3.2-3mdk.i586 net-snmp-mibs-5.0.8-5mdk.i586 net-snmp-utils-5.0.8-5mdk.i586 perl-5.8.0-27mdk.i586 perl-Authen-PAM-0.14-2mdk.i586 perl-GTK-0.7008-30mdk.i586 perl-GTK2-0.0.cvs.2003.04.07.1-2mdk.i586 perl-HTML-Parser-3.28-3mdk.i586 perl-Locale-gettext-1.01-8mdk.i586 perl-MDK-Common-1.1.4-2mdk.i586 perl-Net_SSLeay-1.22-2mdk.i586 perl-URI-1.23-3mdk.noarch perl-URPM-0.91-11mdk.i586 perl-XML-Writer-0.4-7mdk.noarch perl-base-5.8.0-27mdk.i586 perl-devel-5.8.0-27mdk.i586 perl-doc-5.8.0-27mdk.i586 perl-libwww-perl-5.69-2mdk.noarch popt-1.8-12mdk.i586 popt-devel-1.8-12mdk.i586 rpm-4.2-12mdk.i586 rpm-build-4.2-12mdk.i586 rpm-devel-4.2-12mdk.i586 rpm-python-4.2-12mdk.i586 rpmdrake-2.1-24mdk.i586 rpmstats-0.4-1mdk.i586 rpmtools-4.5-11mdk.i586 tcp_wrappers-7.6-22mdk.i586 urpmi-4.4-8mdk.noarch userdrake-0.92-9mdk.i586 Is this OK? (Y/n) Installation failed, some files are missing: /data/cooker/i586/Mandrake/RPMS/perl-devel-5.8.0-27mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/drakxtools-http-9.2-0.12mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/libelfutils1-0.80-1mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/elfutils-0.80-1mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/libuser1-0.51.7-2mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/libnet-snmp50-5.0.8-5mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/perl-URPM-0.91-11mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/perl-base-5.8.0-27mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/perl-doc-5.8.0-27mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/libelfutils1-devel-0.80-1mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/userdrake-0.92-9mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/drakxtools-9.2-0.12mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/perl-5.8.0-27mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/perl-GTK2-0.0.cvs.2003.04.07.1-2mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/drakxtools-newt-9.2-0.12mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/gurpmi-4.4-8mdk.noarch.rpm /data/cooker/i586/Mandrake/RPMS/drakconf-9.2-0.5mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/urpmi-4.4-8mdk.noarch.rpm /data/cooker/i586/Mandrake/RPMS/rpmdrake-2.1-24mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/net-snmp-mibs-5.0.8-5mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/harddrake-ui-9.2-0.12mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/harddrake-9.2-0.12mdk.i586.rpm /data/cooker/i586/Mandrake/RPMS/net-snmp-utils-5.0.8-5mdk.i586.rpm You may want to update your urpmi database Output from urpmi --auto-select: *** The following packages have to be removed for others to be upgraded: MySQL-Max-4.0.11a-5mdk (due to conflicts with MySQL[ 4.0.11]) fax4CUPS-hylafax-1.22-2mdk (due to missing hylafax) hylafax-4.1.5-1mdk (due to conflicts with mgetty-sendfax) mad-0.14.2b-4mdk (due to missing libid3tag.so.0, due to unsatisfied libmad0 == 0.14.2b) mod_ssl-2.8.12-8mdk (due to unsatisfied mm == 1.2.2) mod_sxnet-1.2.4-22mdk (due to unsatisfied mod_ssl *) perl-RPM-0.40-4mdk (due to
[Cooker] gendistrib without SRPMS ?
In trying to upgrade 9.1 to Cooker from a local file mirror, I was running gendistrib and getting the message unable to find a valid root directory which contains SRPMS. I didn't realize that this was actually causing gendistrib to abort before updating the lists. My local cooker mirror only contains the i586 directory, since I didn't think I needed source RPMs for anything. Is there a way to run gendistrib without SRPMS ? Is it correct that gendistrib run against --distrib /xxx/cooker/i586 should be requiring SRPMS ?
[Cooker] MakeCD has taken a step backwards
We're back to this again: URPM object version 0.91 does not match bootstrap parameter 0.90 at ..//Mandrake/mdkinst/usr/lib/perl5/5.8.0//i386-linux-thread-multi/DynaLoader.pm line 249. Compilation failed in require at ..//misc/perl5/vendor_perl/5.8.0/Mkcd/Tools.pm line 12. BEGIN failed--compilation aborted at ..//misc/perl5/vendor_perl/5.8.0/Mkcd/Tools.pm line 12. Compilation failed in require at ..//misc/mkcd line 11. BEGIN failed--compilation aborted at ..//misc/mkcd line 11.
Re: [Cooker] urpmi woes
François Pons wrote: If you mirror directly packages on your machine, you may want to use --virtual of newer urpmi in order to avoid using urpmi.update, because the medium will *always* be up-to-date, but it works only for local medium (ie file:// protocol). The 9.1 urpmi doesn't support --virtual, and it doesn't even show up in the May 18 cooker (which was the last one I was able to build). As of today's cooker updates, things have changed. Both urpmi urpmi and urpmi --auto-select still fail, but now they get to the stage of actually printing out all of the things that are going to be installed, and then find things that are missing. So, I'm assuming that this wasn't something wrong with my setup, and that after the current torrent of cooker updates dies down, I'll be able to try again with better luck.
Re: [Cooker] split lists?
Guillaume Cottenceau wrote: There are two categories of bugs: 1- bugs on install, or hardware support. We are burried under such bugs, with the following problems: - many duplicates - many poorly qualified bugs (and reporting accurate hardware information is non trivial) - the install changes much during beta stabilization - we have little resource to investigate hardware problems hence, many of them stay open forever. Maybe I don't understand the categories correctly, but I have submitted many install bugs with reproducible cases which seem to stay UNCONFIRMED forever. I'm assuming that UNCONFIRMED means that nobody has had time to try the reproducible case, and that if they had, the bug would have changed status to either ASSIGNED, WORKSFORME, or NEEDSINFO. So not all install bugs which stay open forever do so because they are poor bug reports. The indication is that many of them simply don't get looked at because of resource issues.
[Cooker] urpmi woes
Following advice from the ML to install 9.1 and use urpmi --auto-select to upgrade packages to cooker level using a urpmi source of the local cooker directory image, I have been getting the same results for about a week, except the the list of packages in the final Installation failed message just seems to keep increasing. I had also seen a post which intimated that you needed to urpmi urpmi before trying to upgrade everything else. That failed too. Here's the output from urpmi urpmi: *** The following packages have to be removed for others to be upgraded: mkcd-3.4.7-1mdk (due to missing perl-RPM) perl-RPM-0.40-4mdk (due to missing librpmio-4.0.4.so, due to missing librpmdb-4.0.4.so, due to missing librpm-4.0.4.so) do you agree ? (Y/n) To satisfy dependencies, the following packages are going to be installed (101 MB): drakconf-9.2-0.3mdk.i586 drakxtools-9.2-0.8mdk.i586 drakxtools-http-9.2-0.8mdk.i586 drakxtools-newt-9.2-0.8mdk.i586 elfutils-0.79-1mdk.i586 ethereal-0.9.13-1mdk.i586 gnorpm-0.96-13mdk.i586 gurpmi-4.4-8mdk.noarch harddrake-9.2-0.8mdk.i586 harddrake-ui-9.2-0.8mdk.i586 libelfutils1-0.79-1mdk.i586 libnet-snmp50-5.0.8-5mdk.i586 net-snmp-mibs-5.0.8-5mdk.i586 net-snmp-utils-5.0.8-5mdk.i586 perl-5.8.0-25mdk.i586 perl-Authen-PAM-0.14-2mdk.i586 perl-GTK-0.7008-30mdk.i586 perl-GTK2-0.0.cvs.2003.04.07.1-2mdk.i586 perl-HTML-Parser-3.28-3mdk.i586 perl-Locale-gettext-1.01-8mdk.i586 perl-MDK-Common-1.1.4-2mdk.i586 perl-Net_SSLeay-1.22-2mdk.i586 perl-URI-1.23-3mdk.noarch perl-URPM-0.91-7mdk.i586 perl-XML-Writer-0.4-7mdk.noarch perl-base-5.8.0-25mdk.i586 perl-devel-5.8.0-25mdk.i586 perl-doc-5.8.0-25mdk.i586 perl-libwww-perl-5.69-2mdk.noarch popt-1.8-8mdk.i586 popt-devel-1.8-8mdk.i586 rpm-4.2-8mdk.i586 rpm-build-4.2-8mdk.i586 rpm-devel-4.2-8mdk.i586 rpm-python-4.2-8mdk.i586 rpmdrake-2.1-24mdk.i586 rpmstats-0.4-1mdk.i586 rpmtools-4.5-11mdk.i586 tcp_wrappers-7.6-22mdk.i586 urpmi-4.4-8mdk.noarch Is this OK? (Y/n) Installation failed, some files are missing: /data/cooker/Mandrake/RPMS/perl-doc-5.8.0-25mdk.i586.rpm /data/cooker/Mandrake/RPMS/perl-devel-5.8.0-25mdk.i586.rpm /data/cooker/Mandrake/RPMS/harddrake-ui-9.2-0.8mdk.i586.rpm /data/cooker/Mandrake/RPMS/rpm-devel-4.2-8mdk.i586.rpm /data/cooker/Mandrake/RPMS/perl-URPM-0.91-7mdk.i586.rpm /data/cooker/Mandrake/RPMS/drakxtools-newt-9.2-0.8mdk.i586.rpm /data/cooker/Mandrake/RPMS/drakxtools-http-9.2-0.8mdk.i586.rpm /data/cooker/Mandrake/RPMS/rpm-python-4.2-8mdk.i586.rpm /data/cooker/Mandrake/RPMS/perl-5.8.0-25mdk.i586.rpm /data/cooker/Mandrake/RPMS/elfutils-0.79-1mdk.i586.rpm /data/cooker/Mandrake/RPMS/libelfutils1-0.79-1mdk.i586.rpm /data/cooker/Mandrake/RPMS/popt-devel-1.8-8mdk.i586.rpm /data/cooker/Mandrake/RPMS/drakxtools-9.2-0.8mdk.i586.rpm /data/cooker/Mandrake/RPMS/rpm-build-4.2-8mdk.i586.rpm /data/cooker/Mandrake/RPMS/perl-base-5.8.0-25mdk.i586.rpm /data/cooker/Mandrake/RPMS/harddrake-9.2-0.8mdk.i586.rpm /data/cooker/Mandrake/RPMS/rpm-4.2-8mdk.i586.rpm /data/cooker/Mandrake/RPMS/popt-1.8-8mdk.i586.rpm /data/cooker/Mandrake/RPMS/drakconf-9.2-0.3mdk.i586.rpm You may want to update your urpmi database * Here's the output from urpmi --auto-select: ** The following packages have to be removed for others to be upgraded: MySQL-Max-4.0.11a-5mdk (due to conflicts with MySQL[ 4.0.11]) fax4CUPS-hylafax-1.22-2mdk (due to missing hylafax) hylafax-4.1.5-1mdk (due to conflicts with mgetty-sendfax) mad-0.14.2b-4mdk (due to missing libid3tag.so.0, due to unsatisfied libmad0 == 0.14.2b) mod_ssl-2.8.12-8mdk (due to unsatisfied mm == 1.2.2) mod_sxnet-1.2.4-22mdk (due to unsatisfied mm == 1.2.2, due to unsatisfied mod_ssl *) perl-RPM-0.40-4mdk (due to missing librpmdb-4.0.4.so, due to missing librpm-4.0.4.so, due to missing librpmio-4.0.4.so) smalltalk-1.95.12-2mdk (due to missing libtk8.3.so) xemacs-extras-21.4.12-1mdk (due to conflicts with xemacs-extras) do you agree ? (Y/n) To satisfy dependencies, the following packages are going to be installed (2612 MB): ADVX-build-9.2-1mdk.noarch BitchX-1.0-0.c19.7mdk.i586 Borges-0.9.0-2mdk.noarch Borges-DocBook-0.9.0-2mdk.noarch Borges-doc-0.9.0-2mdk.noarch Eterm-0.9.2-3mdk.i586 FaxMail-2.3-9mdk.i586 GConf-1.0.9-7mdk.i586 GConf2-2.3.3-1mdk.i586 Guppi-0.40.3-6mdk.i586 ImageMagick-5.5.4.4-8mdk.i586 Mesa-5.0.1-4mdk.i586 Mesa-demos-5.0.1-4mdk.i586 MySQL-4.0.13-2mdk.i586 MySQL-client-4.0.13-2mdk.i586 MySQL-common-4.0.13-2mdk.i586 ORBit-0.5.17-4mdk.i586 ORBit2-2.7.2-2mdk.i586 OpenOffice.org-1.0.3-2mdk.i586 OpenOffice.org-help-en-1.0.3-2mdk.i586 OpenOffice.org-l10n-en-1.0.3-2mdk.i586 OpenOffice.org-libs-1.0.3-2mdk.i586 OpenSP-1.5-4mdk.i586 SysVinit-2.85-1mdk.i586 X11R6-contrib-4.3-10mdk.i586 XFree86-4.3-10mdk.i586
Re: [Cooker] split lists?
Buchan Milne wrote: Aren't people supposed to be searching Bugzilla first ? The bugzilla mails to the list are not mainly for people who would otherwise post bugs, it is mainly for people who can help fix bugs. It allows distributed bug resolution. Before the bugzilla list, bugzilla traffic was much lower. Sorry if I'm being dense (I'm pretty new to the undocumented Cooker social protocols and I'm not about to try searching the ML for bugzilla), but if someone is interested in working on bugs in one or several product areas, wouldn't it just be simpler to run a Bugzilla query every day/week/month/whatever for all bugs in those areas since the last time you queried ? That would avoid wading through all of the bug reports for those areas that don't interest you. Of course, every time I try to run complex bugzilla searches, I get 500 Internal Server Errors If they run into something, searching Bugzilla seems a lot more straightforward than hoping that you remember some mail that flew by a month or so ago. For me, it is much quicker to search through my 3-6-month archive of the bugzilla mails, than to even get to the bugzilla search page. Again, I ask from ignorance, but would it be possible to register with Bugzilla for a product category (rather than a specific bug) so that you would get mails for any activity in that product category ? Also, one of the problems with Bugzilla is that certain categories (like Installation) bypass sending mail to anybody at Mandrake *except* the Cooker ML. File a bug on Bugzilla (not bugzilla ... IIRC). I'll be happy to, but since it was only the Installation category where I noticed this (you get told that mail will *not* be sent to [EMAIL PROTECTED]), I sort of figured it was intentional :-) Thanks, Frank
Re: [Cooker] split lists?
Ben Reser wrote: Back when Bugzilla was put in place I suggested that bugs only be sent to the list for the first bug report (so people see the reports) and then the rest would happen off the list. If you cared about a bug you subscribed to it. If you didn't then you wouldn't see it after the first entry. We really aren't taking full advantage of the power of Bugzilla... Why shove every bug report and comment down everyone's email pipe? It's entirely wasteful of our time to delete stuff we just don't care about. I agree, but I'd question even sending the first report. Aren't people supposed to be searching Bugzilla first ? If they run into something, searching Bugzilla seems a lot more straightforward than hoping that you remember some mail that flew by a month or so ago. Also, one of the problems with Bugzilla is that certain categories (like Installation) bypass sending mail to anybody at Mandrake *except* the Cooker ML.
Re: [Cooker] split lists?
Buchan Milne wrote: I agree, but I'd question even sending the first report. Well, if it were a seperate list, you wouldn't have to question it. I should have pointed out that if the Bugzilla stuff went to a separate list, then my query about other ways to do this through Bugzilla would be moot...
Re: [Cooker] split lists?
Levi Ramsey wrote: On Tue Jul 08 17:26 -0400, Frank Griffin wrote: I agree, but I'd question even sending the first report. Aren't people supposed to be searching Bugzilla first ? If they run into something, searching Bugzilla seems a lot more straightforward than hoping that you remember some mail that flew by a month or so ago. I think the logic wasn't to prevent duplicate bug reports but to help get more testing of specific bugs (and votes for bugs). Umm, Ok, here's another obvious newbie question. Is there a link that describes the voting process for Cooker (Bugzilla) bugs. I've just been submitting bug reports in the (probably naive) belief that they would be routed to the category owners and acted upon according to whatever priority scheme is dictated by Mandrake's internal business practices. However, comments on those bugs have hinted at some sort of voting practice. As far as I knew, the only voting was on Club packages. How do you vote for bugs (and what is the protocol of doing so) ?
Re: [Cooker] split lists?
Ben Reser wrote: Also, one of the problems with Bugzilla is that certain categories (like Installation) bypass sending mail to anybody at Mandrake *except* the Cooker ML. If you mean that bugs have to be voted on before they get seen by the developers then that is good and useful. Otherwise, I'm not sure what you're talking about here. Somehow I missed the description of bugs having to be voted upon to be seen by developers. I've just been submitting them. Can you provide a link that describes this practice ?
Re: [Cooker] split lists?
Ben Reser wrote: Of course, every time I try to run complex bugzilla searches, I get 500 Internal Server Errors If that's the case then those problems should be filed as bug reports against Bugzilla itself so Warly can fix them. If we were making full use of Bugzilla there would be a way to either watch categories of bugs or at least specific packages bugs. Maintainers already basically get this. I don't see why other people who are interested in a package shouldn't get this benefit too. It helps pull the signal from the noise. Should increase contributions and would hopefully cause more bugs to be fixed without the maintainer/employee of having to do the foot work of figuring the issue out. But I've made these suggestions in the past and nobody seemed all that interested in them. I didn't report them because I assumed they were Bugzilla bugs (and thus outside of Mandrake's province). If Warly maintains Bugzilla, then I will.
Re: [Cooker] split lists?
Rather than split by product, I would suggest splitting out the Bugzilla mails. By definition, anybody interested in changes to bugs ought to be watching the bug (and therefore be mailed) anyway.
Re: [Cooker] Can not install Cooker
[EMAIL PROTECTED] wrote: Then it checks the partition and displays the message Looking for available packages... and it is completely frozen. CTRL-ALT-F1 shows the last message was Entering step Choose Packages to Install. The cooker has been broken in this way for several weeks. Am I just not building it properly? This used to work just fine. Check out bug 4057.
Re: [Cooker] Unable to install Cooker (fresh from mirrors)
Mark Draheim wrote: When I try, the install freezes at looking for available packages - no apparent errors in the other consoles and the CPU gets pegged to 100%. same here Check out bug 4057. Warly says this is an fpons bug which has not yet been fixed.
Re: [Cooker] PLEASE address build/install problems
Warly wrote: Yes and I replied that I tried to fix it. Have you tried again? The bug you replied about was a different one, where MakeCD would immediately error out without creating images. That one was fixed, and I closed the Bugzilla report for it. 4057 is different, in that the ISO images are built without error, but either stuff is missing or the dependencies are screwed, because the install hangs or loops at looking for available packages. I have re-tried this every day or every other day since you fixed the first bug. Thanks, Frank
Re: [Cooker] PLEASE address build/install problems
Thanks, I'll make a note of the urpmi --auto-select technique, but most of the bugs I report are with the install itself, not just with this package or that. Without being able to do an install, those aren't going to get reported, and I can't tell if the old ones have been fixed (when bugzilla asks me every month). I fully understand that it is the nature of cooker than things break. But some things are more critical to the intended goals of cooker than others. If you want people to test a system install, the install has to work at least to the point of completing and producing a bootable system. Have I got this backwards ? Is it really the intention to address only application package issues in the early part of the release cycle, and leave install issues until the betas/RCs ?
[Cooker] PLEASE address build/install problems
I've been building and testing cooker systems for over a year, primarily by creating ISOs with MakeCD and burning them. I've tried harddrive and network installs, but they don't work for me (as documented in bug 3295). Recently, as documented in bug 4057, burning ISOs stopped working as well, and a previous comment in this list drew a response that indicates that at least one other person is having the same problem with MakeCD. There's been a lot of traffic on this list about telling people that they'd better report certain problems by a certain point in the development cycle if they want to see the bug addressed in the final release. That's quite reasonable, but it falls apart if the build/install process itself is broken and just stays that way. I entered bug 3295 on March 14 (hd/network installs giving package errors that ISO installs from ISOs burned from the exact same directories do NOT give), and nobody's commented on it but me. After finding bug 4057 (installs from ISOs hang at looking for available packages), I no longer have any way to build or install a Cooker system. This bug showed up sometime between the beginning of June and mid-June. There was another mkcd bug which Warly addressed around mid-June that had made it impossible to build ISO images at all from about June 3 to June 12, so I can't tell when in that timeframe bug 4057 surfaced. I would very much like to continue testing Cooker for you, and I realize that you are short-staffed. If I had never gotten a build to work, I could understand a certain amount of skepticism about bug reports saying my cooker won't work. But that's not the case here. If I can build/burn ISOs which install without problems, yet the exact same directories won't install without problems if I do an hd or network install, then I don't think the problem is in the directories. Likewise, if I have been building/installing burned ISO systems all along, and they suddenly stop installing when I haven't changed my mirroring code, AND someone else is seeing the same problem, the problem isn't in my copy of Cooker. Recent posts to this list indicate that we're coming close to a beta milepost for 9.2, and it is really annoying to not be able to test anything, since I will no doubt hear later on that I should have reported that bug back in whenever. Please take a look at these bug reports and provide some feedback.
[Cooker] Can't install Cooker ISOs since Jun 3
I've entered bug 4057 about this, but I wanted to ask here just to see if I'm the only one experiencing this. I've been building and installing Cooker ISOs for over a year without problems. The installs are always full installs, with Custom Disk Assignment of just a root ext2 partition and a full format of same. Nothing fancy. Somewhere on or after June 3, the ISOs stopped installing. The install formats the root partition, goes to looking for available packages, and hangs or loops there (I can't tell which) until you reboot. I've left it for several hours on some occasions. I've tried running gendistrib, with no effect. I've even re-downloaded a fresh Cooker tree, still with no effect. I've included report.bug as well as the stdout/stderr from MakeCD in the bug report, but MakeCD puts out so much stuff that I never know what's significant and what's not. Has anyone else been able to install from CDs burned from Cooker since June 3 ?
[Cooker] Re: Rsync Mirrors
Just what I needed, thanks.
[Cooker] Rsync Mirrors
Is there a list of rsync cooker mirrors along with the module names ? I've tried several of the regular mirrors at random, but those that respond to an rsync connect attempt seem to have inconsistent module structures. So far the ones I got to work are sunsite.uio.no and ftp.sunet.se, but as I'm in the USA I'd rather use one closer to home
[Cooker] MakeCD fails due to perl mismatch ?
Trying to build ISO images from today's cooker, I'm getting: URPM object version 0.90 does not match bootstrap parameter 0.81 at ..//Mandrake/mdkinst/usr/lib/perl5/5.8.0//i386-linux-thread-multi/DynaLoader.pm line 249. Compilation failed in require at ..//misc/perl5/vendor_perl/5.8.0/Mkcd/Tools.pm line 12. BEGIN failed--compilation aborted at ..//misc/perl5/vendor_perl/5.8.0/Mkcd/Tools.pm line 12. Compilation failed in require at ..//misc/mkcd line 11. BEGIN failed--compilation aborted at ..//misc/mkcd line 11. I tried selectively deleting and re-fetching control directories like mdkinst, base, and misc, but got the same result.
Re: [Cooker] [Bug 427] [drakxtools] Drakfont freezes the whole partition
Szakacsits Szabolcs wrote:
Hi,
Somebody who can reproduce the problem (I couldn't) would please
'strace' drakfont and send the last 50-100 lines or put all output to
some downloadable place to see where the NTFS driver hangs? You can
try by starting drakfont like
strace -f -e trace=file drakfont
or after it's started but before importing Windows font, what makes
the driver hang, attaching to drakfont like
strace -f -e trace=file -p pid_of_drakfont
The last lines using the latter technique are:
open(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/GARAIT.TTF,
O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE, 0666) = 6
fstat64(6, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
stat64(/mnt/win_d/winnt/fonts/GARAIT.TTF, {st_mode=S_IFREG|0555,
st_size=188916, ...}) = 0
chmod(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/GARAIT.TTF, 0100555) = 0
stat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp,
{st_mode=S_IFDIR|0755, st_size=8192, ...}) = 0
lstat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgia.ttf,
0x804b678) = -1 ENOENT (No such file or directory)
stat64(/mnt/win_d/winnt/fonts/georgia.ttf, {st_mode=S_IFREG|0555,
st_size=142964, ...}) = 0
lstat64(/mnt/win_d/winnt/fonts/georgia.ttf, {st_mode=S_IFREG|0555,
st_size=142964, ...}) = 0
open(/mnt/win_d/winnt/fonts/georgia.ttf, O_RDONLY|O_LARGEFILE) = 3
fstat64(3, {st_mode=S_IFREG|0555, st_size=142964, ...}) = 0
open(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgia.ttf,
O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE, 0666) = 6
fstat64(6, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
stat64(/mnt/win_d/winnt/fonts/georgia.ttf, {st_mode=S_IFREG|0555,
st_size=142964, ...}) = 0
chmod(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgia.ttf, 0100555) = 0
stat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp,
{st_mode=S_IFDIR|0755, st_size=8192, ...}) = 0
lstat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiab.ttf,
0x804b678) = -1 ENOENT (No such file or directory)
stat64(/mnt/win_d/winnt/fonts/georgiab.ttf, {st_mode=S_IFREG|0555,
st_size=139584, ...}) = 0
lstat64(/mnt/win_d/winnt/fonts/georgiab.ttf, {st_mode=S_IFREG|0555,
st_size=139584, ...}) = 0
open(/mnt/win_d/winnt/fonts/georgiab.ttf, O_RDONLY|O_LARGEFILE) = 3
fstat64(3, {st_mode=S_IFREG|0555, st_size=139584, ...}) = 0
open(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiab.ttf,
O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE, 0666) = 6
fstat64(6, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
stat64(/mnt/win_d/winnt/fonts/georgiab.ttf, {st_mode=S_IFREG|0555,
st_size=139584, ...}) = 0
chmod(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiab.ttf, 0100555) = 0
stat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp,
{st_mode=S_IFDIR|0755, st_size=8192, ...}) = 0
lstat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiai.ttf,
0x804b678) = -1 ENOENT (No such file or directory)
stat64(/mnt/win_d/winnt/fonts/georgiai.ttf, {st_mode=S_IFREG|0555,
st_size=156668, ...}) = 0
lstat64(/mnt/win_d/winnt/fonts/georgiai.ttf, {st_mode=S_IFREG|0555,
st_size=156668, ...}) = 0
open(/mnt/win_d/winnt/fonts/georgiai.ttf, O_RDONLY|O_LARGEFILE) = 3
fstat64(3, {st_mode=S_IFREG|0555, st_size=156668, ...}) = 0
open(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiai.ttf,
O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE, 0666) = 6
fstat64(6, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
stat64(/mnt/win_d/winnt/fonts/georgiai.ttf, {st_mode=S_IFREG|0555,
st_size=156668, ...}) = 0
chmod(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiai.ttf, 0100555) = 0
stat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp,
{st_mode=S_IFDIR|0755, st_size=8192, ...}) = 0
lstat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiaz.ttf,
0x804b678) = -1 ENOENT (No such file or directory)
stat64(/mnt/win_d/winnt/fonts/georgiaz.ttf, {st_mode=S_IFREG|0555,
st_size=158796, ...}) = 0
lstat64(/mnt/win_d/winnt/fonts/georgiaz.ttf, {st_mode=S_IFREG|0555,
st_size=158796, ...}) = 0
open(/mnt/win_d/winnt/fonts/georgiaz.ttf, O_RDONLY|O_LARGEFILE) = 3
fstat64(3, {st_mode=S_IFREG|0555, st_size=158796, ...}) = 0
open(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiaz.ttf,
O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE, 0666) = 6
fstat64(6, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
stat64(/mnt/win_d/winnt/fonts/georgiaz.ttf, {st_mode=S_IFREG|0555,
st_size=158796, ...}) = 0
chmod(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/georgiaz.ttf, 0100555) = 0
stat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp,
{st_mode=S_IFDIR|0755, st_size=8192, ...}) = 0
lstat64(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/GOTHIC.TTF,
0x804b678) = -1 ENOENT (No such file or directory)
stat64(/mnt/win_d/winnt/fonts/GOTHIC.TTF, {st_mode=S_IFREG|0555,
st_size=137568, ...}) = 0
lstat64(/mnt/win_d/winnt/fonts/GOTHIC.TTF, {st_mode=S_IFREG|0555,
st_size=137568, ...}) = 0
open(/mnt/win_d/winnt/fonts/GOTHIC.TTF, O_RDONLY|O_LARGEFILE) = 3
fstat64(3, {st_mode=S_IFREG|0555, st_size=137568, ...}) = 0
open(/usr/X11R6/lib/X11/fonts/drakfont/tmp/tmp/GOTHIC.TTF,
O_WRONLY|O_CREAT|O_TRUNC|O_LARGEFILE, 0666) = 6
fstat64(6, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
