Re: [gentoo-user] web ftp server
Hi, Hi guys, i have a web server with IP based virtual hosts (3-4) and i want to make a ftp account for every host. I'm thinking to use proftpd and to make the root dir in /var/www so then every user will have home directory to different web account (/var/www/domain1, /var/www/domain2, etc.). For the ownership of the files i'm thinking to start the ftp server with user and group apache to make possible for the web server to read the files. Is my ideas right? I'm open for other ideas :) well, in general you are right, but please allow me some comments: ProFTPD is supporting alternative user/passwd sources like files or a mysql db. While mysql is a good solution for mass hosting, I would go for authuserfile in your case. This allows you to work with FTP users without adding any system users to your system. As well you can also give the user an invidual UID and GID, so I would suggest the UID and GID of apache. I am not sure, if the ftpasswd program to handle this accounts is included in the gentoo package. You will find it in the contrib directory of proftpd. Usuage: (example taken from an SuSE System I worked on last week): ftpasswd --file /usr/local/etc/proftpd.user --home HOME --name NAME --shell /bin/false --passwd --uid 30 --gid 8 proftpd.conf: AuthUserFile /usr/local/etc/proftpd.user RequireValidshell off the syntax should be clear, otherwise just ask. If I am correct you have to recompile ProFTPD with the correct Useflag. Also please note, that you still can login as a system user. To prevent this you have to add the authorder option to your proftp.conf Hope that helps Stonki -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] web ftp server
Thanks a lot it was very helpfull :) [EMAIL PROTECTED] wrote: Hi, Hi guys, i have a web server with IP based virtual hosts (3-4) and i want to make a ftp account for every host. I'm thinking to use proftpd and to make the root dir in /var/www so then every user will have home directory to different web account (/var/www/domain1, /var/www/domain2, etc.). For the ownership of the files i'm thinking to start the ftp server with user and group apache to make possible for the web server to read the files. Is my ideas right? I'm open for other ideas :) well, in general you are right, but please allow me some comments: ProFTPD is supporting alternative user/passwd sources like files or a mysql db. While mysql is a good solution for mass hosting, I would go for authuserfile in your case. This allows you to work with FTP users without adding any system users to your system. As well you can also give the user an invidual UID and GID, so I would suggest the UID and GID of apache. I am not sure, if the ftpasswd program to handle this accounts is included in the gentoo package. You will find it in the contrib directory of proftpd. Usuage: (example taken from an SuSE System I worked on last week): ftpasswd --file /usr/local/etc/proftpd.user --home HOME --name NAME --shell /bin/false --passwd --uid 30 --gid 8 proftpd.conf: AuthUserFile /usr/local/etc/proftpd.user RequireValidshell off the syntax should be clear, otherwise just ask. If I am correct you have to recompile ProFTPD with the correct Useflag. Also please note, that you still can login as a system user. To prevent this you have to add the authorder option to your proftp.conf Hope that helps Stonki -- Cyberly yours, Nikolay Balov mail: [EMAIL PROTECTED] Linux Registered User #382280 http://keyserver.linux.it/ Key fingerprint = D80E A05B 5727 B40C 7431 2CC0 0845 E79E 428A 1109 -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 100% CPU usage with no processes to blame?
On 2/15/07, Alan McKinnon [EMAIL PROTECTED] wrote: On Wednesday 14 February 2007, brullo nulla wrote: Most likely you only looked at user cpu % and neglected to list the system and niced times as well. p.s. golden rule: ps lies. top lies. free lies. Don't believe the readings they give, rather interpret them in context. sob. it's not the first time I hear this. What should I believe to really know my system state? You should use ps, top and free of course! Just realize that they lie... Seeing this thread reminded me of a blog article I saw on Virtual Threads a while back... http://virtualthreads.blogspot.com/2006/02/understanding-memory-usage-on-linux.html He does a pretty good job of explaining where top gets its numbers and how to properly interpret them. HTH -mike -- Michael E. Crute http://mike.crute.org God put me on this earth to accomplish a certain number of things. Right now I am so far behind that I will never die. --Bill Watterson -- gentoo-user@gentoo.org mailing list
[gentoo-user] Short history in terminal (without X)
Hi, does someone know, how to increase the history-buffer's size of the terminal? (and I mean the pure terminal, without an X-Server) I mean, if I use some commands producing plenty of output, I cannot scroll to the beginning of the text quite often, because the history buffer is to small. Another inconvenient thing is that the buffer seems to forget everything except the last screen of text, if I switch to another terminal. (alt + F2 for instance). How can I make the history buffer larger, or - if possible - set it infinitely large. (Just as the Konsole of KDE.) Thanks, Roman Naumann. pgpbMK8jMfPxJ.pgp Description: PGP signature
Re: [gentoo-user] Short history in terminal (without X)
On 2/15/07, Roman Naumann [EMAIL PROTECTED] wrote: Hi, does someone know, how to increase the history-buffer's size of the terminal? (and I mean the pure terminal, without an X-Server) I mean, if I use some commands producing plenty of output, I cannot scroll to the beginning of the text quite often, because the history buffer is to small. Another inconvenient thing is that the buffer seems to forget everything except the last screen of text, if I switch to another terminal. (alt + F2 for instance). How can I make the history buffer larger, or - if possible - set it infinitely large. (Just as the Konsole of KDE.) Thanks, Roman Naumann. I don't know about making it infinitely large, but you can have a scrollable console history with vesafb (probably with other framebuffers aswell, but I've only had experience with vesafb). Check out /usr/src/linux/Documentation/fb/vesafb.txt for more information. You'll have to compile your kernel with vesafb, and run the vga kernel argument to set the size of the console. I believe there is some sort of kernel option (that you should be able to find with a minimum of searching) that sets the size of the buffer in bytes... From there, it is as easy as pressing shift-pgup and shift-pgdown to scroll through your buffer. Hope this helps. -- -·=»Ðŧħ«=·-
Re: [gentoo-user] Short history in terminal (without X)
Roman Naumann 写道: Hi, does someone know, how to increase the history-buffer's size of the terminal? (and I mean the pure terminal, without an X-Server) I mean, if I use some commands producing plenty of output, I cannot scroll to the beginning of the text quite often, because the history buffer is to small. Another inconvenient thing is that the buffer seems to forget everything except the last screen of text, if I switch to another terminal. (alt + F2 for instance). How can I make the history buffer larger, or - if possible - set it infinitely large. (Just as the Konsole of KDE.) Thanks, Roman Naumann. Try screen. http://www.gnu.org/software/screen But I like to use tee: command 21 | tee output -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Short history in terminal (without X)
Roman Naumann wrote: Hi, does someone know, how to increase the history-buffer's size of the terminal? (and I mean the pure terminal, without an X-Server) I mean, if I use some commands producing plenty of output, I cannot scroll to the beginning of the text quite often, because the history buffer is to small. Another inconvenient thing is that the buffer seems to forget everything except the last screen of text, if I switch to another terminal. (alt + F2 for instance). How can I make the history buffer larger, or - if possible - set it infinitely large. (Just as the Konsole of KDE.) Thanks, Roman Naumann. To make it infinitely large you can set the variable HISTSIZE to some huge value like export HISTSIZE=1 You could try setting the history to infinitely in Konsole and then do echo $HISTSIZE Cheers, Jay -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Short history in terminal (without X)
Jakob Buchgraber wrote: Roman Naumann wrote: Hi, does someone know, how to increase the history-buffer's size of the terminal? (and I mean the pure terminal, without an X-Server) I mean, if I use some commands producing plenty of output, I cannot scroll to the beginning of the text quite often, because the history buffer is to small. Another inconvenient thing is that the buffer seems to forget everything except the last screen of text, if I switch to another terminal. (alt + F2 for instance). How can I make the history buffer larger, or - if possible - set it infinitely large. (Just as the Konsole of KDE.) Thanks, Roman Naumann. To make it infinitely large you can set the variable HISTSIZE to some huge value like export HISTSIZE=1 You could try setting the history to infinitely in Konsole and then do echo $HISTSIZE Cheers, Jay HISTSIZE is used in shell as the number of commands to save in a history list, but cannot increase the history-buffer's size of the terminal. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild keeps building gcc
Old bug. You have to fix some paths in the two mentioned files. I can't remember if it's both files or just one of them. But at least in one of the files the paths are wrong. Check bugs.gentoo.org to find more. I've had 'em too ;) Kristian Poul Herkild -Original Message- From: Peter Lewis [EMAIL PROTECTED] Date: Thu, 15 Feb 2007 17:37:22 + To: gentoo-user@lists.gentoo.org Subject: [gentoo-user] revdep-rebuild keeps building gcc Hi all, I've just run revdep-rebuild, after updating world, and it rebuilt gcc-4.1.1-r3, due to the following libraries: /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/libgcjawt.la /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/libgij.la Once it finished, I ran revdep-rebuild once again, just to check that all was okay, but it wanted to rebuild gcc again, due to the same libraries. In fact, I let it do it again, and it's not fixed it. Does this point to some kind of inconsistency on my system? Thanks, Pete. -- gentoo-user@gentoo.org mailing list -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Short history in terminal (without X)
FuziOK wrote: Jakob Buchgraber wrote: Roman Naumann wrote: Hi, does someone know, how to increase the history-buffer's size of the terminal? (and I mean the pure terminal, without an X-Server) I mean, if I use some commands producing plenty of output, I cannot scroll to the beginning of the text quite often, because the history buffer is to small. Another inconvenient thing is that the buffer seems to forget everything except the last screen of text, if I switch to another terminal. (alt + F2 for instance). How can I make the history buffer larger, or - if possible - set it infinitely large. (Just as the Konsole of KDE.) Thanks, Roman Naumann. To make it infinitely large you can set the variable HISTSIZE to some huge value like export HISTSIZE=1 You could try setting the history to infinitely in Konsole and then do echo $HISTSIZE Cheers, Jay HISTSIZE is used in shell as the number of commands to save in a history list, but cannot increase the history-buffer's size of the terminal. Yes. I was just reading too fast. I noticed this myself after already having sent the e-mail. I am sorry. Cheers, Jay -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild keeps building gcc
On Thursday 15 February 2007 17:46, [EMAIL PROTECTED] wrote: You have to fix some paths in the two mentioned files. I can't remember if it's both files or just one of them. But at least in one of the files the paths are wrong. Check bugs.gentoo.org to find more. Will do... Thanks! Pete. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild keeps building gcc
On Thursday 15 February 2007 18:37:22 Peter Lewis wrote: I've just run revdep-rebuild, after updating world, and it rebuilt gcc-4.1.1-r3, due to the following libraries: /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/libgcjawt.la /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/libgij.la Once it finished, I ran revdep-rebuild once again, just to check that all was okay, but it wanted to rebuild gcc again, due to the same libraries. In fact, I let it do it again, and it's not fixed it. Does this point to some kind of inconsistency on my system? In the future please create a new thread by choosing new mail rather than replying to another thread and changing the subject... https://bugs.gentoo.org/show_bug.cgi?id=125728#c29 -- Bo Andresen pgpKBeWmmCPY1.pgp Description: PGP signature
Re: [gentoo-user] revdep-rebuild keeps building gcc
Peter Lewis wrote: Hi all, I've just run revdep-rebuild, after updating world, and it rebuilt gcc-4.1.1-r3, due to the following libraries: /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/libgcjawt.la /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/libgij.la Once it finished, I ran revdep-rebuild once again, just to check that all was okay, but it wanted to rebuild gcc again, due to the same libraries. In fact, I let it do it again, and it's not fixed it. Does this point to some kind of inconsistency on my system? Thanks, Pete. Following a suggestion on this list, here's what I did to correct this problem: Use your favorite text editor to correct the dependency_libs line in each of these files. Each line starts with something like dependency_libs=' /usr/lib/gcc/some library ... but should start with dependency_libs=' /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/some library John -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild keeps building gcc
On Thursday 15 February 2007 18:20, Bo Ørsted Andresen wrote: In the future please create a new thread by choosing new mail rather than replying to another thread and changing the subject... Sorry... I didn't know there was a difference. https://bugs.gentoo.org/show_bug.cgi?id=125728#c29 Many thanks, Peter. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild keeps building gcc
On Thursday 15 February 2007 18:32, John Blinka wrote: Following a suggestion on this list, here's what I did to correct this problem: Use your favorite text editor to correct the dependency_libs line in each of these files. Each line starts with something like dependency_libs=' /usr/lib/gcc/some library ... but should start with dependency_libs=' /usr/lib/gcc/i686-pc-linux-gnu/4.1.1/some library Thanks... there's also a symlink option on the bug page which Bo linked to. It's working now. :-) -- gentoo-user@gentoo.org mailing list
[gentoo-user] Re: revdep-rebuild keeps building gcc
On 2007-02-15, Peter Lewis [EMAIL PROTECTED] wrote: On Thursday 15 February 2007 18:20, Bo Ørsted Andresen wrote: In the future please create a new thread by choosing new mail rather than replying to another thread and changing the subject... Sorry... I didn't know there was a difference. When you hit reply the message will contains headers that indicate what message it's a reply to. Real mail/news programs use that header information to sort messages into threads. So the message that you intended to be a new thread actually shows up buried down deep inside an existing thread even though it has absolutely nothing to do with that thread. The net result is that all the people who had decided to stop reading that thread will never see your message. -- Grant Edwards grante Yow! Oh, FISH sticks, at CHEEZ WHIZ, GIN fizz, visi.comSHOW BIZ!! -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] disk capacity mismatch
On Thursday 08 February 2007 21:45, Dan Farrell wrote: On Thu, 8 Feb 2007 13:34:21 -0800 Michael Higgins [EMAIL PROTECTED] wrote: Hello, list -- # df -h FilesystemSize Used Avail Use% Mounted on /dev/hda3 20G 12G 7.5G 61% / udev 236M 2.7M 233M 2% /dev shm 236M 0 236M 0% /dev/shm /dev/hda5 14G 13G 1.3G 91% /home/col/dump /dev/hda6 14G 12G 2.0G 86% /home/col/music so here the sizes added up are ~48.5 gigs, right? and here... Disk /dev/hda: 80.0 GB, 80026361856 bytes we can see the 80 gig drive recognized as such. 16 heads, 63 sectors/track, 155061 cylinders and you have 155,061 cylinders on the disk, but Device Boot Start End Blocks Id System /dev/hda1 * 1 497 250456+ 83 Linux /dev/hda2 4982482 1000440 82 Linux swap / Solaris /dev/hda32483 4410320976984 83 Linux /dev/hda4 44104 99582279614165 Extended /dev/hda5 44104 7184313980928+ 83 Linux /dev/hda6 71844 9958213980424+ 83 Linux you only fill to cylinder 99,582. So 99,582 of 155,061 leaves us only about 64% of the drive used, and your 30 'missing' gigs simply not partitioned off. Unfortunately, since you haven't any more primary partitions, you have space after /dev/hda4 and no way to use it. Hopefully you know something about nondestructive partition resizing. good luck! Or, boot off a LiveCD, tar the last partition contents somewhere off disk, optionally you could delete the files/directories (use shred if you wish), then use fdisk to delete the last partition, create a new extended partition and the desired number and sizes of logical partitions, reboot with the LiveCD, create a new fs type on each of your new partitions and untar back your old partition. There's a catch. Your first new logical partition will need to be at least as large as the data you had in your old partition. If you want to move some of the directories mount points into a new different partition, this would be the time to do it. Instead of tar-ring the complete partition, just tar separately the relevant directories. I'm sure there must be some LVM, EVM type of trick that you could use to achieve the above, but I have always used this, aheam, conventional method to do it. HTH. -- Regards, Mick pgpq5GNg2rVsd.pgp Description: PGP signature
Re: [gentoo-user] Re: thunar won't build?
I hope so. In the past I was always baffled when my searches didn't find anything relevent but all the bugs I reported were duplicates. [Actually, I think there was one what wasn't, but the experience was enough to make me swear off reporting bugs.] I think there is nothing bad in this. At least you know: - the bug you are reporting is already known - some dev has seen it repeteadly, and repetitia iuvant* *Repeating helps, for the non-Latin speakers :) Please don't give up reporting bugs. It is always important. The open bug reporting and solving process is one of the main goodies that makes OSS so marvellously different from proprietary software. m. -- gentoo-user@gentoo.org mailing list
[gentoo-user] Re: thunar won't build?
On 2007-02-15, b.n. [EMAIL PROTECTED] wrote: I hope so. In the past I was always baffled when my searches didn't find anything relevent but all the bugs I reported were duplicates. [Actually, I think there was one what wasn't, but the experience was enough to make me swear off reporting bugs.] I think there is nothing bad in this. At least you know: - the bug you are reporting is already known - some dev has seen it repeteadly, and repetitia iuvant* *Repeating helps, for the non-Latin speakers :) Except that it generates extra work for maintainers who have to mark the reported bug as a duplicate. If you've tripped over a bug that's already been reported, perhaps adding a comment (including how you triggered the bug) to an existing bug is probably more productive. Please don't give up reporting bugs. It is always important. The open bug reporting and solving process is one of the main goodies that makes OSS so marvellously different from proprietary software. OK, I'll start again. -- Grant Edwards grante Yow! My ELBOW is a remote at FRENCH OUTPOST!! visi.com -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge -pv world output: strange differences after Portage upgrade
On Thursday 15 February 2007 23:21, b.n. wrote: Hi, I recently upgraded Portage (and a couple of other unrelated packages). After the upgrade, I noticed that emerge -pv world before and after give two very different results. In particular: 1) I had a lot of sound-related packages that are no more in portage still installed. Okay, I should get rid of them. BEFORE: emerge -pv warned me of all these packages. AFTER: emerge -pv warns me of only one package (djplay) 2) I still have to upgrade dbus to 1.0.2 BEFORE: the dbus upgrade was in the -pv output AFTER: It seems the dbus upgrade is no more in emerge -pv world. 3) I am currently running GCC 4.1.x, but I am keeping gcc 3.3 and 3.4 installed (I could probably get rid of gcc 3.3.x, but whatever...) BEFORE: no gcc upgrade was required AFTER: Portage wants me to upgrade gcc 3.3.6 and 3.4.6 to their minor upgrades. My guess is that at least one of your issues has something to do with the new portage behavior regarding buid time dependencies. This is the relevant message displayed when upgrading: In portage-2.1.2, installation actions do not necessarily pull in build time dependencies that are not strictly required. This behavior is adjustable via the new --with-bdeps option that is documented in the emerge(1) man page. For more information regarding this change, please refer to bug #148870. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge -pv world output: strange differences after Portage upgrade
On Thursday 15 February 2007 23:21:02 b.n. wrote: Hi, I recently upgraded Portage (and a couple of other unrelated packages). After the upgrade, I noticed that emerge -pv world before and after give two very different results. In particular: 1) I had a lot of sound-related packages that are no more in portage still installed. Okay, I should get rid of them. BEFORE: emerge -pv warned me of all these packages. AFTER: emerge -pv warns me of only one package (djplay) Due to bug #48195 [1] being fixed. Installed packages can now satisfy a dependency even if it isn't in the tree or in an overlay that's present. 2) I still have to upgrade dbus to 1.0.2 BEFORE: the dbus upgrade was in the -pv output AFTER: It seems the dbus upgrade is no more in emerge -pv world. This seems weird. What's the output of `emerge -pv dbus` with portage 2.1.2? 3) I am currently running GCC 4.1.x, but I am keeping gcc 3.3 and 3.4 installed (I could probably get rid of gcc 3.3.x, but whatever...) BEFORE: no gcc upgrade was required AFTER: Portage wants me to upgrade gcc 3.3.6 and 3.4.6 to their minor upgrades. Due to bug #4698 [2] and some of the referenced bugs on that being fixed. Portage 2.1.1 only upgraded the latest installed slot of any slotted package. [1] https://bugs.gentoo.org/48195 [2] https://bugs.gentoo.org/4698 -- Bo Andresen pgpPxglx1szzg.pgp Description: PGP signature
Re: [gentoo-user] Re: thunar won't build?
Grant Edwards ha scritto: On 2007-02-15, b.n. [EMAIL PROTECTED] wrote: I hope so. In the past I was always baffled when my searches didn't find anything relevent but all the bugs I reported were duplicates. [Actually, I think there was one what wasn't, but the experience was enough to make me swear off reporting bugs.] I think there is nothing bad in this. At least you know: - the bug you are reporting is already known - some dev has seen it repeteadly, and repetitia iuvant* *Repeating helps, for the non-Latin speakers :) Except that it generates extra work for maintainers who have to mark the reported bug as a duplicate. If you've tripped over a bug that's already been reported, perhaps adding a comment (including how you triggered the bug) to an existing bug is probably more productive. Sure, but if you weren't aware of the previous bug, a duplicate is better than nothing, perhaps. OK, I'll start again. Sincere thanks. m. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge -pv world output: strange differences after Portage upgrade
Etaoin Shrdlu ha scritto: My guess is that at least one of your issues has something to do with the new portage behavior regarding buid time dependencies. This is the relevant message displayed when upgrading: In portage-2.1.2, installation actions do not necessarily pull in build time dependencies that are not strictly required. This behavior is adjustable via the new --with-bdeps option that is documented in the emerge(1) man page. For more information regarding this change, please refer to bug #148870. This could explain the issue (2)about the disappearing of the dbus upgrade. Still it doesn't explain the (1) and the (3), imho. m. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge -pv world output: strange differences after Portage upgrade
On Friday 16 February 2007 01:09:30 b.n. wrote: In portage-2.1.2, installation actions do not necessarily pull in build time dependencies that are not strictly required. This behavior is adjustable via the new --with-bdeps option that is documented in the emerge(1) man page. For more information regarding this change, please refer to bug #148870. This could explain the issue (2)about the disappearing of the dbus upgrade. It could not as dbus is both a DEPEND and an RDEPEND of at the very least wine (the dbus USE flag is enabled) which is being upgraded... -- Bo Andresen pgpqN7VfOAsHG.pgp Description: PGP signature
Re: [gentoo-user] emerge -pv world output: strange differences after Portage upgrade
Bo Ørsted Andresen ha scritto: 1) I had a lot of sound-related packages that are no more in portage still installed. Okay, I should get rid of them. BEFORE: emerge -pv warned me of all these packages. AFTER: emerge -pv warns me of only one package (djplay) Due to bug #48195 [1] being fixed. Installed packages can now satisfy a dependency even if it isn't in the tree or in an overlay that's present. Oh, ok. This may make sense. (I preferred the old behaviour, though. We grow accustomed to bugs as features) 2) I still have to upgrade dbus to 1.0.2 BEFORE: the dbus upgrade was in the -pv output AFTER: It seems the dbus upgrade is no more in emerge -pv world. This seems weird. What's the output of `emerge -pv dbus` with portage 2.1.2? These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild U ] sys-apps/dbus-1.0.2 [0.62-r2] USE=X -debug -doc (-selinux) (-gtk%*) (-mono%) (-python%*) (-qt3%*) (-qt4%*) 1,368 kB Total: 1 package (1 upgrade), Size of downloads: 1,368 kB Note that before dbus 0.62 was somehow blocking, now it is no more blocking. Etaoin Shrdlu referred to another bug that may explain this, but it's quite unclear to me. 3) I am currently running GCC 4.1.x, but I am keeping gcc 3.3 and 3.4 installed (I could probably get rid of gcc 3.3.x, but whatever...) BEFORE: no gcc upgrade was required AFTER: Portage wants me to upgrade gcc 3.3.6 and 3.4.6 to their minor upgrades. Due to bug #4698 [2] and some of the referenced bugs on that being fixed. Portage 2.1.1 only upgraded the latest installed slot of any slotted package. I thought it was a feature too. Although the new behaviour makes a lot of sense. m. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: revdep-rebuild keeps building gcc
On Thursday 15 February 2007 19:30, Grant Edwards wrote: When you hit reply the message will contains headers that indicate what message it's a reply to. Real mail/news programs use that header information to sort messages into threads. I've just turned threads on in my mail client and I see what you mean! Thanks for the heads up. :-) -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge -pv world output: strange differences after Portage upgrade
On Friday 16 February 2007 01:33:00 b.n. wrote: 2) I still have to upgrade dbus to 1.0.2 BEFORE: the dbus upgrade was in the -pv output AFTER: It seems the dbus upgrade is no more in emerge -pv world. This seems weird. What's the output of `emerge -pv dbus` with portage 2.1.2? These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild U ] sys-apps/dbus-1.0.2 [0.62-r2] USE=X -debug -doc (-selinux) (-gtk%*) (-mono%) (-python%*) (-qt3%*) (-qt4%*) 1,368 kB Total: 1 package (1 upgrade), Size of downloads: 1,368 kB Note that before dbus 0.62 was somehow blocking, now it is no more blocking. Etaoin Shrdlu referred to another bug that may explain this, but it's quite unclear to me. I'll suggest you file a bug with emerge --info in a comment and emerge --debug -vp dbus attached. This clearly seems like like something is wrong and if nothing else at least we'll find an explanation. -- Bo Andresen pgpqxsfmS6R16.pgp Description: PGP signature
Re: [gentoo-user] emerge -pv world output: strange differences after Portage upgrade
On Friday 16 February 2007 00:44:16 Bo Ørsted Andresen wrote: On Friday 16 February 2007 01:33:00 b.n. wrote: 2) I still have to upgrade dbus to 1.0.2 BEFORE: the dbus upgrade was in the -pv output AFTER: It seems the dbus upgrade is no more in emerge -pv world. This seems weird. What's the output of `emerge -pv dbus` with portage 2.1.2? These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild U ] sys-apps/dbus-1.0.2 [0.62-r2] USE=X -debug -doc (-selinux) (-gtk%*) (-mono%) (-python%*) (-qt3%*) (-qt4%*) 1,368 kB Total: 1 package (1 upgrade), Size of downloads: 1,368 kB Note that before dbus 0.62 was somehow blocking, now it is no more blocking. Etaoin Shrdlu referred to another bug that may explain this, but it's quite unclear to me. I'll suggest you file a bug with emerge --info in a comment and emerge --debug -vp dbus attached. This clearly seems like like something is wrong and if nothing else at least we'll find an explanation. On second thought this does make sense. Portage 2.1.2 allows an upgrade within the same slot despite the block. And the reason it doesn't get pulled in by `emerge -pv world` is because dbus isn't in world and the later version isn't required by anything in world. You need to use --update to get direct dependencies upgraded or --deep to get all dependencies upgraded to the latest version. -- Bo Andresen pgpUugGLZWkBv.pgp Description: PGP signature
Re: [gentoo-user] emerge -pv world output: strange differences after Portage upgrade
On Friday 16 February 2007 02:00:26 Bo Ørsted Andresen wrote: On Friday 16 February 2007 00:44:16 Bo Ørsted Andresen wrote: On Friday 16 February 2007 01:33:00 b.n. wrote: 2) I still have to upgrade dbus to 1.0.2 BEFORE: the dbus upgrade was in the -pv output AFTER: It seems the dbus upgrade is no more in emerge -pv world. [SNIP] On second thought this does make sense. Portage 2.1.2 allows an upgrade within the same slot despite the block. And the reason it doesn't get pulled in by `emerge -pv world` is because dbus isn't in world and the later version isn't required by anything in world. You need to use --update to get direct dependencies upgraded or --deep to get all dependencies upgraded to the latest version. Just to close this entirely dbus-0.62-r2 has had most of it's keywords dropped. This means if you type: # emerge -pv =sys-apps/dbus-0.62-r2 you'll likely get a masked by: missing keyword message. So again bug #48195 explains why dbus-0.62-r2 is able to satisfy dependencies on dbus because it's installed with Portage 2.1.2 but isn't able to satisfy it with Portage 2.1.1 because it's not installable. :) -- Bo Andresen pgp94W9pHhevk.pgp Description: PGP signature
[gentoo-user] udev and glibc update
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, I gave a sync 2 days ago and my world update had that output: # emerge -upvD world These are the packages that would be merged, in order: Calculating world dependencies | !!! Ebuilds for the following packages are either all !!! masked or don't exist: sys-fs/raidtools ... done! [ebuild U ] sys-fs/udev-104-r11 [104-r9] USE=(-selinux) 0 kB [ebuild U ] sys-libs/glibc-2.5 [2.4-r4] USE=nls nptl nptlonly -build -glibc-compat20 -glibc-omitfp -hardened (-multilib) -profile (-selinux) 0 kB If I start to update both upgrade failed! Udev gives: [...] Source unpacked. Compiling source in /var/tmp/portage/sys-fs/udev-104-r11/work/udev-104 ... /usr/bin/x86_64-pc-linux-gnu-ar get_libdir = lib64 GENHDR udev_version.h CC udev_device.o CC udev_config.o CC udev_node.o CC udev_db.o CC udev_sysfs.o CC udev_rules.o CC udev_rules_parse.o CC udev_utils.o CC udev_utils_string.o CC udev_utils_file.o CC udev_utils_run.o CC udev_sysdeps.o AR libudev.a RANLIB libudev.a make: x86_64-pc-linux-gnu-ranlib: Kommando nicht gefunden make: *** [libudev.a] Fehler 127 !!! ERROR: sys-fs/udev-104-r11 failed. and glibc: [...] checking whether we need to use -P to assemble .S files... no checking whether .text pseudo-op must be used... yes checking for assembler global-symbol directive... .globl checking for .set assembler directive... yes checking for assembler .type directive prefix... @ checking for .symver assembler directive... yes checking for ld --version-script... yes checking for .previous assembler directive... yes checking for .protected and .hidden assembler directive... yes checking whether __attribute__((visibility())) is supported... yes checking for broken __attribute__((visibility()))... no checking for broken __attribute__((alias()))... no checking whether to put _rtld_local into .sdata section... no checking for .preinit_array/.init_array/.fini_array support... /var/tmp/portage/sys-libs/glibc-2.5/work/glibc-2.5/configure: line 5513: readelf: command not found no configure: error: Need linker with .init_array/.fini_array support. !!! ERROR: sys-libs/glibc-2.5 failed. How can I solve those problems? I rebuilded gcc but no change... Thanks, Luigi - -- Public key GPG(0xC5CB65CD) on hkp://pgp.mit.edu -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFF1Vx/5ZpKrsXLZc0RAjPPAKCPLYyA6tQkP21nPCUsigMsWugZIwCgmYoT ZmYZ9ho4VVvIrEYSCGN+RxM= =WTyy -END PGP SIGNATURE- -- gentoo-user@gentoo.org mailing list
