Re: [gentoo-user] Debian just voted in systemd for default init system in jessie
On Thu, 20 Feb 2014 19:40:46 -0600 Canek Peláez Valdés can...@gmail.com wrote: [...] But I'm going to save you some bucks: there is nothing fishy. Carry on with the wires on the tin hat. Regards. Perfect. So that nails that bugaboo as well. All is good, then, absolutely nothing to see here. Thanks again for all you've done to convince us there's nothing nefarious going on. Cheers, - mykhyggz
Re: [gentoo-user] Debian just voted in systemd for default init system in jessie
On Thu, 20 Feb 2014 15:38:46 -0600 Canek Peláez Valdés can...@gmail.com wrote: On Thu, Feb 20, 2014 at 3:22 PM, Tanstaafl tansta...@libertytrek.org wrote: On 2014-02-20 4:04 PM, Canek Peláez Valdés can...@gmail.com wrote: No, actually, I think whatever is defined as the current default should dictate which group should be required to do the work. I think this is where we think differently (regarding this particular point). The work must be done by *whomever* wants to do the job. So if the systemd people want to do a profile that's fine (and this already happened); but if they don't want to, nobody can force them to do it (this is academic right now, since they already did the [pretty trivial] work). If the systemd people did not wanted to do the job, then, since you can't force them, the people *not* wanting systemd would be the ones required to do it. And that makes absolutely no sense. I think we agree, but you keep saying we don't... ;) The difference is, since OpenRC is the default, all of the existing profiles are, by definition, OpenRC based profiles. So, people who don't want systemd simply have to do... nothing! OK, I see your point. As I said before, it is people who want systemd who currently have to 'do something', and that is as it should be, unless/until the default of OpenRC is changed to systemd. Agreed. Regards. Been following along, jumping in late. That does settle the issue with regard to Gentoo 'defaults' for now. As I see it, if you are a desktop [l]user, then go ahead and use a quick systemd default. It's good enough. And Gnome folks, from what I understood, didn't have an easy opt-out choice, really. At least not easy enough for me to grok right away. Anyway, some arguments on the other branches of this thread got me thinking. Take the binary log files. This seems to be something we all understand to some extent (as compared to init systems, say), whether a tinkerer or enterprise user. There's now something corporate sitting between me and the evil crap coming out of my daemons. Who logs the loggers? My thought was, well, say if someone else, some other, unrelated team entirely, was maintaining a codebase for handling that logging data, that's a way out of the monoculture of systemd that would give some peace of mind. If they, as also smart, talented coders began to disagree, they might have capacity to keep the project in bounds. So, for the hell of it, I just went to have a look at what the deal might be with that binary data that hangs around and dumps to disk, maybe, in one format or another. This document explains the basic structure of the file format on disk. We are making this available primarily to allow review and provide documentation. Note that the actual implementation in the systemd codebase is the only ultimately authoritative description of the format, so if this document and the code disagree, the code is right. That said we'll of course try hard to keep this document up-to-date and accurate. Instead of implementing your own reader or writer for journal files we ask you to use the Journal's native C API to access these files. It provides you with full access to the files, and will not withhold any data. If you find a limitation, please ping us and we might add some additional interfaces for you. From: http://www.freedesktop.org/wiki/Software/systemd/journal-files/ Wow, that makes me feel all warm, fuzzy and secure as shit. I think I just fell asleep for a moment. (My background would make me read this, essentially advertising copy, as a classic example of perverse use of semiotics, thus raising the reading on my B.S. meter. But I digress.) The best part is the conclusion: *If you find a limitation, please ping us and we might add some additional interfaces for you.* Please! Ping! Second best is, we'll of course try hard to keep this document up-to-date and accurate. Of course. So. If I were paranoid, or even mildly concerned, I'd take this to mean two things: One. There is, or could be, data that the systemd logging magic box doesn't give you an interface for. Unless you know what it is, then they might let you log it. Or you can, of course, contribute code, I'm sure. Though, Note that the actual implementation in the systemd codebase is the only ultimately authoritative description of the format. Two. If I want to have a reasonable degree of surety that the code isn't collecting data that isn't interfaced for you in the journalcontrol, I'd have to delve into the codebase for systemd. Which, I'm sure, is a piece of cake. And, of course, I, and my security team, have to review the basic code on my GNU/Linux F[L]OSS machine regularly anyway. Oh, wait... No I don't. The beauty of the development model I trust is a diverse group of folks with a *variety of interests* contributing to various areas of code, in turn reviewed by some set of that same varied group, eventually committed and
Re: [gentoo-user] Debian just voted in systemd for default init system in jessie
On Thu, 20 Feb 2014 15:37:09 -0800 Michael Higgins li...@evolone.org wrote: Okay, I'll go re-wire my tin hat now. Hope someone found this amusing. One other thought I'd has was, well, as long as systemd doesn't, like, implement some kind of net protocol, so to make it possible to ship logs from systemdjournalcontrol. I mean, so what if then it just sits there doing its job, more or less. You can even offload the job to your favourite logger, and it just sits there, pretty much inoffensively in the pid 1 family. Then I read this: I am curently working on getting log syncing via both a PUSH and PULL model done. This will be based one existing protocols and standards as much as we can (SSH or HTTP/HTTPS as transport, and JSON and more as payload), and is flexible for others to hook into. For example, I think it would be cool if greylog2 and similar software would just pull the data out of the journal on its own, simply via HTTP/JSON. We make everything available to make this smooth, i.e. we provide clients with stable cursors which they can use to restart operation. So, I don't know. That 'journalctl' seems to be really just a front-end to some routines actuall coded in sytemd. So, I guess, windows all over again? Really, a net protocol at that level? I'm sure there's no way to exploit that. What I do know is, that if any point I work again in IT, it will just be part of the way it is done, for the same reasons most shops don't use Gentoo, for example. No one will blame you for using whatever, if pretty much everyone else does so too. Plus, if there's some massive codebase with a flaw causing a security breach, you're not alone. But, seriously, flawed (carnivorous or nefarious) by design does seem to be the upshot here, the more I read about the design decisions. I don't think I've presented any compelling argument against anyone else using it, per se, more than questioning the logic of it becoming ubiquitous. It does seem to solve a lot of real world problems, in one way, or another. And I'm done looking at it, I think, and will leave it to others to take my notes with the grains of salt required. For me, I don't really have a need for what it offers, nor a real world need to worry about possible implications of its design or rate of adoption. Cheers, Cheers, - mykhyggz
Re: [gentoo-user] Is there a way to keyword a whole overlay as ~arch ?
On Sun, 26 Jan 2014 20:14:17 -0500 Michael Orlitzky m...@gentoo.org wrote: On 01/26/2014 05:12 PM, Thanasis wrote: I am following stable (ACCEPT_KEYWORDS=amd64). In order to install the mate desktop I need to install the mate overlay and keyword all its packages as ~amd64. Is there a way to do it easily for the whole overlay? Yep, you just need to know the overlay name. In package.accept_keywords: */*::overlay-name ~amd64 The */* syntax means any category, any package name. The :: then specifies the repository name. If you don't know the repository name, it can often be found in path/to/overlay/profiles/repo_name. I wonder, can we get a hint of this function echo-ed at the end of every new layman install? It would make using, say, perl-experimental, far less unwieldy than without. (Apologies in advance for the noise.) I had no idea this was possible, but it seems the only way to use the overlay without making of mess of package.accept_keywords, which is what I have had when installing anything useful in the perl development area. Does this make any sense? Do all the overlays work that way, that is, kw-masking everything so you have to enable the ~arch per package? This always seemed absurd to me, as I added the overlay, I must have meant to use it... but anyway... I suppose it should be printed in red use if you know what it means, kind of thing. I can see it being a PITA if it breaks stuff in the main gentoo tree. . . . FWIW, I tried adding that incantation and emerging world, which gave no changes to my install. Then I tried: emerge -av dev-perl/Catalyst-Model-DBIC-Schema (Which created a bloated keyword file on my other machine.) This time, it only wants to unmask things perl in the gentoo tree... but --autounmask-write proposes to list every overlay dep as a comment. Ugh. Before, I'd do something like: mv /etc/portage/package.accept_keywords ~/ mkdir /etc/portage/package.accept_keywords cp package.accept_keywords /etc/portage/package.accept_keywords/99_portage eix -c --only-in-overlay 0 -C dev-perl|grep '(~' |cut -d ' ' -f 2 | while read a ; do echo $a ~x86 ;done /etc/portage/package.accept_keywords/00_perl emerge -auDNtv world Nothing to merge; quitting. ... Which I did. Then /me thinks, for a change. If */*::overlay-name ~amd64, then: dev-perl/*::gentoo ~amd64 should work too. Now, emerge -av dev-perl/Catalyst-Model-DBIC-Schema Gives: Total: 37 packages (34 upgrades, 3 new), Size of downloads: 5,905 kB Would you like to merge these packages? [Yes/No] And no guff from portage. OMG, what a treat. I wept. . . . So wow, if I'd known about that before now, it would have saved me hours, if not days, worth of hassle. I think that should be the recommendation for anyone who installs the perl overlay. Maybe I'm crazy, but it seems to DWIW. Thanks Michael. Very helpful to know, at least. Does anyone who may have read this far think this would be a good thing to mention from the start for a new user of an overlay, like echoed at the end of emerging layman, or adding a new overlay? And, yes, it's a full moon, so I'm posting to the list. As is my habit. ;-) Cheers [I recommend Flat Tail seriously budget barley wine], - mykhyggz (who lost his .sig)
Re: [gentoo-user] llvm compile error
On Fri, 24 Jan 2014 19:44:38 +0100 Silvio Siefke siefke_lis...@web.de wrote: Hello, On Fri, 24 Jan 2014 09:00:26 +0200 Alan McKinnon alan.mckin...@gmail.com wrote: llvm-3.4 is not hardmasked. It is marked ~x86 which is something very different. Correct, i mean Clang in version 3.4 is hardmasked. I want install clang alone llvm i not need. I think mesa use llvm too but i has not set the use flag. https://packages.gentoo.org/package/sys-devel/clang I don't know why I'm replying, as I'm no expert. So, I could be way off, but: From a google search: Clang /ˈklæŋ/ is a compiler front end for the C, C++, Objective-C and Objective-C++ programming languages. It uses LLVM as its back end ... The mistake you made is using --autounmask-write That feature writes local package.unmask entries to satisfy dependencies. It's a very blunt tool, it blindly keywords whatever it thinks it needs to and when it goes wrong, it goes very wrong quickly. The tool write package.accept_keywords, a package.unmask i have not on system. Yes its a shit tool, normal i make echo package ~x86 package.a.. What should do? That's what you should do. The output will tell you what files need updating. In some cases, I just let it write, like for a big perl package with lots of modules from the overlay. Otherwise it's faster to just copy and paste the output. Should i compile without portage? No. Should i not use? When i want emerge clang is masked, ok why? https://bugs.gentoo.org/show_bug.cgi?id=408963 maybe? FreeBSD use it and say is stable. FreeBSD maybe not the reference on earth but the BSD's make a good job. When i saw all versions of Clang is masked. Isn't FBSD more similar than, say, Linux, to what OSX is, sort of? [...] So i think for some packages is accept between risk and compromiss. I find better x86 flag better as install software in /usr without portage. Yep. Sometimes you have to unmask stuff, but portage can handle it then. Anyway, I hope any of this helps. Good luck! Cheers, -- - - - Michael Higgins - michael_higg...@iinet.com 503-473-5882
[gentoo-user] Help to upgrade perl?
Hey, all -- I have two systems, one of which got perl 5.16.1, somehow. My other system is still at perl 5.12... and I'm having a heck of a time trying to upgrade that system to 5.16.1. Is there some trick that I should recall? This is what I tried: USE=-build emerge -v =dev-lang/perl-5.16.1 And this is what I got: (dev-lang/perl-5.12.4-r1::gentoo, installed) pulled in by ~dev-lang/perl-5.12.4 required by (virtual/perl-i18n-langtags-0.35::gentoo, installed) (dev-lang/perl-5.16.1::gentoo, ebuild scheduled for merge) pulled in by =dev-lang/perl-5.16.1 =dev-lang/perl-5.16 required by (dev-perl/IO-Socket-SSL-1.840.0::gentoo, installed) =dev-lang/perl-5.16* required by (virtual/perl-digest-base-1.170.0-r1::gentoo, installed) (and 144 more with the same problems) 144 more with the same problem? There must be some incantation that handles all that, no? Thanks for any pointers, foax! Cheers, - Michael Higgins
Re: [gentoo-user] Help to upgrade perl? [SOLVED]
On Mon, 02 Dec 2013 22:24:35 +0200 Alan McKinnon alan.mckin...@gmail.com wrote: emerge -avuND world /me slaps forehead. Of course. :( Why 5.16.1? that is the lowest version that is ~arch; your next sync and update is going to want to upgrade it anyway. Ah, well, I'm not running ~arch anywhere and so not sure why my public server got 5.16.1 to begin. So, I'm just updating the development machine to match. Lots of things have been moving around in perl land, modules moving into core etc etc. A proper world update takes care of all the necessary blockers. Yes. Perl surely is a PITA for the distro to maintain. Or, seeing as you are going to remerge everything perl related anyway, you might as well just undelete everything perl-related (keeping an accurate list), and then merge the whole lot back This kind of on-the-side bookkeeping is exactly what I'm trying to avoid. But I do get your point. ;-) Looks like I'm good to go, as perl is updating and perl-cleaner --all after the fact should be fine, I'm guessing. And if not, I know what I'm in for, mostly. Thanks! Cheers, - Michael
Re: [gentoo-user] SystemD + Gnome 3.8 I can log in as root but not as normal user
On Sat, 5 Oct 2013 18:08:38 -0600 Carlos Sura carlos.su...@googlemail.com wrote: Update: I was desperate trying to rebuild everything or reinstalling or whatever idea I had. So, my logic said: if root works why others users wont?, then I decided to create another user, and guess what? it worked. So I removed my user and then created a new one and it worked. Yeah this is fine and great that worked, but I would like to know: Why? Thanks. On 5 October 2013 17:10, Carlos Sura carlos.su...@googlemail.com wrote: Hi, I do have -consolekit systemd policykit and yes, my kernel is properly configured. To avoid playing around I did reinstalled it again! and guess what? After installing gnome-tweak-tools and shell-extensions and use the tweak tools to do some changes, after reboot, again: blank screen trying to start gnome. So, I unmerged the packages and did: emerge @preserved-rebuild revdep-rebuild emerge -1 gnome but seems not to work, again, I can log in as root, not as user. On 1 October 2013 06:49, Canek Peláez Valdés can...@gmail.com wrote: On Tue, Oct 1, 2013 at 1:30 AM, Carlos Sura carlos.su...@googlemail.com wrote: Hello Mates, I finally thought that I got this working, so now I am dealing this issue: I choose the latest kernel with systemd and GDM starts good, but I cannot log in to gnome as a normal user, I can only log in to gnome as root. What I have done before this started: emerge -uDvaN world emerge --depclean --ask emerge @preserved-rebuild Then it just happened. What I've tried so far: - .xinitrc with exec gnome-session for user and root - emerge -1 $( qlist -IC x11-drivers/ ) - emerge @x11-module-rebuild - X -configure Another note: Starting gnome as normal user just hang, apparently it wants to start but it cannot start. Neither doing: startx. I am using for xorg: VIDEO_CARDS=intel vesa Any help? Could you please boot up your machine, let GDM start, and then from a console do systemctl --all --full, and post here its output? Also, can you try to do: USE=systemd -consolekit policykit emerge -uDNvp world Does any package wants to be reemerged? Lastly, does your kernel contains all the configuration options required by systemd/udev? Whenever I have trouble with gnome, the first thing I do is remove any .gnome*-type folders in ~/. My guess is that in some cases gnome saved data isn't compatible with the new gnome configuration. Cheers, -- - - - Michael Higgins -
Re: [gentoo-user] Fwd: Gentoo Prefix Issue...
On Wed, 6 Jun 2012 15:22:45 +1000 Nima Talebi n...@it.net.au wrote: Hi All, Could someone point out what's the deal with the following: % emerge --oneshot virtual/os-headers Calculating dependencies... done! Verifying ebuild manifests Starting parallel fetch Emerging (1 of 2) sys-kernel/linux-headers-3.3 * gentoo-headers-base-3.3.tar.xz RMD160 SHA1 SHA256 size ;-) ... [ ok ] * gentoo-headers-3.3-2.tar.xz RMD160 SHA1 SHA256 size ;-) ... [ ok ] Unpacking source... Unpacking gentoo-headers-base-3.3.tar.xz to /home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/work This process ultimately fails, because the unpacking drops the files into gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/work as root, and then later things begin to fail as it (running as a non-root user) doesn't have permissions to chmod, chown, etc: ... chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/drm_dp_helper.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/drm_hashtab.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/drm_usb.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/drm_fixed.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/drmP.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/drm_mm.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/vmwgfx_drm.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/i810_drm.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/drm/drm_crtc.h': Operation not permitted chmod: changing permissions of `gentoo-headers-base-3.3/include/config': Operation not permitted make headers_install INSTALL_HDR_PATH=/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/image///home/talebi/gentoo/usr/include/.. ARCH=x86 CROSS_COMPILE=x86_64-pc-linux-gnu- make[1]: Entering directory `/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/work/gentoo-headers-base-3.3' CHK include/linux/version.h /home/talebi/gentoo/tmp/usr/bin/bash: include/linux/version.h.tmp: Permission denied make[1]: *** [include/linux/version.h] Error 1 make[1]: Leaving directory `/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/work/gentoo-headers-base-3.3' emake failed * ERROR: sys-kernel/linux-headers-3.3 failed (install phase): * (no error message) * * Call stack: * ebuild.sh, line 85: Called src_install * environment, line 2673: Called kernel-2_src_install * environment, line 1909: Called install_headers * environment, line 1804: Called die * The specific snippet of code: * emake headers_install INSTALL_HDR_PATH=${D}/${ddir}/.. ${xmakeopts} || die; * * If you need support, post the output of 'emerge --info =sys-kernel/linux-headers-3.3', * the complete build log and the output of 'emerge -pqv =sys-kernel/linux-headers-3.3'. * The complete build log is located at '/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/temp/build.log'. * The ebuild environment file is located at '/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/temp/environment'. * Working directory: '/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/work/gentoo-headers-base-3.3' * S: '/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/work/gentoo-headers-base-3.3' Failed to emerge sys-kernel/linux-headers-3.3, Log file: '/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/temp/build.log' * Messages for package sys-kernel/linux-headers-3.3: * ERROR: sys-kernel/linux-headers-3.3 failed (install phase): * (no error message) * * Call stack: * ebuild.sh, line 85: Called src_install * environment, line 2673: Called kernel-2_src_install * environment, line 1909: Called install_headers * environment, line 1804: Called die * The specific snippet of code: * emake headers_install INSTALL_HDR_PATH=${D}/${ddir}/.. ${xmakeopts} || die; * * If you need support, post the output of 'emerge --info =sys-kernel/linux-headers-3.3', * the complete build log and the output of 'emerge -pqv =sys-kernel/linux-headers-3.3'. * The complete build log is located at '/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/temp/build.log'. * The ebuild environment file is located at '/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/temp/environment'. * Working directory: '/home/talebi/gentoo/var/tmp/portage/sys-kernel/linux-headers-3.3/work/gentoo-headers-base-3.3' * S:
Re: [gentoo-user] vmplayer wont start
On Thu, 08 Sep 2011 10:29:38 +0200 ifj. Stefán István iste...@stef.hu wrote: Hello! I've installed vmware-player, but can't start it. It outputs a lot of messages [ 8 ] It wont start even if I add a virtual machine config file parameter to vmplayer. Is there any trick to start it? Thanks, István WAG here, but maybe this will help: | http://www.google.com/search?q=vmplayer%20USE_SHIPPED_GTK%20ie=utf-8oe=utf-8aq=trls=org.mozilla:en-US:unofficialclient=firefox-asource=hpchannel=np Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] alsa just mucked up my cards on reboot
IDK what happened. I rarely restart the server. Two cards, both share ac_97 module. via_82xx and ice_1712 Has anyone had to fix this recent development? Any clues as to what changed under the hood? Cheers, --
Re: [gentoo-user] alsa just mucked up my cards on reboot
On Wed, 23 Feb 2011 11:53:44 -0800 Michael Higgins li...@evolone.org wrote: IDK what happened. I rarely restart the server. Two cards, both share ac_97 module. via_82xx and ice_1712 Has anyone had to fix this recent development? Any clues as to what changed under the hood? Cheers, Well, I was able to rmmod ice_1712 and modprobe, restart alsa and use it. Does that give a clue to anyone? Anyone else on this list doing multiple audio cards seeing them suddenly disappear, or whatever? Probably I need to load them (the modules) with different/new options? I'd love any feedback or suggestions as to how to fix this. cat /etc/modprobe.d/alsa.conf alias char-major-116 snd alias snd-card-0 snd-via82xx alias sound-slot-0 snd-via82xx alias snd-card-1 snd_mpu401 alias sound-slot-1 snd_mpu401 alias snd-card-2 snd-ice1712 alias sound-slot-2 snd-ice1712 options snd cards_limit=2 options snd slots=snd-via82xx,snd_mpu401,snd-ice1712 Is there any way to guarantee that this setup will survive a reboot? Cheers, --
[solved] Re: [gentoo-user] alsa just mucked up my cards on reboot
On Wed, 23 Feb 2011 11:53:44 -0800 Michael Higgins li...@evolone.org wrote: IDK what happened. I rarely restart the server. Two cards, both share ac_97 module. via_82xx and ice_1712 Has anyone had to fix this recent development? Any clues as to what changed under the hood? Cheers, Well, the setup survived rebooting... *this* time. IDK what the heck that was all about. Too bad I can't say what caused it. :( Sorry for the noise! Cheers, --
Re: [gentoo-user] postgresql 8.2.14 cannot stop
On Tue, 30 Mar 2010 21:31:50 +0800 Xi Shen davidshe...@googlemail.com wrote: hi, my system is gentoo amd64, postgresql 8.2.14. the postgresql service can start without any error; but when i try to stop it, it reports failure. the process of postgresql are actually terminated. if i try to stop the service again, i got start-stop-daemon: fopen `/var/lib/postgresql/data/postmaster.pid': No such file or direc i will have to use zap to reset the service status. how to fix this? There are new start stop scripts for gentoo that work at http://bugs.gentoo.org/show_bug.cgi?id=311047 Enjoy! -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Re: Is the move perl-5.10 involved here
On Thu, 04 Mar 2010 10:52:30 -0600 Harry Putnam rea...@newsguy.com wrote: Use of uninitialized value in subroutine entry at /usr/lib/perl5/5.10.1/i686-linux/DynaLoader.pm line 223. object version 1.609 does not match $::VERSION %_ at /usr/lib/perl5/5.10.1/i686-linux/DynaLoader.pm line 223. BEGIN failed--compilation aborted at /usr/lib/perl5/vendor_perl/5.8.8/i686-linux/DBI.pm line 263. ^ Your perl installation is broken. That is obvious. What you do to fix it probably should start with perl-cleaner... FWIW my experience with updating it has been fine. Basically, why are you asking... So do you think its related to the update to perl-5.10? and Have you had a similar problem and `perl-cleaner' fixed it? ... because either you ran perl-cleaner or you didn't. It does something you need to do if portage manages perl, so just let it work. ;-) Just have a look a perl-cleaner itself if you are concerned. head `which perl-cleaner` #!/bin/bash Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Re: [OT] attach a perl script to daemon services
On Sat, 27 Feb 2010 12:40:17 -0600 Harry Putnam rea...@newsguy.com wrote: reload() { ebegin Reloading evolone_agi configuration start-stop-daemon --signal 1 --pidfile /var/run/evolone_agi.pid eend $? Error reloading evolone_agi } Thanks ... good info. How does the script behave regarding what happens if it receives a sigINT or sigKILL, or some other way dies (even from an internal error)? This script is more proof-of-concept piece hacked together in an hour which just routes my outgoing calls to local free PSTN or SIP. So I know if it needs restarting when I can't make a call from home. ;-) Basically, I think if you perldoc Net::Server[::SIG] In there is: $self-pre_server_close_hook; ... might be your friend to share some Sys::Syslog routines. In the Net::Server code: ### set some sigs $SIG{INT} = $SIG{TERM} = $SIG{QUIT} = sub { $self-server_close; }; ### most cases, a closed pipe will take care of itself $SIG{PIPE} = 'IGNORE'; ### catch children (mainly for Fork and PreFork but works for any chld) $SIG{CHLD} = \sig_chld; ### catch sighup $SIG{HUP} = sub { $self-sig_hup; } ... seems to have it covered for most cases. ;-) You can see sig_hup is almost expected to do something. Does it log to syslog? and notify sysadmin? I think a good use of Net::Server::* and Sys::Syslog can get you where you want. I'd bet you can put anything in that pre_server_close_hook hook... send an email, set off the building alarms... it's perl, after all. But I really haven't done much directly with it. I haven't had occasion to run into an `EXIT trap' in perl... but there must be such a thing. perldoc -q signal perldoc perlipc HTH. -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Re: [OT] attach a perl script to daemon services
On Sat, 27 Feb 2010 01:00:43 +0200 Alan McKinnon alan.mckin...@gmail.com wrote: Where I'm weak is the part where the custom script checks if the daemon is running, before the script itself starts. That part would need to be something pretty fool proof... maybe just grep ps output for the daemon? I have a perl fastAGI script that launches in /etc/init.d/ I just cribbed some code from something there... in my perl script: log_file='Sys::Syslog', pid_file='/var/run/evolone_agi.pid'; Sys::Syslog is an interface to the UNIX syslog(3) program. from /etc/init.d/evolone_agi depend() { need net asterisk postgresql-8.4 } As for reading new information, there are a gazillion ways. Trap a signal to reread the configs? reload() { ebegin Reloading evolone_agi configuration start-stop-daemon --signal 1 --pidfile /var/run/evolone_agi.pid eend $? Error reloading evolone_agi } .. for example. Hope this helps. '-) -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] firefox 3.5.6 doesn't see my CUPS printers
On Fri, 08 Jan 2010 13:18:38 +0100 (CET) Helmut Jarausch jarau...@igpm.rwth-aachen.de wrote: Hi, Firefox (www-client/firefox-bin-3.5.6) doesn't see my CUPS printers anymore. It just shows Any Printer and lateron Print to LPR What am I missing? WAG: Start cupsd? -- Michael Higgins
Re: [gentoo-user] freezing a package
On Sat, 19 Dec 2009 19:01:03 -0500 Michael P. Soulier msoul...@digitaltorque.ca wrote: On 19/12/09 Albert Hopkins said: Create an local overlay. Put the ebuild in there. I'll need to read up on how. The section on overlays left me with the impression that overlays were for experimental code, not for keeping private copies of packages. Yeah, that's too bad, because having your own overlay is a Good Thing, IMO, in that it gives you some starting point for your own packages. Anyway, I think I'll remove it from the world file, and mask it out, and look into the overlay. At least it will be ignored when I emerge world. Why remove it from world file? Just put it in your overlay and mask any later versions, like 'whateverpackage-N.NN-rN' (the version you want to keep) in package.mask, I think. I found that the in the package.mask wasn't always enough though, as if I used the --update argument to emerge and my version wasn't in the portage tree anymore, emerge would get silly and offer to Except that it *will* be in the portage tree, in your personal overlay, unless *you* delete your ebuild. ;-) downgrade. Cheers, -- Michael Higgins
Re: [gentoo-user] [OT] gnome-shell is cool (and how to move my current user)
On Mon, 23 Nov 2009 20:54:13 -0800 Michael Higgins li...@evolone.org wrote: So then, what's the *best way* to (re-) set my current user to get the goodness of a brand new gnome profile, but none of the hassle of reconfiguring all my apps? ;-) I see moving my /home folder to something else, deleting me, adding me back, with new /home and my groups... then moving a few very configured apps .configs over, but I thought there *might* be a sure way that would be easier... Of course, I couldn't wait. User is all set and I'm obviously no linux sysadmin... or whatever. Problem with slowness using gnome-shell was caused by user being in the 'video' group. Hmm. Not entirely needed, I guess. Can't say what's up with that. Anyway, I might stick with this, be ahead of the curve, or suffer needlessly, for a bit. Nice to know I worked on documenting a fix for vertical gnome-panels, but may not need it now... Sorry for the noise. Cheers, -- Michael Higgins
[gentoo-user] [OT] gnome-shell is cool (and how to move my current user)
So I've been trying to get around a long-standing bug in gnome-panel with vertical panels and the application list. Some workarounds, patches exist that work, but still a PITA to fix up my own ebuilds to address it. I received a comment on a bug tracker that gnome-panel likely will be deprecated in favour of gnome-shell in gnome 3.0 (aka 2.32) anyway, so forget an *upstream* commit to fix this problem that's been around for years. Well, my curiosity got the best of me, so I cherry-picked some ebuilds, an eclass (and even built a piece by hand since the ebuild failed to actually install anything...) and I got gnome-shell --replace to launch. What I found was the system became so slow as to be unusable... *but* that adding a new user, so to test on a clean slate as it were, gnome-shell works just fine. Very cool. So then, what's the *best way* to (re-) set my current user to get the goodness of a brand new gnome profile, but none of the hassle of reconfiguring all my apps? ;-) I see moving my /home folder to something else, deleting me, adding me back, with new /home and my groups... then moving a few very configured apps .configs over, but I thought there *might* be a sure way that would be easier... Cheers, -- Michael Higgins
Re: how to force libxcb version below 1.4 when re-installing X (was: [gentoo-user] my xorg-server 1.6 seems a bit unstable - what am I doing wrong?)
On Tue, 13 Oct 2009 22:36:18 -0400 Denis denis@gmail.com wrote: Of course, I forget to attach the file to embarrass myself to the end... Great. :-P Oh, it's not over yet! Here it is! make[1]: *** [ks_tables.h] Error 136 Any constructive thoughts appreciated! :-) Heh. Let me google that *exact error string* for you, by way of illustration... OK, so someone using Gentoo sez: Ultimately, you need to check the version of xproto that you have installed. By masking 7.0.15 and re-emerging 7.0.14, I was able to resume the emerge without any remaining libX11 errors. From: http://blog.olebox.com/tag/linux/ Cheers (and good luck), -- Michael Higgins
Re: [gentoo-user] Gentoo and Ultramonkey
On Mon, 28 Sep 2009 00:37:58 + Nick Khamis sym...@gmail.com wrote: [don't top-post, please] On Sun, Sep 27, 2009 at 11:33 PM, Volker Armin Hemmann volkerar...@googlemail.com wrote: On Montag 28 September 2009, Nick Khamis wrote: Hello Everyone I am new to gentoo, usually I use Debian. I have a fresh install x86 2.6.30 kernel and I was trying to get Ultramonkey going on the box. I looking for the equivilant to the following deb packages for Gentoo please * libdbi-perl dev-perl/DBI ... is dbi-perl * libdbd-mysql-perl * dev-perl/DBD-mysql so, just: emerge [-pv] dev-perl/DBD-mysql, which needs what it needs: These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild N] dev-db/mysql-init-scripts-1.2 0 kB [ebuild N] virtual/perl-Sys-Syslog-0.27 0 kB [ebuild N] dev-perl/Net-Daemon-0.43 28 kB [ebuild N] sys-apps/ed-1.0 67 kB [ebuild N] dev-perl/PlRPC-0.2020-r1 18 kB [ebuild N] dev-perl/DBI-1.601 484 kB [ebuild N] dev-db/mysql-5.0.70-r1 USE=berkdb perl ssl -big-tables -cluster -debug -embedded -extraengine -latin1 -max-idx-128 -minimal (-selinux) -static 28,028 kB [ebuild N] virtual/mysql-5.0 0 kB [ebuild N] dev-perl/DBD-mysql-4.00.5 120 kB and check the use flags... if it looks good, let it rip! * libmysqlclient14-dev Well, I think Gentoo is different, in that you always get the required development libs, say to build DBD::mysql. (If not, there's something wrong with the ebuild.) I think dev-db/mysql-community is the server part of mysql, but not sure. If you feel that there is anything else I need to know about getting Ultramonkey working on Gentoo please let me know. No idea about ultramonkey, but HTH. Hello Gentlemen, Thank you for your responses, I do apologize for the inconvenience. So all I have to do is modify /etc/make.conf by adding mysql and emerge libdbi. I do not need to install mysql server, this machine only needs mysqlclient. Thanks again for your help. Nick You may have to add 'perl' to make.conf. Just --pretend your emerge and have a look... good luck! -- Michael Higgins
Re: [gentoo-user] trying to track down broken dependency
On Fri, 18 Sep 2009 18:49:39 -0700 kashani kashani-l...@badapple.net wrote: Paul Hartman wrote: On Fri, Sep 18, 2009 at 5:40 PM, kashani kashani-l...@badapple.net wrote: kash...@www01 ~ $ emerge -pvt bugzilla These are the packages that would be merged, in reverse order: Calculating dependencies... done! emerge: there are no ebuilds to satisfy ~dev-lang/perl-5.10.1. (dependency required by perl-core/Module-Build-0.35 [ebuild]) (dependency required by dev-perl/DateTime-TimeZone-0.98 [ebuild]) (dependency required by www-apps/bugzilla-3.4.1-r1 [ebuild]) (dependency required by bugzilla [argument]) I don't see anything in man emerge that would help me track down the missing dependency. Is there any easy way to do this or do I have to track it down the Modeule-Build dependency tree which looks to be the culprit. kashani What version of bugzilla are you trying to emerge? That perl is newer than any I see in my portage. Are you using an overlay? www-apps/bugzilla-3.4.1-r1 seems to emerge fine and works with perl 5.8 here. Perl 5.8 is at end-of-life. Gentoo volunteers are *very* (PAINFULLY) slow in getting 5.10 into the official tree. They unfortunately need all the help they can get, since this is a major failure of Gentoo to keep up with the upstream developers and (FWIW) other distros. This isn't news, BTW... Please report the error on b.g.o., so the perl herd (or whoever is really doing the work now) can fix the problem. Also, #gentoo-perl is the only place to get any real help on these gentoo-perl issues... (devolution to IRC chat being yet another systemic failure, IMO, but that's the place the folks making these mistakes may communicate with users). Better yet, take the plunge and go on to install 5.10.1 from the perl-experimental overlay (good luck with *that*) and report how you fixed any issues you come across. It's only going to get to be a worse mess unless everyone who is able picks up the slack for these guys. And you will have to upgrade eventually anyway... so why not now? My $.02, not terribly helpful though, I suppose. :( Cheers, -- Michael Higgins
Re: [gentoo-user] 2.6.30 and vmware-server build.log
On Sat, 5 Sep 2009 13:23:32 + Nick Khamis sym...@gmail.com wrote: Hello Everyone, I have been at this for a week, and I am stumped. Trying to emerge vmware-server 1.0.9 using x86 environment and 2.6.30 kernel. I gave http://lmgtfy.com/?q=vmware-server+1.0.9+kernel+2.6.30 Good luck! Be sure to post your edited ebuild to b.g.o. up on amd64 earlier this week please help. Attached is the build.log Thanks In Advanced! Ninus.
Re: [gentoo-user] Gentoo Virtualization
On Sat, 5 Sep 2009 17:21:57 + Nick Khamis sym...@gmail.com wrote: I do not meean to be persistent regarding this problem LOL!!! -- Michael Higgins
Re: [gentoo-user] A little OT; ssmtp: Is this correct?
On Fri, 17 Jul 2009 09:26:24 -0500 Michael Sullivan msulli1...@gmail.com wrote: mailhub=baby.espersunited.com # Where will the mail seem to come from? rewritedomain=r...@camille.espersunited.com # The full hostname # Gentoo bug #47562 # Commenting the following line will force ssmtp to figure # out the hostname itself. # hostname=_HOSTNAME_ This has worked fine for about a week until yesterday when I rebooted. Is your hostname properly configured on the machine you rebooted? Now I'm seeing these in my log file: Jul 17 00:00:46 camille sSMTP[18912]: Set RewriteDomain=r...@camille.espersunited.com is invalid http://www.lmgtfy.com/?q=ssmtp+RewriteDomain+%22is+invalid%22 First hit: Comment by Alessandro Doro (adoroo) - Monday, 21 July 2008, 15:41 GMT-4 Dawid, your configuration is wrong. rewriteDomain should only contain the domain name, without the username part. Which makes some sense, given the name of the config option. But the mail that camille tries to send ends up in dead.letter. What's going on here? If it ends up in dead.letter, it seems it's probably *your own* mail server rejecting it..?? Maybe have a look at the exim server on baby.espersunited.com to see why? Anyway, you probably figured this out already. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] festival usage (possibly OT)
On Tue, 7 Jul 2009 11:23:55 -0500 William Hubbs willi...@gentoo.org wrote: William -- Thanks VERY much for asking here. Seeing the post on -dev, I wanted (as is often the case) to reply, but... :( This is much better: Here, no replies could be taken as a pretty good sign no one cares. ;-) we have several open bugs on app-accessibility/festival and So, if there are not any objections, I am considering removing festival from the main gentoo portage tree. I was trying to use it with Asterisk until I hit a file collision error. So, maybe look for another bug report soon... I just started emerging the latest keyword-masked version, so maybe this is fixed already. Nope. As for espeak... http://asterisk-espeak.sourceforge.net/ ... not for 1.6.x (current) Asterisk. So, I think Festival is still the only option for this. Perhaps it'd be good to know for sure before you decide to mothball, or give up maintenance of the ebuilds. As for upstream development, it looks like they are busy. Is three years really so long not to do a release for something that (apparently) works? Looking at the mailing list archives, I see developers posting this year. Short of an announcement that the project is abandoned, why are you eager to reap it? Developers are totally allowed to ignore distro packagers -- particularly academics in the summertime, no? Are other distros dropping it? So, what I'm trying to figure out is how popular festival is and what it is being used for. If you use festival, can you reply to this thread and give me an idea of how important this package is to you and how often you use it, etc? Anyway, for my purposes and from what I can see, it is simply needed in the tree. And somehow I need to fix this file collision before I can even install it. Thanks again for asking. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Spurious warnings?
On Tue, 30 Jun 2009 19:51:52 +0100 Neil Bothwick n...@digimed.co.uk wrote: On Tue, 30 Jun 2009 10:58:40 -0700, Michael Higgins wrote: How do others know if a warning applies after any system/world update? LOG: postinst The database format has changed between 0.30 and 0.40. You may need to update all your local databases and caches. To do so, please run the following commands: (for each user) $ update-mime-database ~/.local/share/mime/ (as root) # update-mime-database /usr/local/share/mime/ genlop -u pkgname will show you the version that was unmerged and the new version. It's up to you to decide which side of 0.30 and 0.40 they lie. Ah, very cool indeed! Thanks, Neil. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Why did this happen?
On Tue, 30 Jun 2009 22:03:24 +0200 Dirk Heinrichs dirk.heinri...@online.de wrote: Am Dienstag 30 Juni 2009 19:54:23 schrieb Michael Higgins: Can anyone speculate as to why this error? Because it's /sbin/nologin. Indeed. Is the fix to usermod -s /sbin/nologin for each of these? Yes. Or vipw. Very useful tool I never knew about. Thanks. It gets more interesting, since viewing with that tool, the passwd file shows all the shells look right, pointing to /sbin/nologin. So, where did the incorrect values come from, I wonder? I'll file a bug against the ebuild. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] app-accessibility/speech-tools-1.2.96_beta build fails
Detected file collision(s): /usr/bin/dp Searching all installed packages for file collisions... Press Ctrl-C to Stop mail-client/nmh-1.1-r1 /usr/bin/dp Package 'app-accessibility/speech-tools-1.2.96_beta' NOT merged due to file collisions. If necessary, refer to your elog messages for the whole content of the above message. So, it seems both packages install the same file. WTF? Am I dead in the water now? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] Why did this happen?
Can anyone speculate as to why this error? Is the fix to usermod -s /sbin/nologin for each of these? ebuild log for sys-apps/baselayout-2.0.1 on evolone.org WARN: postinst The following users have non-existent shells! apache - /usr/sbin/nologin cron - /usr/sbin/nologin ldap - /usr/sbin/nologin postfix - /usr/sbin/nologin sshd - /usr/sbin/nologin -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] Spurious warnings?
Here's an example of a (possibly) useless log warning. I say useless because I don't know (how would I?) whether this was updated from version 0.30 or not. I get these kinds of messages regularly. Sometimes, I say, right. I did that last week, why is it telling me again?. How do others know if a warning applies after any system/world update? LOG: postinst The database format has changed between 0.30 and 0.40. You may need to update all your local databases and caches. To do so, please run the following commands: (for each user) $ update-mime-database ~/.local/share/mime/ (as root) # update-mime-database /usr/local/share/mime/ -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] do I need this?
LOG: postinst Install =x11-libs/gtk+-2 if you need command gtk-update-icon-cache. So, do I need that command? Am I wrong in thinking this is a pretty silly message provided as such without any context? -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] g-cpan not pulling in the correct depencies
On Wed, 17 Jun 2009 06:07:31 -0700 Grant emailgr...@gmail.com wrote: You do have the (perl-experimental) overlay? Thanks Michael. I don't have that overlay. What would it do for me? It's where other ebuilds wind up if no one puts them in the official tree. So, if there needs to be some customisation of an ebuild, it *may* already be done in the overlay. And g-cpan should find it and use it, so may save some time and hassle. An improved g-cpan? Unfortunately, not... BTW, there is a patch to g-cpan which hasn't made it into the tree. I haven't tried it but... looking at the code, it may help a bit with some blatant errors. Still, I don't think a CPAN 'bundle' is going to fly very well using this code... since I believe a 'bundle' really relies on CPAN doing the work, not a package manager. HTH. -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] g-cpan not pulling in the correct depencies
On Tue, 16 Jun 2009 10:03:38 -0700 Grant emailgr...@gmail.com wrote: I'm trying to use g-cpan to pull in Bundle-InterchangeKitchenSink: http://search.cpan.org/~MIKEH/Bundle-InterchangeKitchenSink/InterchangeKitchenSink.pm but the list of dependencies is way off. Is g-cpan just broken? That's about it, yes. (I think the way perl is handled in Gentoo is in serious need of review.) Of course, what you are trying to install, to be fair, isn't exactly trivial and g-cpan does trivial well enough... usually. So, perhaps you could just install the pieces you need? I don't think portage will ever be able to handle CPAN 'bundles'. Just gcpan -g for each item in the package from a shell script. Maybe it will work once you have an ebuild for each needed module, or you can see which one is just way off... and edit to work. You do have the (perl-experimental) overlay? Good luck. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] How to IPSEC M$oft VPN client setup
On Sun, 17 May 2009 12:07:33 +0100 Mick michaelkintz...@gmail.com wrote: On Sunday 17 May 2009, Mick wrote: Thanks Graham, On Saturday 16 May 2009, Graham Murray wrote: Here are some samples. [8] The more I try to use VPN the more I love SSH! http://bugs.gentoo.org/87920 Mick -- This is a *very* old bug. But it still happens. WTF... I see you linked to a related bug here in the ML, but you didn't file/reopen a bug. (Is there a reason why?) Anyway, it would appear like there is no Gentoo dev-loving on these packages, so maybe it would be a waste... For myself, I have zero desire to understand VPN technology, but I guess that's not an option if the devs aren't active in making sane choices for, and presenting viable options to, the users. :( So can we agree on the combination of packages that are *supposed* to provide this VPN-IPSEC-L2TP function? The only thing vaguely M$FT about this setup is MS-CHAP. And L2TP, perhaps. (At least, in so far as I understand this crap, that's my conclusion.) I have: net-firewall/ipsec-tools net-dialup/xl2tpd net-dialup/ppp --is this needed? I don't have * net-misc/openswan ... since that seems to be an alternative to ipsec-tools (KAME). (Or, vice-versa. I'm totally getting sick of reading about VPN.) Is there some other package that should be needed to make this all work? Do I need ppp at all? Isn't XL2TPD the full replacement? Anyway, since there doesn't appear to be a Gentoo document for this, I'd be totally willing to take up space on the ML until both of us have this working. Here, I begin: . . . /etc/init.d/xl2tpd start * Starting xl2tpd ...[ ok ] May 19 10:25:04 lappy xl2tpd[5179]: setsockopt recvref[22]: Protocol not available May 19 10:25:04 lappy xl2tpd[5179]: This binary does not support kernel L2TP. May 19 10:25:04 lappy xl2tpd[5180]: xl2tpd version xl2tpd-1.2.3 started on lappy PID:5180 May 19 10:25:04 lappy xl2tpd[5180]: Written by Mark Spencer, Copyright (C) 1998, Adtran, Inc. May 19 10:25:04 lappy xl2tpd[5180]: Forked by Scott Balmos and David Stipp, (C) 2001 May 19 10:25:04 lappy xl2tpd[5180]: Inherited by Jeff McAdams, (C) 2002 May 19 10:25:04 lappy xl2tpd[5180]: Forked again by Xelerance (www.xelerance.com) (C) 2006 May 19 10:25:04 lappy xl2tpd[5180]: Listening on IP address 0.0.0.0, port 1701 So far, there are no errors. (The warning about *kernel* L2TP is a warning, so I understand, not a failure.) /etc/init.d/racoon start * Loading ipsec policies from /etc/ipsec.conf. * Starting racoon ...[ ok ] May 19 10:27:11 lappy hald [ loads additional crypt modules ] Module Size Used by twofish 5568 0 twofish_common 12672 1 twofish serpent15936 0 blowfish7104 0 sha256_generic 10240 0 May 19 10:27:12 lappy racoon: INFO: @(#)ipsec-tools 0.7.2 (http://ipsec-tools.sourceforge.net) May 19 10:27:12 lappy racoon: INFO: @(#)This product linked OpenSSL 0.9.8k 25 Mar 2009 (http://www.openssl.org/) May 19 10:27:12 lappy racoon: INFO: Reading configuration from /etc/racoon/racoon.conf May 19 10:27:12 lappy racoon: DEBUG: call pfkey_send_register for AH May 19 10:27:12 lappy racoon: DEBUG: call pfkey_send_register for ESP May 19 10:27:12 lappy racoon: DEBUG: call pfkey_send_register for IPCOMP May 19 10:27:12 lappy racoon: DEBUG: reading config file /etc/racoon/racoon.conf May 19 10:27:12 lappy racoon: DEBUG2: lifetime = 3600 May 19 10:27:12 lappy racoon: DEBUG2: lifebyte = 0 May 19 10:27:12 lappy racoon: DEBUG2: encklen=0 May 19 10:27:12 lappy racoon: DEBUG2: p:1 t:1 May 19 10:27:12 lappy racoon: DEBUG2: 3DES-CBC(5) May 19 10:27:12 lappy racoon: DEBUG2: SHA(2) May 19 10:27:12 lappy racoon: DEBUG2: 1024-bit MODP group(2) May 19 10:27:12 lappy racoon: DEBUG2: pre-shared key(1) May 19 10:27:12 lappy racoon: DEBUG2: May 19 10:27:12 lappy racoon: DEBUG: compression algorithm can not be checked because sadb message doesn't support it. [ And there is only 'deflate' available anyway... ?? ] May 19 10:27:12 lappy racoon: DEBUG: getsainfo params: loc='ANONYMOUS', rmt='ANONYMOUS', peer='NULL', id=0 May 19 10:27:12 lappy racoon: DEBUG: getsainfo pass #2 May 19 10:27:12 lappy racoon: DEBUG2: parse successed. May 19 10:27:12 lappy racoon: DEBUG: open /var/lib/racoon/racoon.sock as racoon management. May 19 10:27:12 lappy racoon: DEBUG: my interface: 192.168.1.100 (wlan0) May 19 10:27:12 lappy racoon: DEBUG: my interface: 127.0.0.1 (lo) May 19 10:27:12 lappy racoon: DEBUG: configuring default isakmp port. May 19 10:27:12 lappy racoon: NOTIFY: NAT-T is enabled, autoconfiguring ports May 19 10:27:12 lappy racoon: DEBUG: 4 addrs are configured successfully May 19 10:27:12 lappy racoon: INFO: 127.0.0.1[500] used as isakmp port (fd=7) May 19 10:27:12 lappy racoon: INFO: 127.0.0.1[500] used for NAT-T May 19 10:27:12 lappy
Re: [gentoo-user] How to IPSEC M$oft VPN client setup
On Tue, 19 May 2009 13:57:21 -0500 Paul Hartman paul.hartman+gen...@gmail.com wrote: Based on a brief googling I didn't see anyone who has a working connection to a Checkpoint VPN. Thanks, Paul. I've already the solution, as I'm not so much trying to get something accomplished (access machines inside which I can do just fine with SSH tunnel), as to figure out why we have these various, related, open source software packages available but no basic client-to-corporate real-world implementations specifically outlined for the Gentoo community -- that I can find. :( Just a definitive answer to which Gentoo packages and USE flags to I need to emerge so to do this? .. would be a HUGE help (as weeks later I *still* don't know for sure). And if 60%+ of the folks following it got lucky with cut-n-paste from a how-to, then... great! Say if all the related items were configured, tested and ultimately failed, if documented publicly it'd at the least serve as a good template for anyone else trying to troubleshoot a VPN connection when using Gentoo on a client machine. Or, should I instead, just go outside and play? I thought someone else here had hoped to make something like this work... ;-) Anyway, thanks again for taking a look. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] How to IPSEC M$oft VPN client setup
On Tue, 19 May 2009 22:08:10 +0100 Mick michaelkintz...@gmail.com wrote: On Tuesday 19 May 2009, Michael Higgins wrote: On Tue, 19 May 2009 13:57:21 -0500 Paul Hartman paul.hartman+gen...@gmail.com wrote: Based on a brief googling I didn't see anyone who has a working connection to a Checkpoint VPN. Thanks, Paul. I've already the solution, as I'm not so much trying to get something accomplished (access machines inside which I can do just fine with SSH tunnel), as to figure out why we have these various, related, open source software packages available but no basic client-to-corporate real-world implementations specifically outlined for the Gentoo community -- that I can find. :( [...] Or, should I instead, just go outside and play? I thought someone else here had hoped to make something like this work... ;-) I very much share your frustration. On and off (OK, mostly off) I have been trying to get a VPN connection to my router going, and have tried vnpc, kvpn and racoon all of which failed. Meanwhile, a friend tried the shrew VPN client and succeeded after a couple of hours of tweaking his Vista box! Arrrgh! Yeah, I have no problem to get to working, with XP on VMWare. Naturally, I haven't given up. Seems like it's nearly there... also, there are some examples and docs installed. I assume that I have all the right components installed (judging from the wiki pages) Wiki pages? Hmm. Which ones? but I am not sure about my configuration. Unlike your set up which seems to be almost there, mine won't even complete stage 1 handshake. Very, very, very frustrating ... Well, racoon now claims it has started the connexion. It could have been as trivial as a trailing ' ' on my pre-shared secret. Or not... Either way, it's still not working... just a bit closer. racoonctl vc pub.vpn.ip.add VPN connexion established And still nothing useful happens. ping -c 1 192.168.243.140 PING 192.168.243.140 (192.168.243.140) 56(84) bytes of data. --- 192.168.243.140 ping statistics --- 1 packets transmitted, 0 received, 100% packet loss, time 0ms And tons of debug info. Well, it's more than I had, but less than useful. Sorry that I can't be of much help with this. :( No worries. It seems like this really *should* be possible, though. I'll try to post my findings if I get it working. DEBUG: pfkey UPDATE succeeded: ESP/Tunnel pub.vpn.ip.add[0]-192.168.1.100[0] spi=53896550(0x3366566) May 19 16:00:21 lappy racoon: INFO: IPsec-SA established: ESP/Tunnel 198.145.243.130[0]-192.168.1.100[0] spi=53896550(0x3366566) May 19 16:00:21 lappy racoon: phase2(quick): 0.337284 May 19 16:00:21 lappy racoon: DEBUG: === May 19 16:00:21 lappy racoon: DEBUG: pk_recv: retry[0] recv() May 19 16:00:21 lappy racoon: DEBUG: get pfkey ADD message May 19 16:00:21 lappy racoon: INFO: IPsec-SA established: ESP/Tunnel 192.168.1.100[4500]-pub.vpn.ip.add[4500] spi=1021286747(0x3cdf995b) Not much showing for the failure to communicate, though. :( Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] How to IPSEC M$oft VPN client setup
On Tue, 05 May 2009 17:49:06 +0100 Graham Murray gra...@gmurray.org.uk wrote: Michael Higgins li...@evolone.org writes: Is there a useful Gentoo document anyone might suggest describing how one *connects to* a VPN device of the 'Microsoft' flavour with IPSEC? I do not know about a Gentoo document, I've been working on this for *way* too long, with no apparent success. I have racoon and l2tpt running, but no network addresses in the VPN. Does anyone understand the actual procedure(s) for making a VPN like, l2tp, IPSEC pre-shared secret connection, and wish to elaborate just a bit on the issues (config files, possible values) involved? I mean, the ebuild for ipsec-tools doesn't even put in half the config files... as if any of this could work at all without them? Any help appreciated. :( Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Cursor keys stopped working in VMware workstation
On Fri, 8 May 2009 09:46:55 +0800 Mike Mazur mma...@gmail.com wrote: But since you're seeing this issue too with existing VMs, perhaps it's related to the recent Xorg upgrade? I guess, yes. Happened like this for me too. Found the same fix. -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] How to IPSEC M$oft VPN client setup
Is there a useful Gentoo document anyone might suggest describing how one *connects to* a VPN device of the 'Microsoft' flavour with IPSEC? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] ebuild log for net-print/cups-1.3.10-r1
I _remember_ doing this just a few days ago, so I'm not doing it again... but why is it still throwing a warning? Is this a limitation of portage or the ebuild? -- WARN: postinst /usr/lib/cups exists - You need to remerge every ebuild that installed into /usr/lib/cups and /etc/cups, qfile is in portage-utils: # FEATURES=-collision-protect emerge -va1 $(qfile -qC /usr/lib/cups /etc/cups | sed s:net-print/cups$::) FEATURES=-collision-protect is needed to overwrite the compatibility symlinks installed by this package, it won't be needed on later merges. You should also run revdep-rebuild -- Did they forget to tell me what to do so that the warning stops? '-) Is it possible that it's necessary to re-emerge all those packages on every cups *ebuild* update? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] telephony
On Thu, 23 Apr 2009 19:08:48 +0100 Stroller strol...@stellar.eclipse.co.uk wrote: [...] If you want Asterisk to answer your conventional POTS phone line then you can use an X100P card which you can buy for c £17. AIUI this is basically a modem based on a certain chipset that Digium have written drivers for. They have unequivocally dropped support for these cheap cards. They suck anyway, but this isn't to say you can't play with one I still do, after all. At one time Digium sold this hardware at quite a premium, but people realised that other models would work just as well, and Asterisk (who are sponsored by / part of Digium) has been very fair about supporting these clones in the codebase. Indeed, mine is a clone, and they are equally unconcerned about my problems with it. '-) They're obviously not supported if you buy an official support package, and IIRC I have seen posters on the Asterisk mailing list being snobby and refusing to help posters using the clones because it's not supporting the developers. This may be true, but I believe it's more because the cards, as every one will tell you straight up (unless they are selling you the card, of course) are of poor quality and design. I don't know how well the X100P works, or if there are any gotyas to look out for, but I'm pretty sure plenty of people are using them. Yep. There are driver issues, voltage/signalling problems... and in the end, even if working, they won't sound good. There's a reason they are, like, $10 on Ebay. Basically, they are decent winmodems (if such a thing is possible)... that they can be used for telephony is a fluke. A couple of friends of mine (who I considered going into IT consulting with) implemented Asterisk after I mentioned it to them and I'm sure they've used the X100P; I think those lads have deployed Asterisk for customers since. Yep. Definitely a way to get your hands dirty. By the time you figure out what you need to know to get a decent answering machine with your new toy, you can go buy real hardware and make telephony appliances. Meanwhile, anyone likely to be of any real help while you experiment, is doing just that, and has no interest in watching/helping you suffer, from what I've gathered. My time figuring out the first glitch between my card and the (sort of) supporting driver would have been saved/paid for by buying a real FXO/FXS card initially. I didn't do that, but you, or the OP, still can. And, finally, if I want to ever *use* this experiment in the real world, I'll have to replace the X100p with a decent sounding device anyway. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] telephony
On Wed, 22 Apr 2009 12:30:52 -0400 Simon turne...@gmail.com wrote: hi there, i'm looking for suggestions and guidance. [] I tend to tell folks up front to never listen to me or believe anything I say, so here's my... $.02. I've looked into several ideas, but the last one that remains a good 'deal' is to set it all up myself, for free (but lots of work)! I'm thinking on setting up asterisk, [] Don't. Your GF will hate you for wasting your time. I've read those, but I'm sure there are a lot of surprises not covered by these... http://www.voip-info.org/wiki-Asterisk+Linux+Gentoo http://www.gentoo-wiki.info/Asterisk Thanks for any help! Asterisk is unlike most other open source projects I've seen. Free help doesn't exist, really -- there's just too much money to be made replacing proprietary telco hardware for folks to enjoy educing that aha! moment. All would tell you, at your hourly rate, choosing to roll your own asterisk install on Gentoo is probably the most expensive option available to you. Anyway, it can be done. #gentoo-voip are helpful folks, but unless you have a real reason to know asterisk, really, don't bother... it's huge PITA for just getting a phone. Even if you do decide to kill that gnat with a sledgehammer, you're better off with Trixbox, astlinux or something you can pretty much expect some help with. (So I hear.) If all you want is to use a SIP trunk, my local telco monopoly, for example, will rent me a box to plug a phone into for a fixed monthly fee and unlimited calling. Or, you can probably get any Dahdi compat fxs card and plug a phone into your gentoo box. Or just get a SIP phone... plug into your router. Basically, there are so many options, it's bewildering. From my (limited) experience with it, I'd say that Asterisk is really more of a place to create telco appliances from computers, not an application you run with other things as an afterthought, just because you can. But, then again, you already run Gentoo, so you are familiar with needless self-inflicted pain as a pastime. Asterisk may be just for you, too! ;-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] What annoys you?
On Sat, 4 Apr 2009 13:31:35 +0200 Sebastian Günther sam...@guenther-roetgen.de wrote: * Alan McKinnon (alan.mckin...@gmail.com) [04.04.09 09:57]: emerge --lock some-package-some-version I find this suggestion very good, and would like to ask the more experienced participants, if such thing was thought of before. I'm thinking about some options to freeze a system totally, servers would like this, and some options to gradually move back from testing to stable. This is like what I've been working on for myself, keeping the convenience of emerge world for ebuild revisions but not moving most packages to the next version. That was the thing that annoyed me in the last weeks: a simple possibility to say: hold this packageversion until it is back to stable. So far, with masking the next major version in package.mask, all I've gotten is ‑rN ebuilds. It's an ugly mess of scripting, but it did what I intended so far, that some key packages will stay at current version for a while, some keep closer to current offerings (or from an overlay, or whatever) and others won't register at all every time I sync. So I'm suggesting the following new options for emerge: --freeze: hold this package version *and* revision Mask anything '' current installed ebuild version. --hold: hold this package version, but allow revision updates This by masking '=' next version is what I did, but: It'd be helpful to know for this purpose how the versioning of the release corresponds to the ebuild. Like ${package_major_release_version_position} or something. I don't *think* that's part of the spec yet. --hold-til-stable: hold this package, until it hits stable, and then use the stable version. This should be doable. --testing: set the ~x86 keyword for this package and necessary dependencies I think autounmask is for that, no? Never used it myself. Unmasking specific versions is the trick, rather than a blanket unmask for a package. IDK how that's handled. If you move package.keywords you'll be offered a downgrade for anything not stable, grep out the packages (formerly) in package.keywords from emerge -p world to see if they are still unstable. I know that this is done relatively easy for one package, but with sets this can become a really powerful feature. Haven't looked into sets, but it sounds great. --copy-to-local: make a local overlay entry for that package ++ I would gladly help to implement this, but I did not read anything about becoming a dev. I *think* all you need is some way to read the caches and in your favourite language make something that works. '-) For example: https://projects.gentooexperimental.org/eix/browser/trunk/doc/format.txt.in For /var/cache/eix... and /usr/portage/metadata/cache is pretty straightforward... app-portage/portage-utils small and fast portage helper tools written in C qcache action args : search the metadata cache These tools probably could be incorporated into a set of scripts to run after emerge and do things like you want. Since this is important to me as well, I might just look a little further. ;-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] How to freeze my Gentoo system
On Fri, 3 Apr 2009 10:45:46 +0800 Mark David Dumlao madum...@gmail.com wrote: On Thu, Mar 12, 2009 at 4:13 PM, Alan McKinnon alan.mckin...@gmail.com wrote: On Thursday 12 March 2009 10:07:03 Dale wrote: I do understand that getting something stable and working then wanting to keep it that way. I'm just wondering what his mileage may be in the long run. Here's the first significant result with a sync today: These are the packages that would be merged, in reverse order: Calculating dependencies... done! [ebuild U ] app-text/xpdf-3.02-r2 [3.02-r1] USE=-nodrm LINGUAS=-ar -el -he -ja -ko -la -ru -th -tr -zh_CN -zh_TW 0 kB Total: 1 package (1 upgrade), Size of downloads: 0 kB Ahh. ;-) I guess what's important, unless I see some particular reason to upgrade something, would be this: glsa-check -tv affected This system is affected by the following GLSAs: 200808-09 ( OpenLDAP: Denial of Service vulnerability ) 200903-11 ( PyCrypto: Execution of arbitrary code ) for glsa in `glsa-check -t affected` ; do glsa-check -p $glsa ; done This system is affected by the following GLSAs: Checking GLSA 200808-09 The following updates will be performed for this GLSA: net-nds/openldap-2.4.11-r1 (2.3.41) Checking GLSA 200903-11 The following updates will be performed for this GLSA: dev-python/pycrypto-2.0.1-r8 (2.0.1-r6) In the interest of writing really ugly bash scripts: # for glsa in `glsa-check -t affected` ; do equery d $( glsa-check -p $glsa |grep -P '^\s+\w+-\w+/' | perl -pe 's/^\s+(\w+-\w+\/.+)-\d[\d.].+/$1/' ) ; done This system is affected by the following GLSAs: [ Searching for packages depending on net-nds/openldap... ] app-admin/sudo-1.7.0 (ldap? =net-nds/openldap-2.1.30-r1) app-crypt/gnupg-2.0.10 (!static ldap? net-nds/openldap) (ldap? net-nds/openldap) app-emulation/wine-1.1.12 (ldap? net-nds/openldap) dev-db/postgresql-base-8.3.5 (ldap? net-nds/openldap) dev-libs/apr-util-1.3.4 (ldap? =net-nds/openldap-2*) gnome-base/gconf-2.24.0 (ldap? net-nds/openldap) gnome-extra/evolution-data-server-2.24.5-r2 (ldap? =net-nds/openldap-2.0) mail-client/claws-mail-3.7.1 (ldap? =net-nds/openldap-2.0.7) net-firewall/ipsec-tools-0.7.1 (ldap? net-nds/openldap) net-fs/samba-3.0.33 (ldap? net-nds/openldap) net-misc/curl-7.19.4 (ldap? net-nds/openldap) net-misc/openssh-5.1_p1-r2 (ldap? net-nds/openldap) net-misc/openswan-2.4.13-r2 (ldap? net-nds/openldap) net-print/cups-1.3.9-r1 (ldap? net-nds/openldap) www-servers/apache-2.2.10 (ldap? =net-nds/openldap-2*) [ Searching for packages depending on dev-python/pycrypto... ] sys-apps/portage-2.1.6.7 (!build? =dev-python/pycrypto-2.0.1-r6) Looks like I can fix the use flag and clean out ldap if I want to do so, but I'm stuck with pycrypto (or the build use flag): euse -i build global use flags (searching: build) [-] build - !!internal use only!! DO NOT SET THIS FLAG YOURSELF!, used for creating build images and the first half of bootstrapping [make stage1] ... that's pretty clear. '-) I can only imagine what will happen if he forgets that package.mask and then removes it six months later:-) I too, have spent a couple of days wondering what was masking a package before remembering that it was me. And just to see if there's any upside evident: mv /etc/portage/package.mask /etc/portage/package.mask.bak emerge -puDNtv system mv /etc/portage/package.mask.bak /etc/portage/package.mask These are the packages that would be merged, in reverse order: Calculating dependencies... done! [ebuild U ] net-misc/openssh-5.2_p1-r1 [5.1_p1-r2] USE=X pam tcpd -X509 -hpn -kerberos -ldap -libedit -pkcs11% (-selinux) -skey -smartcard -static 993 kB [ebuild U ] sys-devel/gcc-4.3.3-r2 [4.3.2-r3] USE=fortran gtk mudflap nls openmp (-altivec) -bootstrap -build -doc (-fixed-point) -gcj (-hardened) -ip28 -ip32r10k -libffi (-multilib) -multislot (-n32) (-n64) -nocxx -nopie -objc -objc++ -objc-gc -test -vanilla 58,063 kB Total: 2 packages (2 upgrades), Size of downloads: 59,055 kB Hmm. # mv /etc/portage/package.mask /etc/portage/package.mask.bak emerge -puDNtv world mv /etc/portage/package.mask.bak /etc/portage/package.mask These are the packages that would be merged, in reverse order: Calculating dependencies... done! [ebuild U ] dev-java/sun-jre-bin-1.6.0.13 [1.6.0.12] USE=X alsa nsplugin odbc 78,284 kB [0] (... and some perl modules). So, that's ssh, gcc and java I can pass on today... figure I can unmask in a month and update any of these packages, if I feel like it. But, http://bugs.gentoo.org/buglist.cgi?quicksearch=xpdf (search on the one update I took), it looks like there was a good gentoo reason and maybe a good gentoo response. As I understand it, if the maintainer thinks the recent changes/patches are significant, I'll get a -rN for a new ebuild. OTOH, If there's a new version of something I care about tracking new
[gentoo-user] emerge --prune: should it work?
I attempted to emerge 'twinkle', a soft phone, but whoever made the ebuild neglected to include a dependency on KDE libraries. Of course, since I don't have KDE libs, emerge failed. But before failing, the ebuild had pulled in, built, and installed two *new* packages. As these packages were new dependencies only needed by 'twinkle' (which failed to install), I'd expect running emerge --prune immediately afterward to remove these unnecessary packages. But it didn't. Something should, however. Rather than my asserting that --prune is broken, since it apparently does *something* (just not what I'd expect), can someone give me a helpful clue as to what WILL remove these unneeded libraries? '-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] How to freeze my Gentoo system
On Thu, 12 Mar 2009 09:56:47 +0200 Alan McKinnon alan.mckin...@gmail.com wrote: On Wednesday 11 March 2009 22:40:54 Michael Higgins wrote: Don't know the proper term, but I want to stop version updates for a while, yet allow package-rN updates... This doesn't seem to be a built-in feature of portage after a quick scan of the man pages. But I can think of a method to do it the long way round: The atom syntax you want is Well, I finally got a tuit, the round type, and applied to this question. I don't know about ~ or ~ syntax. Not sure if ~ will work in package mask. Maybe in package.unmask, though... which makes sense, to my understanding. To unmask ~anyversionof-1.2.3. But to package.mask? I can't make '~' work. So... what I want, I think, is = in package.mask for the *next* version update, so it doesn't get pulled in anywhere. Here's the command line I used: qatom $(qlist -ICv) | awk '{print =$1/$2-$3}' |\ sort -u | perl -pe 's/^(.+(?:\.|-))(\d+)(?([-_.\da-z]*))$/$1.($2+1)/ge' | \ grep -Pv $( echo $( eix -inc -\* |cut -d -f 2 ) |tr | ) /etc/portage/package.mask Anyway, that is what I wound up with. Unless I'm totally off, it works by removing funky version strings (that hideous regex) and bumping up by a digit whatever is left, skipping anything that is slotted (eix -inc). (We can't mask in a slot, AFAIK, but maybe someone else does know.) I deeply, newly updated system and world, ran the script above and repeated the deep new update to system and world. The list passed, in that I didn't mask any currently installed packages. '-) And today I got no updates to install after syncing the tree. Good so far. Time will tell, of course, if this really works. If there are no ebuilds offered with revision numbers, or new -rc -pre -whatever, I won't be bothered by them, is the plan. That is, if portage doesn't see 1.2.-rc_ as =1.3 -- which is shouldn't, right? ;-) Then one day I'll remove my additions to package.mask and spend a day pulling out my hair (or watching the new, magic portage do it's beautiful thing...) but that's better than possibly doing that several times a month. And this without ignoring any revisions, bug fixes, and the like. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] kacpid eating alot of cpu
On Mon, 30 Mar 2009 10:57:11 +0200 Johan Blåbäck johan.bluecr...@gmail.com wrote: Is acpi suppose to act like this? No. But you probably really knew that. ;-) Can I fix it? I had to update to latest kernels, as bug patch for kacpid hogging CPU with certain laptops and certain configurations is in 2.6.28, finally. (Happens consistently when returning from sleep or when switching out the extra battery for the DVD. I think Dell and some HP machines are/were affected.) IDK if this will help *you*, but if it smells similar... might be worth checking into. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Usernames in ssh attacks
On Thu, 19 Mar 2009 10:43:13 -0500 Paul Hartman paul.hartman+gen...@gmail.com wrote: On Thu, Mar 19, 2009 at 10:36 AM, Johan Blåbäck johan.bluecr...@gmail.com wrote: I've always had usernames when it comes to sshd's log entries in auth.log, like the following: time hostname sshd[5926]: error: PAM: Authentication failure for username from ip-adress Well, I don't use PAM, just key-based authentication only, so I always see only the IP getting rejected since it doesn't even give them a place to try a user/password :) It's just weird that it is refusing a connection from u...@domain rather than simply the IP. I guess they could be trying to ssh u...@myhost.net or something. The one with [U2FsdGVkX19g32YZVKMsQkl+mouWITILOicY4Iq9OQo=] as the username is interesting. I wonder what that's all about. My $.02: perl -MMIME::Base64 -e 'print decode_base64(U2FsdGVkX19g32YZVKMsQkl+mouWITILOicY4Iq9OQo=)' Salted__`�fT�,BI~���!2 :'���9 I'm not expert, so Google led me to OpenSSL's command-line enc utility: echo U2FsdGVkX19g32YZVKMsQkl+mouWITILOicY4Iq9OQo= | openssl enc -d -base64 -a -idea enter idea-cbc decryption password: ... or like that. Seems like an attempt to send user and password together. I suppose if you know what are possible user/pass combos on your system, and can suss the crypt type from the signature (I've no idea if possible), you can see if it's a real hack attempt. It is interesting, I think... but I'm just guessing. ;-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] How to freeze my Gentoo system
On Thu, 12 Mar 2009 09:51:05 + Neil Bothwick n...@digimed.co.uk wrote: On Thu, 12 Mar 2009 10:13:30 +0200, Alan McKinnon wrote: Could he just not sync and call it a day? I suspect this is going to bite him one day tho. We know Gentoo likes to be updated fairly regular. I been around Gentoo for years and I don't think I would want to do this. I'm not sure how much experience the OP has tho. No worries. If I break it, I get to keep the pieces... Michael's been around a while, his name is familiar. He did say he wants -rN updates so I take that to mean he wants bug fixes and security updates but everything else to stay that same and especially no potential ABI/API changes One potential problem is ebuilds disappearing from the portage tree as packages are updated, so it would be worth copying everything he uses (or the whole tree) into an overlay. Thanks to you both for all the suggestions and caveats... I'll report back when I've done the script to populate package.mask with atoms *pre-*pended by '~'. As one of you mentioned, it's not an unreasonable thing to want to freeze a system, but OTOH Gentoo does like regular updating. If something drops from the tree, that's okay... My goal is, for packages I've unmasked for the architecture, that they don't keep being updated to the latest available, but eventually come into concordance with 'stable'. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] How to freeze my Gentoo system
Don't know the proper term, but I want to stop version updates for a while, yet allow package-rN updates... I spent most of the last couple of days killing two bugs that were a serious drag on my laptop, involving kacpid hogging the CPU on a resume, or bay swap, and gnome panel freezing on 7 open windows (a real deal killer). I'd like to spend a few months just using it now that it all works... So with the latest kernel in the tree unmasked (kacpid bug fix) and a couple of patches and ebuilds in my overlay for a pair of unmasked x11 and gnome packages, what is the method to keep this 'world' in a 'set' and 'forgotten' state? '-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Re: how to switch laptop to usb keyboard
On Wed, 4 Mar 2009 00:16:22 +0100 Miernik pub...@public.miernik.name wrote: Michael Higgins li...@evolone.org wrote: What the *keypad* does, after plugging in the usb keyboard, is act as a mouse. I've never seen this... it's kinda cool, actually. Like using an etch-a-sketch. You can turn such behaviour on and off on any keyboard under X with Shift+Ctrl+NumLock. Try it, maybe it will bring back numbers. 1234567890/*-+ THANKS! Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] how to switch laptop to usb keyboard
On Tue, 3 Mar 2009 09:24:38 +0500 Mike Kazantsev mike_kazant...@fraggod.net wrote: On Mon, 2 Mar 2009 18:31:41 -0800 Michael Higgins li...@evolone.org wrote: Still no luck here, as this is new to me. Really, I'm lost on the process... figured it would just work, which it does, except that I don't have number keys... on either keyboard. So, why is *that*? What should I be trying to do next? If xev (x11-apps/xev) shows generated events and a keycodes when you press these keys (on either KB), It doesn't. What the *keypad* does, after plugging in the usb keyboard, is act as a mouse. I've never seen this... it's kinda cool, actually. Like using an etch-a-sketch. This is what the device is: usb 2-1: new low speed USB device using uhci_hcd and address 3 usb 2-1: configuration #1 chosen from 1 choice input: Logitech Logitech USB Keyboard as /class/input/input8 input: USB HID v1.10 Keyboard [Logitech Logitech USB Keyboard] on usb-:00:1d.1-1 Bus 002 Device 003: ID 046d:c315 Logitech, Inc. Some of the bug reports regarding this keyboard have to do with misidentification as a mouse, so maybe that satisfies me enough to let it drop. Maybe at some point in the future it'll just work... Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] how to switch laptop to usb keyboard
Hey, all -- I picked up a logitech usb keyboard to use with my laptop. Plugged it in and it works... almost. No number keys. How to I tell X to use this different device when I plug it in? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] how to switch laptop to usb keyboard
On Mon, 2 Mar 2009 11:42:53 -0600 Paul Hartman paul.hartman+gen...@gmail.com wrote: On Mon, Mar 2, 2009 at 11:34 AM, Michael Higgins li...@evolone.org wrote: Hey, all -- I picked up a logitech usb keyboard to use with my laptop. Plugged it in and it works... almost. No number keys. How to I tell X to use this different device when I plug it in? Cheers, Using HAL and .fdi files you can set up rules for that device ID to treat it as a certain kind of keyboard. Search gentoo forums or google for some examples and more info about it. Still no luck here, as this is new to me. Really, I'm lost on the process... figured it would just work, which it does, except that I don't have number keys... on either keyboard. So, why is *that*? What should I be trying to do next? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] OT -- superuser file manager access to remote via ssh with no root login?
On Wed, 25 Feb 2009 08:50:00 + Stroller strol...@stellar.eclipse.co.uk wrote: On 25 Feb 2009, at 03:42, Mike Kazantsev wrote: ... 3. Since it sounds like you have no need to do it repeatedly, why not open root and do the stuff? Provided you don't have '123' as password. The voice of reason has entered the thread. Indeed. Thanks to all for the helpful replies. In the end, being on a deadline to complete this idiotic task (moving nested shared IMAP folders), I just emerged a comparatively lightweight file manager, 'pcmanfm' (as the server has some X libs on it) and did it via ssh with -Y... possibly the least secure option, but as was suggested, the easiest. '-) Anyway, Gentoo community rocks! Some very clever proposals. Virtual beverages all around... Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] OT -- superuser file manager access to remote via ssh with no root login?
I can't figure this one out. Have disallowed root login, public key auth. Have a bunch of random renaming to do on that machine though, so would like to point and click for a change. Is this possible? No GUI libs on the remote machine... I was thinking sshfs, but since I can't login directly as root, is there some other way? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] is it enchant stable on x86 -- oh, my spell checker, where did it go?
Claws-mail version 3.7.0 Locale: en_US.UTF-8 (charset: UTF-8) Spell checker could not be started. Couldn't initialize en dictionary: (null) Couldn't initialize en speller. eix app-dicts/*en [I] app-dicts/aspell-en Available versions: 0.51.0 0.51.1 6.0.0 Installed versions: 6.0.0(08:11:33 AM 04/15/2008) Homepage:http://aspell.net Description: English (US, British, Canadian) language dictionary for aspell ??? Did someone change an ebuild, or profile or something, without an announcement somewhere? grumble I get the feeling, more and more, that there isn't a rudder, or even common sense. If someone makes changes that, simply put, break systems, shouldn't they make an effort to advertise it beforehand? With a suggested fix? /grumble http://bugs.gentoo.org/show_bug.cgi?id=250115 http://bugs.gentoo.org/show_bug.cgi?id=250573 And even http://www.thewildbeast.co.uk/claws-mail/bugzilla/show_bug.cgi?id=1801 So, maybe I'm not the only one who found a problem. eix app-dicts/*en [I] app-dicts/aspell-en Available versions: 0.51.0 0.51.1 6.0.0 Installed versions: 6.0.0(08:11:33 AM 04/15/2008) Homepage:http://aspell.net Description: English (US, British, Canadian) language dictionary for aspell equery depends app-dicts/aspell-en [ Searching for packages depending on app-dicts/aspell-en... ] app-text/aspell-0.60.5 (!linguas_vi !linguas_uk !linguas_sv !linguas_sr !linguas_sl !linguas_sk !linguas_ru !linguas_ro !linguas_pt !linguas_pl !linguas_no !linguas_nl !linguas_it !linguas_is !linguas_hr !linguas_he !linguas_gl !linguas_ga !linguas_fr !linguas_fo !linguas_fi !linguas_et !linguas_es !linguas_eo !linguas_en !linguas_el !linguas_de !linguas_da !linguas_cy !linguas_cs !linguas_ca !linguas_br !linguas_bg !linguas_be !linguas_af? app-dicts/aspell-en) (linguas_en? app-dicts/aspell-en) app-text/enchant-1.4.2 (aspell? virtual/aspell-dict) [I] app-text/enchant Available versions: 1.2.5 1.3.0 1.4.2 {aspell hunspell zemberek} Installed versions: 1.4.2(07:07:10 PM 01/25/2009)(hunspell -aspell -zemberek) Homepage:http://www.abisource.com/enchant/ Description: Spellchecker wrapping library So, what is a user supposed to do in this case? It appears that the profile, or something, is defaulted to something other than what it did before (aspell -- hunspell maybe?), but this change maybe wasn't propogated to the user who needs to do something, I suppose. At least, not this user... where did the change originate? This was updated yesterday, so I assume this really is the culprit. Anyway, I put it to the list, should this be something we are told about before we have to go searching for the solution to yet another ebuild-related problem? Is there a dev-announce list I should be subscribing to? Why wouldn't those announcements go here, for example? (Ah, the dead horse isn't moving.) If so, where is/was it? Anyway, it would seem the solution is to change my enchant use flags, or build a different set of dictionaries. Shouldn't there be a post-install check, or something? alculating dependencies... done! [ebuild R ] app-text/enchant-1.4.2 USE=hunspell -aspell -zemberek 0 kB Total: 1 package (1 reinstall), Size of downloads: 0 kB Would you like to merge these packages? [Yes/No] n Quitting. lappy ~ # grep enchant /var/lib/portage/world Geez! I was about to pollute world. Hello, WTF app-office/abiword-2.6.4 (spell? =app-text/enchant-1.2) app-text/gtkspell-2.0.14 (=app-text/enchant-1.1.6) mail-client/claws-mail-3.7.0 (spell? =app-text/enchant-1.0.0) So, is abiword spellcheck broken too? Yep. I just rebuild 'enchant', there were no messages that my system would be broken. [ebuild R ] app-office/abiword-2.6.4 USE=gnome spell -debug -xml 0 kB So, spell enabled. Why not working? Anyway, I don't know the best route to proceed. It seems like 'aspell' is just orphaned on my system, since it isn't used. I didn't make that choice to orphan it, or break the spellchecking. Checking again, nowhere is 'aspell' or 'hunspell' specified in my use flags. Thoughts, anyone? Google search causes many hits, no doubt the solution is either to remove 'aspell' and emerge hunspell en dicts, or add 'aspell' use, but since this is a dependency enabled by default somewhere, why doesn't it try to, or prompt to, install a proper dictionary if the default changed by devs? diff enchant-1.4.2.ebuild /usr/portage/app-text/enchant/enchant-1.4.2.ebuild 1c1 # Copyright 1999-2008 Gentoo Foundation --- # Copyright 1999-2009 Gentoo Foundation 3c3 # $Header: /var/cvsroot/gentoo-x86/app-text/enchant/enchant-1.4.2.ebuild,v 1.2 2008/07/20 06:04:56 serkan Exp $ --- # $Header: /var/cvsroot/gentoo-x86/app-text/enchant/enchant-1.4.2.ebuild,v 1.5 2009/01/23 22:37:39 maekke Exp $ 14c14 KEYWORDS=~alpha ~amd64 ~hppa ~ia64 ~ppc
[gentoo-user] perl-5.10.0 [?!?!]
On Fri, 12 Dec 2008 10:44:51 -0800 Michael Higgins li...@evolone.org wrote: On Thu, 11 Dec 2008 12:15:36 -0800 Andrey Falko ma3ox...@gmail.com wrote: It is about catalystframework, which is in the perl-experimental overlay (a misnomer if ever there was one, experimental). It is about perl 5.10.0, which is long overdue for making it into the tree (never mind just into an overlay). You want to upgrade to perl 5.10.0 on an experimental/testing box? Or do you need perl 5.10.0 for a production environment? Its an upgrade. New features. I want it. Others are already using it. Wah! '-) Seriously, 5.8 is being deprecated. Perl6 is what is experimental. Just check what distros have perl 5.10 and which don't. It doesn't look good for Gentoo! [8] Bad form to reply to myself and change the subject, but for anyone who *does* happen to care, a postscript. '-) There doesn't *appear* to be a plan to move perl-5.10.0 into the tree any time soon, but there are plenty of requests on b.g.o. My experience was after fixing the ebuild from the perl-experimental overlay to remove a call to non-existing command-line patch file and just emerging it, perl-cleaner didn't work well for me. Neither does g-cpan, it *seems*... anyway, these are just utilities for managing the perl installation, not *really* needed... So after re-emerging all the modules with some shell scripting, everything strictly perl-ish, catalsytframework included, worked fine. Yeah! OTOH, in the gentoo tree area, it broke emerging perl-tk and inkscape... :( After a lot of googling, I sent related links for reference, two ebuilds and some patches to b.g.o. as now everything on this system installs and plays nicely with perl 5.10.0. I hope it saves someone some time and gets us closer to current perl. If anyone else wants to try it, but runs into these show-stoppers, the (possibly) needed bits are all up there. Worked for me. '-) As far as slotting the perl 5.8 and 5.10 ebuilds, I have no idea if they will. Anyway I think it's traditional to have different versions of perl on a machine, since you just call from /usr/bin/perl and it actually just points to the symlink for the perl you want, so it should be dead easy. But then emerge empties the old perl binary and modules from the 5.8 tree, so who knows... is still the question, I think. -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Perl woes MRO-Compat requires perl-5.10.0 [?!?!]
On Thu, 11 Dec 2008 21:14:26 +0100 Daniel Pielmeier daniel.pielme...@googlemail.com wrote: Michael Higgins schrieb am 11.12.2008 20:54: On Thu, 11 Dec 2008 19:39:34 +0100 Daniel Pielmeier daniel.pielme...@googlemail.com wrote: and why do you blame Gentoo in bug #250632? Because I don't know how to find out where the problem comes from. They at BGO say it is in overlay, resolved the bug invalid. Fine with me. [8] Well the bug was resolved as invalid as the ebuild is from an overlay (although Gentoo developers contribute to that one) and that are not supported. I think you just expressed yourself in the wrong way and you did not provide all the necessary information in your report. When looking at the MRO-Compat perl script it says need 5.006. So I think perl-5.10 is really an unnecessary high dependency. Um, exactly. ;-) The first sign of trouble. Being given via sync an ebuild which blocked updating my system with a *requirement* for 5.10.0 was the first sign of major ebuild problem... And, yes, it was fixed in response to my bug report after all. And as you may have seen that bug has been reopened (comment 4) after my comment and your issue is now resolved (comment 5) by removing this dependency. And, the person *actually responsible* for the wider error regarding suddenly masked packages (where they weren't marched before) emailed me *privately* with explanation. So, one fix via BGO, and one piece of private information for me to quiet my mind. All good! Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Perl woes MRO-Compat requires perl-5.10.0 [?!?!]
On Thu, 11 Dec 2008 12:15:36 -0800 Andrey Falko ma3ox...@gmail.com wrote: It is about catalystframework, which is in the perl-experimental overlay (a misnomer if ever there was one, experimental). It is about perl 5.10.0, which is long overdue for making it into the tree (never mind just into an overlay). You want to upgrade to perl 5.10.0 on an experimental/testing box? Or do you need perl 5.10.0 for a production environment? Its an upgrade. New features. I want it. Others are already using it. Wah! '-) Seriously, 5.8 is being deprecated. Perl6 is what is experimental. Just check what distros have perl 5.10 and which don't. It doesn't look good for Gentoo! Also, it is about the (likely) painful migration that will come and how to get informed now. If there is a migration to perl 5.10.0 in the future, it will be as smooth as possible for users, especially those who stick to all non-~x86 or custom packages. I hate to even reply to this, but install catalystframework (which isn't experimental in the least, BTW) on a non-~arch system and get back to me. '-) And what do you mean by 'if' there is a migration? Is it possible Gentoo will always only offer perl 5.8, and never the current version 5.10? I mean, I can get 5.10 on Windows from Active State! Worst case scenario is that you'll have to follow an official guide that has you run some extra commands to ensure that upgrade goes through without problems for Gentoo users. perl 5.10.0 will not be marked stable until all upgrade issues are What upgrade issues (where's the link to this list)? cleared either via appropriate documentation (official upgrade guide) or via bug fixes elsewhere. What bugs? What documents? (Could you please provide the links to items you reference?) I hate to seem grouchy, but really I'm looking for sources of related information, not a pat on the hand. '-) It seems you have information I have yet to discover, so please provide some documentation if it exists. However, it pretty much confirms, by omission, that there are no plans/actions by the gentoo tree maintainers to address perl 5.10.0, I have to assume, or someone would have mentioned it. (But how to find out for sure?) It seems this still stands absent contradicting evidence, there are no plans/actions by the gentoo tree maintainers to address perl 5.10.0. True, or false? How to know for sure? If you need perl 5.10.0 in a production environment, now or in the future, then you might as well file a bug in bugs.gentoo.org requesting that perl 5.10.0 be put into the tree and stabilized. Yeah, they are already there. Are the bugs filed so far insufficiently illustrating the problem? If there is enough interest by users and developers in getting 5.10.0 into the tree and stable, then you will likely see this bug resolved. If on the other hand you are the only one interested, then unfortunately you'll have to do the work in doing the work. So, unless a certain number of folks file bugs about a version bump, we don't get current version upgrades? That seems shortsighted, at best. I am confused about what you're intentions and needs are. Hopefully I've addressed some of your concerns or worries. Well, as you didn't provide any documentation, nothing is addressed to my satisfaction. I'm not even sure -- are you part of the herd? -- where/how you base your replies... OTOH, what you've asserted, if not in fact baseless, makes me worry even more. :( Thanks, sincerely, for your reply, though. '-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] Perl woes MRO-Compat requires perl-5.10.0 [?!?!]
Hey, folks -- IMO, maintaining a decent perl library isn't difficult in the least. With Gentoo, there is an extra level of confusion when using Portage to manage the perl libraries and modules. Even so, it hasn't been too bad. Until now. I've come across this issue several times, solved it by unmasking everything perl in the trees, but now it's affecting 'real world' boxes. Folks on IRC haven't given any explanation of the problem, or a reason as to why it hasn't been fixed. So, maybe someone on this -user list can shed light? emerge -puDNtv world (Standard invocation, no?) . . . emerge: there are no ebuilds to satisfy =dev-perl/perl-5.10.0. (dependency required by dev-perl/MRO-Compat-0.09 [installed]) Folks on #gentoo-perl suggested upgrading, emerge dev-perl/MRO-Compat-0.09. (No one assigned blame for this dependency on a non-existing package.) D] dev-perl/MRO-Compat [1] Available versions: ~0.05 ~0.07 ~0.09-r1 Installed versions: 0.09(01:15:13 PM 07/27/2008) Homepage: http://search.cpan.org/search?query=MRO-Compatmode=dist Description: Lets you build groups of accessors !!! All ebuilds that could satisfy dev-perl/MRO-Compat-0.09 have been masked. !!! One of the following masked packages is required to complete your request: - dev-perl/MRO-Compat-0.09-r1 (masked by: ~x86 keyword) So... joy is at hand, right? vim package.keywords... And finally, emerge -av dev-perl/MRO-Compat-0.09 These are the packages that would be merged, in order: Calculating dependencies | !!! All ebuilds that could satisfy =dev-lang/perl-5.10.0 have been masked. !!! One of the following masked packages is required to complete your request: - dev-lang/perl-5.10.0 (masked by: package.mask, ~x86 keyword) For more information, see MASKED PACKAGES section in the emerge man page or refer to the Gentoo Handbook. (dependency required by dev-perl/MRO-Compat-0.09-r1 [ebuild]) Well, at least that _does_ exist in the overlay. Either way, it appears I'm forced to upgrade to perl-5.10. Which is, of course, ludicrous. . . . Well, I'll send this. (No help on IRC today.) Maybe someone out in user land has a clue about this mess. I'll file some bug reports too, since that at least leaves an ugly mark until addressed. :( Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Perl woes MRO-Compat requires perl-5.10.0 [?!?!]
On Thu, 11 Dec 2008 19:39:34 +0100 Daniel Pielmeier daniel.pielme...@googlemail.com wrote: Michael Higgins schrieb am 11.12.2008 19:19: [8] dev-perl/MRO-Compat is not in portage, so from which overlay do you have it. The perl overlay (where you get catalystframework ebuilds). This overlay should also contain perl-5.10 which is not in portage too. It does indeed. For testing. So what is this all about It is about catalystframework, which is in the perl-experimental overlay (a misnomer if ever there was one, experimental). It is about perl 5.10.0, which is long overdue for making it into the tree (never mind just into an overlay). Also, it is about the (likely) painful migration that will come and how to get informed now. and why do you blame Gentoo in bug #250632? Because I don't know how to find out where the problem comes from. They at BGO say it is in overlay, resolved the bug invalid. Fine with me. However, it pretty much confirms, by omission, that there are no plans/actions by the gentoo tree maintainers to address perl 5.10.0, I have to assume, or someone would have mentioned it. (But how to find out for sure?) And anyone else who runs into this (I must be the only one, though) will see a resolved bug. Is that a problem..?? Please accept my apologies for the noise. I'll never blame Gentoo again! Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] package masked perl left and right... what now??
After the better part of an hour spent manually unmasking inexplicably newly-masked dependencies of installed packages, I finally: grep dev-perl /usr/portage/profiles/package.mask /etc/portage/package.unmask ... so I could [EMAIL PROTECTED]*()_ get on with an updated system. So, did anyone on this list have forewarning that gentoo's perl maintainability was about to be trashed? Or like, did I miss some new portage feature that would have solved this for me? (I emailed the dev apparently responsible for this borkage directly, but he hasn't replied, yet.) I've often wondered (more and more often now it seems) where I can get a heads up that the devs are about to cause major pain. Is there such a beast? Never do I see an announcement on the -dev list like, hey guys, check it out: next week I'm gonna make perl users maintaining Gentoo servers wish they used python instead, or something useful like this. '-) Any thoughts or insight to this from the list? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] package masked perl left and right... what now??
On Tue, 18 Nov 2008 11:51:06 -0800 Michael Higgins [EMAIL PROTECTED] wrote: To follow up, since I posted to the list, I got a reply from the maintainer who did this, who referred me here: https://bugs.gentoo.org/show_bug.cgi?id=247413#c1 Which also contains the explanation, Some perl modules were moved from dev-perl/ to perl-core/. This took longer than the time between two rsync mirror updates. So you have fetched an inconsistent portage tree. All I can say (beyond WTF) is *wow*, someone did something knowing it would/could/did create havoc, but didn't bother to broadcast it to those it would affect. (I say this knowing that maybe there is some way to find out about stuff before getting burned by it, but I don't know about it if there is such a beast. I suppose I could read every new post at B.G.O.?) At least now I know what happened, but not why it was allowed to happen. I suppose having an inconsistent tree is a risk with any packages maintained by gentoo folks? If so, why doesn't this happen more often? Hmm. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] [OT?] /etc/shadow perms group shadow?
On Sat, 15 Nov 2008 02:01:54 +0100 Michele Schiavo [EMAIL PROTECTED] wrote: /etc/sudoers ?? I think I'm trying to avoid running under sudo. Yes, that works, but must have other security implications? In researching the problem, the workaround I posted was cribbed from other distros which have a 'shadow' group. This is why I posted here, to see if this is common (as I now suspect), why isn't it used in Gentoo? A decision must have been made at some point? Ultimately, the apache:apache user will be running this code. I expect to have to add apache to the group shadow to be able to use the app. I don't want apache in the sudoers file, nor do I think it'd solve the problem, since my user is in the sudoers file but only can access /etc/shadow when running under sudo. I don't see this as a way to launch my webserver..?? I admit I'm flying in the dark, as is usual. I could be totally wrong. Thanks for the input, though! Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] [OT?] /etc/shadow perms group shadow?
On Sat, 15 Nov 2008 06:45:54 + Stroller [EMAIL PROTECTED] wrote: On 15 Nov 2008, at 00:57, Michael Higgins wrote: ... An application runs as a web server. In this application I have hooks to PAM. The results I was getting from attempting to authorize against PAM were fruitless, until I looked at making a way for the user running this to read /etc/shadow. At any rate, I wound up making a group shadow and making /etc/ shadow owned by group shadow and group-readable, adding my user to this group. Now it works great. Isn't this something Gentoo should have a mechanism for handling already, or am I totally off the mark here? Does anyone know if this ability to read /etc/shadow to authenticate on a system is somehow deprecated in favor of something else, or just overlooked in Gentoo land... or what? '-) Isn't this depreciated in favour of PAM? Well, my point was to use PAM. But, it would seem my regular user needs higher privileges for this. I think you want to be looking at why that wasn't working at fixing it. It wasn't working, to all appearances, because my user didn't have permission to read /etc/shadow. I didn't write the PAM hook code, just observing results of trying to use it. '-) What if an administrator wants to install your app on a system where users authenticate against LDAP? They'd use an LDAP hook and probably wouldn't have this problem. '-) Sorry to sound negative, but there must be some books / HOWTOs about PAM which show minimal programming examples. I'd copy one of those and see why it won't work on your system or how your code differs. Ah, as I said, the code with the hook is not mine. I'm just observing the behavior of using the code, and all experiments show that giving the user permission to read /etc/shadow is the fix. Other distros _seem_ to include a group to allow use of PAM by arbitrary users added to this group. Unfortunately, I don't run any other distros so to be able to confirm or deny this. (Was hoping someone else might.) If I can get some feeling as to why Gentoo *doesn't* include this group, it would inform my reply to the maintainer of the PAM hook code. In other words, if the PAM suite was modified at some point to provide access to the needed information without superuser privileges, I'd need to have some references to this fact. I've not found any. :( Rather, it seemed from posts about several other similar problems solved that in other distros a 'shadow' group has been created and the perms to /etc/shadow modified to allow reading by this group. Maybe filing a bug report would get me some Gentoo reasoning for why this isn't in place already, but I wanted to grab a sanity check here first. '-) Thanks! Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] [OT?] /etc/shadow perms group shadow?
I have a question which may or may not be Gentoo-specific, but here goes: An application runs as a web server. In this application I have hooks to PAM. The results I was getting from attempting to authorize against PAM were fruitless, until I looked at making a way for the user running this to read /etc/shadow. At any rate, I wound up making a group shadow and making /etc/shadow owned by group shadow and group-readable, adding my user to this group. Now it works great. Isn't this something Gentoo should have a mechanism for handling already, or am I totally off the mark here? Does anyone know if this ability to read /etc/shadow to authenticate on a system is somehow deprecated in favor of something else, or just overlooked in Gentoo land... or what? '-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] asterisk hardware recommendations?
Anyone have a success story with asterisk to voip phones? I'd like to start with a hardware recommendation for the T1 interface card and or CSU/DSU, if anyone can share. Is Digium, FEX the best way to go? TIA for any help or FMs to RT. '-) -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
[gentoo-user] how touchy is /var really? And how to keep tabs on a new disk?
So, in setting up a huge repository of junk, I mean, important business documents, I nearly ran out of disk space on rootfs. Much of it was living in /var, like half the disk's worth. I'd just dropped a new disk in for /home... to move some Outlook files to IMAP maildir folders. Had I been thinking ahead, I would have partitioned it for /var as well, but I didn't. So, I rsyncd /var to /home/varlink, moved /var to /oldvar, 'soft' linked /var to /home/varlink/var and restarted some services that were less than happy with the change, like the mail servers, mysql. Everything seems to work now. Now, was that a stupid thing to do, or should everything under /var continue to work still, without issues? /home is mounted in fstab, just like rootfs. I'm worried about a reboot not working for some reason . As for the new disk, it is only new to this machine. It's SCSI and sits on its own controller, reiserfs format. I think perhaps it originally was intermittently failing in a higher-speed controller and got swapped out. Is there a way I can keep regularly checking it for performance issues, should any crop up? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] how touchy is /var really? And how to keep tabs on a new disk?
On Tue, 12 Aug 2008 20:18:39 +0200 Volker Armin Hemmann [EMAIL PROTECTED] wrote: On Dienstag, 12. August 2008, Michael Higgins wrote: Now, was that a stupid thing to do, or should everything under /var continue to work still, without issues? afaik it should work. /home is mounted in fstab, just like rootfs. I'm worried about a reboot not working for some reason . it should work. I would probably use bind mounting, but that is a matter of taste. This was what I thought I should do, but wasn't sure how to do it. Makes sense now, seems to have worked well. As for the new disk, it is only new to this machine. It's SCSI and sits on its own controller, reiserfs format. I think perhaps it originally was intermittently failing in a higher-speed controller and got swapped out. Is there a way I can keep regularly checking it for performance issues, should any crop up? smartd Is this in portage? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] how touchy is /var really? And how to keep tabs on a new disk?
On Tue, 12 Aug 2008 12:52:59 -0700 kashani [EMAIL PROTECTED] wrote: expire_logs_days = 7 is your friend. You must be psychic. Thanks for all your helpful suggestions, this one too! Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org
Re: [gentoo-user] Mail on multiple laptops
On Tue, 15 Jul 2008 10:11:08 -0400 (EDT) James [EMAIL PROTECTED] wrote: On Tue, July 15, 2008 9:05 am, Michael Pobega wrote: I've always had just one laptop so I've been a POP user for years now, but I'm finally buying a second laptop for travels (Asus Eee PC) and I was wondering what gentoo-user would recommend as a good way to keep the mail synced on both of my machines. IMAP++ I was looking into IMAP but I can't really figure out how it works, and Google doesn't offer IMAP, and I'd like to keep this e-mail address. Can anyone make any suggestions? I've had some luck with unison. Just sync to a common storage before and after using your laptops. Good luck. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Re: gentoo install on Dell restore partition... live?
On Tue, 15 Jul 2008 08:52:20 +0100 Neil Bothwick [EMAIL PROTECTED] wrote: On Tue, 15 Jul 2008 10:42:20 +0300, Nikos Chantziaras wrote: I assume you meant 'cp -ax' :P I did, it's tar that uses -l for this. rsync -ax / /root/hd/ Well, /root/hd is empty so there's nothing to win with rsync other than it being slower than cp. But it does the job too, anyway. It preserves directory time stamps, which cp does not. It's not slower because the destination is empty so the initial file scan takes no longer than with cp. Oh, and you can stop and restart it. OK Guys, thanks for the tips. I'll be trying later this week, just in case Saturday means a trip to the physical location. :( Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] gentoo install on Dell restore partition... live?
So, I finally got the OK to nuke another Dell winbox. My dream is to: * Partition the restore partition (3.6 GB) for any small linux installation * make that partition active, bootable, with grub installed * reboot and ssh into that linux machine to finish by toasting NTFS partition and making it my gentoo system '/' as per usual... Can this be done without physical access to the machine which is currently running windows XP? I am administrator and get 'there' via RDP. Any suggestions appreciated in advance. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] sanity check: fetchmail exim courier-*
Hi all — First question, where is $MAIL set for the users? It is set now to /var/mail/[user] somehow, or is by default. If so, where is that set? Anyway, I need to change to ~/.maildir. If not set to change, where should I set it? The situation (needing a reality check) is we have an ISP that gives small mailboxes. We need to empty them. I work in many locations and have several of these accounts that can fill quickly. I have been leaving mail on the ISP server and using fetchmail to pull to storage in my local user account ~/.maildir whenever I got a 'quota warning'. If I needed any of these emails after the fact, I could open with nail and forward. ('mail' seems to work if I reset $MAIL var as above...) Next, I thought to make these available via IMAP, so set up exim and courier. I can't get IMAP to work, not sure why (perhaps Q#1 above is related? Duh...). But at this point I can send email via smtp and receive via pop3. SMTP is limited to within the network the pop3 is working from anywhere. And local delivery works too. So, I'll just set up a cron job for my user to run fetchmail on the various accounts so to keep my ISP storage empty and pull the emails via POP3 (until I get IMAP working). Anything obviously awry with this kind of setup? Local users can send via inside the net, world users can reply to email originating on the machine. Setting the email address in the mail client to the ISP mail address means failed emails and replies come back to the user... (and SMTP doesn't also reject for failed reverse DNS lookup)... does this seem all good? '-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] DBD-ODBC installation problem
Folks — I'm having a problem getting a working *-perl/DBD-ODBC. I had one that was working just fine, but un-merged unixODBC in a fit of pique, or something. I re-emerged it, but then DBD-ODBC segfaults. So, getting to debug I: export LD_LIBRARY_PATH=/usr LANG= ODBCHOME=/usr ; FEATURES=nostrip emerge -aDv DBD-ODBC : ... Using ODBC in /usr Umm, this looks like a unixodbc type of driver manager. We expect to find the sql.h, sqlext.h and (which were supplied with unixODBC) in $ODBCHOME/include directory alongside the /usr/lib/libodbc.a /usr/lib/libodbc.so library in $ODBCHOME/lib Warning: LD_LIBRARY_PATH doesn't include /usr Checking if your kit is complete... Looks good Using DBI 1.601 (for perl 5.008008 on i586-linux) installed in /usr/lib/perl5/vendor_perl/5.8.8/i586-linux/auto/DBI/ Using DBI 1.601 (for perl 5.008008 on i586-linux) installed in /usr/lib/perl5/vendor_perl/5.8.8/i586-linux/auto/DBI/ Writing Makefile for DBD::ODBC The DBD::ODBC tests will use these values for the database connection: DBI_DSN=e.g. dbi:ODBC:demo DBI_USER= DBI_PASS= Warning: not all required environment variables are set. [But these are just for the tests which aren't enabled anyway.] . . . ' isql -vv TREX user pass ' works just fine. So, unixODBC is working as it should. perl -MDBI -e 'use DBI; \ my $db=DBI-connect(dbi:mysql:dbtest,user,pass)' ... also works, so problem not with DBI. Anyway, a simple script: perl -MDBI -e 'use DBI; \ my $db=DBI-connect(dbi:ODBC:TREX,_user,XX)' Segmentation fault I've tried different versions no joy. Always, the same result. GDB output: Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0xb7dbd6c0 (LWP 23716)] _SQLGetDiagRec (handleType=2, handle=value optimized out, numRecord=1, szSqlState=0xbfdfee2e , pfNativeError=0xbfdfee34, szErrorMsg=0xbfdfec2e , cbErrorMsgMax=511, pcbErrorMsg=0xbfdfee3a) at error.c:525 525 error.c: No such file or directory. in error.c (gdb) bt' #0 _SQLGetDiagRec (handleType=2, handle=value optimized out, numRecord=1, szSqlState=0xbfdfee2e , pfNativeError=0xbfdfee34, szErrorMsg=0xbfdfec2e , cbErrorMsgMax=511, pcbErrorMsg=0xbfdfee3a) at error.c:525 result = value optimized out errs = (struct _sql_errors *) 0x8280584 p = value optimized out odbc_ver = value optimized out msgprefix = [FreeTDS][SQL Server] #1 0xb7b8cdfb in SQLError (henv=0x827ffd8, hdbc=0x8280568, hstmt=0x0, szSqlState=0xbfdfee2e , pfNativeError=0xbfdfee34, szErrorMsg=0xbfdfec2e , cbErrorMsgMax=511, pcbErrorMsg=0xbfdfee3a) at error.c:590 result = value optimized out errs = (struct _sql_errors *) 0x8280584 handle = (SQLHANDLE) 0x8280568 #2 0xb7c6eb4d in AllODBCErrors () from /usr/lib/perl5/vendor_perl/5.8.8/i586-linux/auto/DBD/ODBC/ODBC.so No symbol table info available. #3 0xb7c75f97 in odbc_db_login6 () from /usr/lib/perl5/vendor_perl/5.8.8/i586-linux/auto/DBD/ODBC/ODBC.so No symbol table info available. #4 0xb7c6e206 in XS_DBD__ODBC__db__login () ---Type return to continue, or q return to quit--- from /usr/lib/perl5/vendor_perl/5.8.8/i586-linux/auto/DBD/ODBC/ODBC.so No symbol table info available. #5 0x080bb04f in Perl_pp_entersub () No symbol table info available. #6 0x080b98c5 in Perl_runops_standard () No symbol table info available. #7 0x08065450 in ?? () No symbol table info available. #8 0x0813cff4 in ?? () No symbol table info available. #9 0x08140428 in PL_reg_oldsavedlen () No symbol table info available. #10 0xbfdffaf8 in ?? () No symbol table info available. #11 0x08066004 in Perl_call_sv () No symbol table info available. Backtrace stopped: frame did not save the PC strace -otracedump perl -MDBI -e 'use DBI; my $db=DBI-connect(dbi:ODBC:TREX,user,pass)' output viewable here: http://evolone.org/dbdodbc_stracedump What do I do next? I have a working installation on another machine. Versions are all the same. And this was working previously, somehow. Can anyone make a clue out of the debug output? Are there any gentoo-type methods to employ in an attempt to fix this problem, like some kind of massive re-emergence? I already re-emerged perl, ran perl-cleaner to rebuild all the modules. This after I re-emerged (in order) FreeTDS, unixODBC, DBD-ODBC. No difference. What next? This can't be impossible, though I did note that this package is proposed for bugday under Randomly selected open bugs. Maybe I can just wait 'till Monday (or do I really need to hang out on IRC on Saturday?) Emerging 'pidgin', just in case. :( emerge --info: Portage 2.1.4.4 (default-linux/x86/2006.1, gcc-4.1.2, glibc-2.6.1-r0, 2.6.23-gentoo-r8 i686) = System uname: 2.6.23-gentoo-r8 i686 AMD Athlon(tm) XP 2400+ Timestamp of tree: Fri, 30 May 2008 02:06:01 + ccache version 2.4 [enabled] app-shells/bash: 3.2_p33 dev-java/java-config: 1.3.7, 2.1.4
Re: [gentoo-user] Install Windows XP on Gentoo Laptop
On Sat, 3 May 2008 10:44:01 +0100 Stroller [EMAIL PROTECTED] wrote: On 2 May 2008, at 19:03, Mark Knecht wrote: On Fri, May 2, 2008 at 10:52 AM, Michael Higgins [EMAIL PROTECTED] wrote: (Saw a similar thread, going the wrong way.) I have a laptop with a spare partition waiting for WinXP, to install from Dell OEM disks that came originally. I would be very careful about installing from OEM disks. My HP OEM disks will actually blow ALL the partitions on the drive away, repartition and reformat the whole drive back to the way it was shipped from the factory. What he said. Better if you can find a regular retail copy of XP. Better if he can find a regular _OEM_ copy of XP. Note that in the case of this HP Vista license it only works with the OEM install. The license is no good with a normal copy of Vista. Right. Same with XP. [8] The OEM license numbers don't work with a retail installation CD vice-versa. So what Michael needs is a Microsoft-branded OEM installation CD. These work with any OEM license number (even if the sticker says Dell or HP on it), as long as the Home / Professional versioning is correct. I would suggest - as long as you live in Sweden - the famous 9-in-1 OEM CD, from your favourite swashbuckling sea-dog. Ar, me hearties! This be perfectly legal because the the sticker on the underside of the the laptop is the license for XP, not the CD itself. Stroller. Thanks to you folks for all the great info. Just happen to have an OEM copy of XP Pro kicking about and a sticker on the bottom of the laptop with a product key for the same. Looks like I'm good to go. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] Install Windows XP on Gentoo Laptop
(Saw a similar thread, going the wrong way.) I have a laptop with a spare partition waiting for WinXP, to install from Dell OEM disks that came originally. Is this possible? I understand XP will overwrite the MBR. So, I'd have to re-install grub that's it? '-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] What overwrites resolv.conf
I received a used laptop a week or so ago, wiped the tinker-toy OS offered with it and proceeded to do the right thing. So far, I have got a machine I can (manually) put to sleep and use on a wireless network. So far, so good. At home, I don't have a wireless AP, but a 50-ft. ethernet cable. When I connect via wireless (at the office, say), then use my machine at home, resolv.conf is toasted, where I use fixed IP and put my DNS servers in there. DHCP is used everywhere else. So, what overwrites it, when, how, and how to stop it? Is there a definitive guide to the syntax of the various config files? Or, BETTER YET, is there anyone who has a smoothly-functioning configuration to switch between wireless DHCP and connected hard-wired net setups and would like to share? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] What overwrites resolv.conf
On Wed, 23 Apr 2008 13:03:55 -0400 Mike Edenfield [EMAIL PROTECTED] wrote: Michael Higgins wrote: So, what overwrites it, when, how, and how to stop it? Is there a definitive guide to the syntax of the various config files? Or, BETTER YET, is there anyone who has a smoothly-functioning configuration to switch between wireless DHCP and connected hard-wired net setups and would like to share? Thanks to all for your replies. I think I've got it now! -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] installing vmware?
On Wed, 2 Apr 2008 19:10:20 -0300 luis jure [EMAIL PROTECTED] wrote: hello list, [8] i found a few pages on the net explaining how to install vmware on gentoo, but i'm not clear about those issues. thanks for any hint. http://archiver.mailfighter.net/gentoo-user/2008/January/1600.html My recent experience. HTH. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] killing gnome light - pathetic cry for help.
On Tue, 26 Feb 2008 13:02:58 + Stroller [EMAIL PROTECTED] wrote: On 26 Feb 2008, at 04:18, Iain Buchanan wrote: On Mon, 2008-02-25 at 13:15 -0800, Michael Higgins wrote: (And I don't want to trash my .gnome* folders in a WAG at a fix. I have customization that will disappear. Again.) Make a new user with default .blah files. Does it happen for the new user? If so, start singling them out one by one... yes it's tedious. Can't he just `rename .gnome .gnome.old .gnome*`? Stroller. All, thanks for the info. I'll have to learn a bit of strace, I guess. No magic bullet, is what I find. Could this -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] killing gnome light - pathetic cry for help.
I use Gnome ['gnome-light'] as my WM. For the past few months (many months) I've had the 'gnome-panel' lock up on me. Nothing is clearly causing this. Rebuilding has not seemed to help. Of course, what to rebuild? Everything? I've adopted the strategy of kill -INT and (next time) kill -HUP (and repeat). This gets me working again 98% of the time. This is slightly better than rebooting every time you have a problem. Slightly. So, of course I have a lot of questions. Anyone know what is going on? Anyone know how I can find out what is going on? Adding a gig and a half of ram didn't change the behaviour. Changing video cards didn't either. (And I don't want to trash my .gnome* folders in a WAG at a fix. I have customization that will disappear. Again.) Finally, I'd like to just get the PID of gnome panel in a one-liner to send to kill, if I can't get this fixed. Can anyone volunteer a gimme only the PID one-liner? [ And why does it always complain about not finding GDM... when was that required? I've never used it, nor wanted it. I want the warnings to go away! ] Any replies appreciated. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] [OT] fetchmail to procmail (or something) to arbitrary dir?
Hello, OT post here, but: I (the office, actually) have this lousy ISP that sells mailboxes limited to 50MB. Whatever, I can't change that just now. I have need to keep all mail in one place... for safekeeping, mostly. 50MB is not enough and I get a quota warning. So, knowing this situation is BS, I thought, what's the quickest and dirtiest way (short of hacking up another perl script) that I can make this just go away, without losing that backup storage? Anyway, I puzzled a bit and decided 'fetchmail' sounds pretty good, pretty much what I want to do here. But, it needs sendmail...?? I don't want a MTA on this box. So, I see 'procmail' is an alternative target. Hmm. I see the process as, getting a quota warning and then running 'fetchmail' as a user. It worked, but not how I want. I got the mail off the server, but now it's in my 'own' .maildir folder. As I will need to set up a dump folder for a bunch of different accounts, this won't do. So, what part did I miss about setting the MAILDIR? For some reason my config selected the 'DEFAULT'. How can I set up multiple procmail targets and choose which one I want based on the .fetchmailrc? (Yeah, I don't want to actually learn procmail rules or anything.) poll pop.lousyfreakinisp.com protocol POP3 user [EMAIL PROTECTED] password job fetchall mda /usr/bin/procmail -d %T Or, is there some other more lightweight brainless set-and-forget way I can approach this? Quota-warning-dumpALLpop3email to local folders one each for six or seven email accounts? Without setting up new users? I don't really even care what the format of the folder is, just that it isn't the multi-gigabyte .pst files everyone else has. Would rcvstore[?] work? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] VM Ware or not?
On Fri, 1 Feb 2008 16:47:28 + Peter Humphrey [EMAIL PROTECTED] wrote: On Friday 01 February 2008 13:21:38 Michael Sullivan wrote: Hos anyone been able to get 6.0.2 to work? Yes. I posted pretty much the whole session to the list. app-emulation/vmware-workstation-6.0.2.59824 is what I installed. (BTW, I started the thread. Perhaps you can find the post where I show what I ran into in your mail storage or some online archive.) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] VM Ware or not?
Hello, folks -- Most days, I work remotely from the 'ndoze net via rdesktop when I have to access it for some reason. But the bandwidth there is getting choked as others follow suit. I'm thinking I'd have a better time of it if I just had my own little toybox -- I mean, Windows XP installation -- at home. Sure, I can dual-boot, but then I lose the whole of my reasons for having a real OS. And I'd put up another machine, but I have enough anxiety over the ones I have to maintain already. Why bring that kind of trouble into your home? So, the quick question is, is VM Ware the best way to go? I think XP is still lurking on a partition here... and I do need to run Internet Extorter, Outlook, etc... pretend I'm in their domain... At any rate, I'm hoping some on the list can give me a leg up on this. My first question is, what does this mean to me: * app-emulation/vmware-workstation Available versions: 4.5.3.19414-r7 5.5.5.56455 ~6.0.1.55017!f ~6.0.2.59824!f ... when I downloaded the evaluation license and a tar.gz, does that satisfy the !f, for example? What do I do with it, then? I do notice the version number matches. And the other question I hate to ask -- I'm pretty sure there is one, but, if there is a Gentoo HOW-TO on this process, where is it and IS IT UP TO DATE??? ;-) Any and all pointers welcome! . . . I just read through a bunch of stuff on-line, found the vmware overlay, and I feel less sure of how to proceed than before writing this. '-) Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] VM Ware or not?
On Tue, 29 Jan 2008 19:49:30 + Neil Bothwick [EMAIL PROTECTED] wrote: On Tue, 29 Jan 2008 11:20:38 -0800, Michael Higgins wrote: So, the quick question is, is VM Ware the best way to go? I think XP is still lurking on a partition here... and I do need to run Internet Extorter, Outlook, etc... pretend I'm in their domain... VMware is a good way to go, whether it's the best is open to debate. While you can run it from a physical partition, trying to use an existing install may cause problems. At the very least you'll need to re-activate is with MS because it is now running on a different computer (the virtual one). * app-emulation/vmware-workstation Available versions: 4.5.3.19414-r7 5.5.5.56455 ~6.0.1.55017!f ~6.0.2.59824!f ... when I downloaded the evaluation license and a tar.gz, does that satisfy the !f, for example? What do I do with it, then? I do notice the version number matches. If you try to emerge it, you will be told which file to download and where to put it. Put the one you have in $DISTDIR and use the -v option with emerge. If it shows a 0KB download, you know you have the right file. So, here's how it went, more or less: emerge -av =app-emulation/vmware-workstation-6.0.2.59824 [ nope... ] echo app-emulation/vmware-workstation ~x86 \ app-emulation/vmware-modules ~x86 \ dev-cpp/libsexymm ~x86 \ x11-libs/libview ~x86 /etc/portage/package.keywords [ so: ] [ebuild N] app-emulation/vmware-modules-1.0.0.17-r1 483 kB [ebuild N] dev-cpp/libsexymm-0.1.9 USE=-debug -doc 341 kB [ebuild N] x11-libs/libview-0.6.1 USE=-debug 274 kB [ebuild N F ] app-emulation/vmware-workstation-6.0.2.59824 213,943 kB Total: 4 packages (4 new), Size of downloads: 215,039 kB Fetch Restriction: 1 package (1 unsatisfied) [...] http://download3.vmware.com/software/wkst/VMware-workstation-6.0.2-59824.i386.tar.gz = `/portdir/portage/distfiles/VMware-workstation-6.0.2-59824.i386.tar.gz' ... * Fetch failed for 'app-emulation/vmware-workstation-6.0.2.59824' [ so I: ] mv dump/downloads/VMware-workstation-6.0.2-59824.i386.tar.gz \ /portdir/portage/distfiles/ # emerge -av =app-emulation/vmware-workstation-6.0.2.59824 Total: 1 package (1 new), Size of downloads: 0 kB [ ... thanks, Neil! ] Fetch Restriction: 1 package ... * checking vmware-any-any-update115.tar.gz ;-) ... [ ok ] [ but I must first: ] * Rebuild dev-cpp/gtkmm with USE=accessibility echo dev-cpp/gtkmm accessibility /etc/portage/package.use emerge -av dev-cpp/gtkmm ... Machine crashes, locks hard... I clean the CPU heatsink fan... :( emerge --resume * Messages for package app-emulation/vmware-workstation-6.0.2.59824: * Fallback PaX marking -m * VMWare allows for the potential of overwriting files as root. Only * give VMWare access to trusted individuals. * Remember, in order to run VMware Workstation, you have to * be in the 'vmware' group. * You need to run * /opt/vmware/workstation/bin/vmware-config.pl * to complete the install. * After configuring, run vmware-workstation to launch * Also note that when you reboot you should run: * /etc/init.d/vmware start * before trying to run vmware. Or you could just add it to * the default runlevel: * rc-update add vmware default * GNU info directory index is up-to-date. Linux lbg2 2.6.23-gentoo-r6 #1 Sun Jan 27 22:25:30 PST 2008 i686 AMD Athlon(tm) XP 2400+ AuthenticAMD GNU/Linux passwd -a myuser vmware Adding user myuser to group vmware Running /opt/vmware/workstation/bin/vmware-config.pl [I know no surely appropriate answer to any of the questions, but: ] * Loading vmblock for drag and drop support ... [ ok ] * Starting VMware services: [ ok ] * Virtual machine monitor [ ok ] * Bridged networking on /dev/vmnet0[ ok ] * DHCP server on /dev/vmnet1 [ ok ] * Host-only networking on /dev/vmnet1 [ ok ] * DHCP server on /dev/vmnet8 [ ok ] * NAT service on /dev/vmnet8 [ ok ] * Host-only networking on /dev/vmnet8 [ ok ] * Virtual ethernet [ ok ] The configuration of VMware Workstation 6.0.2 build-59824 for Linux for this running kernel completed successfully. You can now run VMware Workstation by invoking the following command: /opt/vmware/workstation/bin/vmware. I installed it. It runs. Now what? After configuring, run vmware-workstation to launch # which vmware-workstation which: no vmware-workstation in (/sbin:/bin:/usr/sbin:/usr/bin) updatedb locate vmware-workstation |grep -v -P 'db/pkg|cache|portage' ... Okay, so that was a ruse. Hmm. Yes, I
Re: [gentoo-user] tone generator
On Fri, 25 Jan 2008 10:59:25 -0800 (PST) maxim wexler [EMAIL PROTECTED] wrote: Not sure if that's what you want, but speaker-test from alsa-utils can generate sine waves, pink and white noise. [EMAIL PROTECTED] ~ $ speaker-test -t 2 speaker-test 0.0.8 Playback device is plughw:0,0 Stream parameters are 48000Hz, S16_LE, 1 channels Sine wave rate is 440.Hz But it's silent. Doesn't even say /dev/dsp already in use or something like that. I use alsa and my audio works OK otherwise. More mysteries to unravel. speaker-test -D default -r 44100 -c 2 -f 880 -t sine Worked for me. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] tone generator
On Thu, 24 Jan 2008 20:15:14 -0800 (PST) maxim wexler [EMAIL PROTECTED] wrote: something like this should do it: gcc -D LINUX -include string.h -lm -o tonegen tonegen.c Yup. Thanks Iain. ++ Very handy. ;-) -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] RAM upgrade, kernel and swap
On Tue, 22 Jan 2008 00:44:53 +0100 Hemmann, Volker Armin [EMAIL PROTECTED] wrote: On Montag, 21. Januar 2008, Michael Higgins wrote: So, I just got 2 GB of RAM in the mail. Whoo hoo. Before I pop these in, soliciting any thoughts about the following: zcat /proc/config.gz |grep MEM CONFIG_SHMEM=y # CONFIG_TINY_SHMEM is not set # CONFIG_NOHIGHMEM is not set CONFIG_HIGHMEM4G=y # CONFIG_HIGHMEM64G is not set CONFIG_HIGHMEM=y CONFIG_SELECT_MEMORY_MODEL=y CONFIG_FLATMEM_MANUAL=y # CONFIG_DISCONTIGMEM_MANUAL is not set # CONFIG_SPARSEMEM_MANUAL is not set CONFIG_FLATMEM=y CONFIG_FLAT_NODE_MEM_MAP=y # CONFIG_SPARSEMEM_STATIC is not set CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y # CONFIG_BLK_DEV_UMEM is not set # CONFIG_INPUT_FF_MEMLESS is not set # CONFIG_DEBUG_HIGHMEM is not set CONFIG_HAS_IOMEM=y NameFlags Part Type FS Type [Label] Size (MB) --- --- hda1BootPrimary Linux ext2 256.50 hda2Primary Linux swap / Solaris 1024.46 hda3Primary Linux ext3 21480.44 hda5Logical Linux ReiserFS 28771.84 hda6 Logical Linux ReiserFS 28493.15 IOW, do I need to/should I recompile my kernel or change partition sizes? if you don't plan to try suspend-to-disk. No even if you plan to try suspend-to-disk it might work with a swapfile. So still no. Thanks for the feedback. I figured I'd be covered with the setup as it is, but you never know... unless you know. This thing never sleeps... so no worries there. Anyway, I just popped 'em in: $ free total used free sharedbuffers cached Mem: 2074548 1660481908500 0 6812 77284 -/+ buffers/cache: 819521992596 Swap: 1000432 01000432 [ Nice. I like that last line a lot. ] Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] cron and ssmtp
Hello, folks -- Is anyone using cron (vixie-cron) without a real MTA and getting emails? The test is simple, just echo hello world. I expect this to come as an email to the address I put in MAILTO=. But it doesn't. No dead.letter to be found either. Not having found any examples of folks using ssmtp to mail cron errors, does anyone have an experience to share, or advice (short of installing an MTA)? Or, if this is totally possible, can I get help to eliminate possible config errors by some method? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] cron and ssmtp
On Tue, 22 Jan 2008 15:48:05 + Neil Bothwick [EMAIL PROTECTED] wrote: On Tue, 22 Jan 2008 07:36:46 -0800, Michael Higgins wrote: Is anyone using cron (vixie-cron) without a real MTA and getting emails? The test is simple, just echo hello world. I expect this to come as an email to the address I put in MAILTO=. But it doesn't. It works on several boxes here. Is ssmtp working correctly? Try echo -e To: Me\nSubject: ssmtp test\n\nssmtp test | /usr/sbin/ssmtp youraddress If it fails, check /etc/ssmtp/ssmtp.conf, particularly the mailhub setting. No fail. Worked like a charm... I edit my crontab like sudo crontab -u mykhyggz -e At the top, is MAILTO=[EMAIL PROTECTED] My user belongs to cron group: groups wheel cron users mykhyggz Cron is running: top bn1|grep cron 5068 root 18 0 3072 692 552 S 0.0 0.7 0:00.00 cron 1508 * * *echo hello world should have sent me an email with 'hello world' in the body at 8:15, or so I believe, but didn't. What else can I check? Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] cron and ssmtp
On Tue, 22 Jan 2008 22:37:32 +0100 Jan Seeger [EMAIL PROTECTED] wrote: On Tue, 22 Jan 2008 07:36:46 -0800, Michael Higgins wrote: Is anyone using cron (vixie-cron) without a real MTA and getting emails? The test is simple, just echo hello world. I expect this to come as an email to the address I put in MAILTO=. But it doesn't. snip I am also getting mails from cron with ssmtp. However, I had to twiddle my mail server setup: Since ssmtp does not specify a complete From: (or envelope from) adress, the server refused the mail. I needed to allow messages without a from: adress to receive any mail from cron. Regards Jan Well, to f/u with all of this. I re-emerged vixie-cron and cronbase too... * Portage doesn't enforce proper permissions on already existing * directories (bug 141619). Appropriate permissions are now being set * on //etc/cron.{hourly,daily,weekly,monthly}, * //var/spool/cron/ and //var/spool/cron/lastrun/ * (see bug 182998) IDK if that mattered. Again, I have no problem sending mail from a shell. Looking at world, ssmtp isn't listed. I installed that too, for kicks. But, I've been able to send and receive mail, so that shouldn't matter... In fact, I believe now what I do have is a problem getting cron to do it's thing. At this point, I'm baffled. I started finding this in the logs, however, so something changed: Jan 22 16:04:01 devserv cron[22211]: Cannot make/remove an entry for the specified session Jan 22 16:05:01 devserv cron[22215]: Cannot make/remove an entry for the specified session Jan 22 16:06:01 devserv cron[22216]: Cannot make/remove an entry for the specified session Jan 22 16:07:01 devserv cron[22217]: Cannot make/remove an entry for the specified session Well, I guess now I know it's trying... I'll remove 'pam' from my use flags for this? Re-emerge vixie-cron... ... oh, hello, now it's working??? WTF Important: After upgrading PAM, from any version to any version, you have to restart those services that are using it to avoid internal ABI mismatches. This includes sshd, vixie-cron (and probably any other cron service), mail servers, and in general almost every service that accepts users. Oh, REALLY? So there's a connection... hmm. I recall an upgrade howto and check qfile -o /etc/pam.d/* /etc/pam.d/cron ... but it seems fine... Referring to a working config, I uncommented the following line: session optionalpam_limits.so Re-emerged with pam useflag enabled again. Joy. Thanks for all of your input, everyone... I'll be damned if I can figure out what exactly fixed it, though. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] RAM upgrade, kernel and swap
So, I just got 2 GB of RAM in the mail. Whoo hoo. Before I pop these in, soliciting any thoughts about the following: zcat /proc/config.gz |grep MEM CONFIG_SHMEM=y # CONFIG_TINY_SHMEM is not set # CONFIG_NOHIGHMEM is not set CONFIG_HIGHMEM4G=y # CONFIG_HIGHMEM64G is not set CONFIG_HIGHMEM=y CONFIG_SELECT_MEMORY_MODEL=y CONFIG_FLATMEM_MANUAL=y # CONFIG_DISCONTIGMEM_MANUAL is not set # CONFIG_SPARSEMEM_MANUAL is not set CONFIG_FLATMEM=y CONFIG_FLAT_NODE_MEM_MAP=y # CONFIG_SPARSEMEM_STATIC is not set CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y # CONFIG_BLK_DEV_UMEM is not set # CONFIG_INPUT_FF_MEMLESS is not set # CONFIG_DEBUG_HIGHMEM is not set CONFIG_HAS_IOMEM=y NameFlags Part Type FS Type [Label] Size (MB) -- hda1BootPrimary Linux ext2 256.50 hda2Primary Linux swap / Solaris 1024.46 hda3Primary Linux ext3 21480.44 hda5Logical Linux ReiserFS 28771.84 hda6 Logical Linux ReiserFS 28493.15 IOW, do I need to/should I recompile my kernel or change partition sizes? Any thoughts appreciated. Cheers, -- |\ /|| | ~ ~ | \/ ||---| `|` ? ||ichael | |iggins\^ / michael.higgins[at]evolone[dot]org -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] apache configuration woes
Since upgrading one of my servers, I've been getting the standard HTACCESS password required dialogue. I didn't choose this consciously and I don't want it. The whole machine is not visible to the 'net. Where did this get enabled? I'm desperately needing to use one of the applications on the machine that requires I be able to access the files. Thanks for any quick reply. -- Michael Higgins [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] apache configuration woes
On Fri, 7 Sep 2007 11:15:39 -0700 Michael Higgins [EMAIL PROTECTED] wrote: Since upgrading one of my servers, I've been getting the standard HTACCESS password required dialogue. I didn't choose this consciously and I don't want it. The whole machine is not visible to the 'net. Where did this get enabled? I'm desperately needing to use one of the applications on the machine that requires I be able to access the files. Thanks for any quick reply. Well, removing every and all module references with 'auth*' revealed an error with an .htaccess file -- apparently either newly dropped into my htdocs folder (doubtful) or suddenly enabled by one of the directives. Either way, it seems like this should have been optional in the upgrade, somehow. But I can get to my files, it seems. Cheers, -- Michael Higgins [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] apache can't start after updating expat to 2.0.1
On Wed, 15 Aug 2007 16:28:03 +0800 Shaochun Wang [EMAIL PROTECTED] wrote: After updating to expat-2.0.1, apache2 couldn't be started. It says that /usr/sbin/apache2 can't find shared library libexpat.so.0. I did some searching on the internet and found that I need do the following: revdep-rebuild -X --library=libexpat.so.0 Unfortunately, doing this doesn't get apache2 running again and the same problem still exits. Observing the output of the above command, I notice /usr/lib64/libaprutil-0.so.0.9.12 is broken. This file belongs to apr-util. By executing 'eix apr-util', I find that there are two versions of apr-util installed in my computer. After re-emergeing the apr-util-0.9.12-r1, apache2 started normally. So I think that the problem is accually caused by the -X option of revdep-rebuild command, which emerges the best (currrently is the latest stable) packages available. In this apr-util case, it will emerge apr-util-1.2.8. So for slotted packages, don't use -X option for revdep-rebuild command. Hoping in the future, revdep-rebuild could handles such situation correctly. I wish my experience be useful to you. Thank you, thank you, thank you! That was a nasty one, glad you sleuthed it out. Cheers, -- Michael Higgins [EMAIL PROTECTED] -- [EMAIL PROTECTED] mailing list