Re: Using a second monitor
On Wed, 13 Aug 2014 13:50:37 -0400 "Robert D. Hilliard" wrote: > Reply To: bobhilli...@gmail.com > > I use a second, larger monitor (my eyes are vary bad). > > [...] > > Any assistance in getting Debian to display on the second > monitor will be greatly appreciated. As other have pointed out, OS handles this. Here's how I solve it: The core utility you need is called `xrandr`, which stands for X Rotate And Resize. It's a command line utility that (using RandR X extension) can set up various things from enabling/disabling monitors to setting resolution to setting DPI. If you prefer GUI, there's ARandR, which is a GUI utility for the same task. Although not as rich on features, it is pretty straightforward and easy to use. One particularly cool feature is that once you are OK with the setup, using "File">"Save" command, it generates a simple (basically one-liner) script that can set the same layout again. I guess in your case, you would want to have two scripts: one for layout with two monitors, and one for only the laptop. It's up to you how you run them, you can call them from your .xinitrc (or copy/paste just the lines), have them on desktop or assign a keyboard shortcut for them. HTH, Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140814004535.17382...@hugo.daonet.home
Re: How to find dirs with single item
On Thu, 29 May 2014 14:31:03 -0500
Dennis Wicks wrote:
> Can't quite figure out how to do this.
>
> I'd like to be able to scan a Volume or directory and find
> all directories that have only one item in them. Either
> directory or file.
>
> Any ideas??
This one
* wraps around find utility so that you can easily add
further constraints (those will only apply to the directory
you search, not to the child),
* works in good-old `sh`,
* returns true if found at least 1 (similar to grep),
* does not "slurp" the whole list, i.e. should start
printing matches ASAP instead of waiting for find to finish:
#!/bin/sh
filter_single_item() {
rv=1
while read path;
do
test $(ls "$path" | wc -l) -eq 1 && echo "$path" && rv=0
done
return $rv
}
find "$@" -type d | filter_single_item
Yes it's quite ineffective as it creates at least 3 processes
per each dir it scans, but it depends on your case if it's
problem.
OTOH, if you are sure no dir contains space, you could use
`echo` instead of ls and add -w option to wc. Eventually you
could even replace wc with some trick with variable expansion.
(probably switching to bash for better tricks).
aL.
--
Alois Mahdal
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/20140602034936.18a1d...@hugo.daonet.home
Re: OT: n guilty men: what is n? [Was: Re: FSF condemns partnership between Mozilla and Adobe...]
On Tue, 20 May 2014 11:26:41 -0400 The Wanderer wrote: > [...] > > If that's not (related to) what you're asking, then I'm > afraid I don't understand the question. No, I think you understood; my confusion came from not understanding the text correctly; perhaps because I'm not a native English speaker and perhaps also because of my impatience. (And apparently reading tons of technical material in my life did not make my understanding of other language close to perfect :)) Thank you for clarification, I'll re-read the article once more. aL. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140520182029.5208c...@hugo.daonet.home
OT: n guilty men: what is n? [Was: Re: FSF condemns partnership between Mozilla and Adobe...]
On Tue, 20 May 2014 09:45:00 -0400 Celejar wrote: > > >> On 5/20/14, Celejar wrote: > > > > >>> But this is precisely the problem with some of the > > >>> dogmatic idealists here - by this logic, we should > > >>> abolish criminal justice entirely, as it's virtually > > >>> impossible to guarantee that "no one blameless" will > > >>> ever be "persecuted": > > >>> http://www2.law.ucla.edu/volokh/guilty.htm [...] > If you take the trouble to follow the link I posted above, > you'll see an entire paper - one of the most brilliantly > erudite and funniest things I have ever read - devoted to > that question. I am confused about the meaning of n. He first states that n = (P - 10) / 10; # P being population of Sodom, so n has no particular known weight or meaning: Is it n = 1 if we save 1 innocent for 1 guilty? Is it n = 10 if we save 1 innocent for 10 guilty? That would almost make sense except that it would silently imply P = 110. Then, in the rest of the article, he refers to n but, failing to explain the meaning of it, I don't see any point of reading it. Did I miss something? Thanks, aL. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20140520170321.75d0d...@hugo.daonet.home
Re: images that play nicely with revision control?
On 2014-02-10 13:46, Darac Marjal wrote: Diff'ing images would probably want to involve a viewer that can flick between the two images (as one would do in astronomy) as well as, perhaps, showing the difference in the 'code' (perhaps the only difference is compression). I use Beyond Compare for this and many other things like syncing FTPs, comparing/merging trees, analyzing changes, migrating home folders... Its image comparing mode is especially cool, try it out or see product page[1]. [1]: http://www.scootersoftware.com/moreinfo.php?zz=moreinfo_viewers It's paid, but really worth the money, and has month evaluation period. Thanks. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/1984a367a0260f4a7d4c061a8d93b...@zxcvb.cz
Changing Week start
Hello, regarding locales and my user account, I have kind of strange taste, but I would like to know if it's possible to achieve this (or at least indicate my preference to applications): * As a language of display (e.g. GUI messages), I want to use English * As date format, I want to use Czech standard, e.g. "Út, 31.12.2013 13:45:31" for last day of this year, quarter to 2pm plus 31 seconds * When a week is displayed, I want to see Monday as the first day Of course I never had problem with the first one. Time format can be specified using LC_TIME along with things like paper format and monetary format, although most of the time I just tweak it in the app (since it's usually only xfce4-datetime-plugin where I really care). However, I have yet to find a proper way how to define first day of week. Is this possible? Or is there an ongoing attempt to add this to standard LC_* arsenal as well? Thanks, aL. PS: Although I said that my taste is strange, I suspect this might be valid case for any English-speaking citizen living in a non-English environment: basically you want all in your native language, except for things like dates, timetables and prices, where you want to be consistent with what you see everywhere around you. (Disclaimer: I'm not that case, I'm native Czech, I just prefer English for GUIs and WWW to avoid distraction from incorrect/incomplete translations and to learn English better. And to explore new use cases ;)) -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131213001527.632fa...@hugo.daonet.home
Re: Installing Debian inside Windows 7 utilizing VirtualBox
Hello, On Sun, 24 Nov 2013 19:30:16 +0200 Andrei POPESCU wrote: > On Du, 24 nov 13, 11:45:11, Wally Lepore wrote: > > > > Has anyone successfully mounted a Debian .iso file directly > > to VirtualBox? It's probably clear, bout just to be complete: the mounting part should be definitely same for Debian ISO as any other ISO :) > Yes, I did this in VirtualBox running on Debian, but it > shouldn't be different on Windows. I can't detail the steps > because it seems I uninstalled VirtualBox, but it was quite > easy. I don't have it at hand either, but and now I'm not sure if it was for Virtualbox, KVM manager or both it was not so trivial as e.g. on Vmware Workstation, where it's two clicks in menu and you choose the file. It wanted you to add the file first to some "repository" before being able to use it as image for virtual optical drive. However, it was possible to get there using intuition. On the other hand, managers usually offer a VM creation wizard which can simplify this since it's pretty common. Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131124212707.5ed62...@hugo.daonet.home
Re: Why Debian
On Tue, 19 Nov 2013 14:40:57 +0100 berenger.mo...@neutralite.org wrote: > > So, I think that the problems you had with untrusted packages > can be: > 1) your fault: did you install the key? > 2) mate developer's fault, if they did not provided one. > 3) your package management software's fault. 4) there is an actual ongoing MITM attack. Isn't it ironic, how we people tend to forget about real meaning of own alarms? *Especially* those of us who really understand them? aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131119211321.2210e...@hugo.daonet.home
Re: [OT] P* language web page (Was: Re: P* - New language for web programming)
On Tue, 19 Nov 2013 21:01:22 +0100 Alois Mahdal wrote: > I see you are using Github for the code and issue tracking. > Many projects do this for the documentation as well, and even > for main presentation page or other related projects, as > separate parallel repos. I think this approach makes it way > easier to contribute on various levels. Oh my bad, I haven't checked properly. Now I can see you are already doing that (all in one repo)... Please ignore the above then :) > > Thanks, > aL. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131119210515.6180d...@hugo.daonet.home
Re: [OT] P* language web page (Was: Re: P* - New language for web programming)
On Tue, 19 Nov 2013 00:50:10 +0100 Atle Solbakken wrote: > Den 18. nov. 2013 22:45, skrev Alois Mahdal: > > Anyway, I changed the page and used the good old "900px wide > centered page with 16px font size"-trick, looks better now? I think it's almost perfect. (Note that with us, QA people, you never get better than that :D) With this out of way, some other little things come out. (I think we're getting little bit OT-spammy, the rest of this mail tries to address that as well): * Reference to the Google mailing list does not make it clear how to register there, or give link to Howto page. I'm assuming there is similar interface as for this list? * Cosmetic, but: the header logo is exactly touching the top of the window. http://imgur.com/Yf3RzGn * `/` is automatically redirected to `/cgi-bin/what_is`. I understand that you want to have the page in P* (which is great approach) and it might not yet support proper routing, but this can cause unnecessary pain in future: everyone (including bots) will bookmark this URL so once you change it to a nicer form (either by implementing proper routing or changing structure of the page), everyone will get 404s or you will need to keep this URI and send 3xx's for some time. So if possible, I'd redirect at least `/` in some transparent manner, i.e. serve identical content as for `/cgi-bin/what_is`. * Also I think it's kind of unwritten standard to have the main logo link to the page home, i.e. `/`. * I just randomly spotted a typo at `/cgi-bin/how_to_run` -- s/favourute/favourite/ I see you are using Github for the code and issue tracking. Many projects do this for the documentation as well, and even for main presentation page or other related projects, as separate parallel repos. I think this approach makes it way easier to contribute on various levels. Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131119210122.0fd8b...@hugo.daonet.home
[OT] P* language web page (Was: Re: P* - New language for web programming)
On Mon, 18 Nov 2013 22:55:33 +0100
Atle Solbakken wrote:
> Den 18. nov. 2013 10:06, skrev Sandro CAZZANIGA:
> > You see my point ? Cheers.
>
> Yes, I see your point :)
>
> The manual page you saw was auto-generated docbook pages, so
> it's no P*.
>
> Today, however, I've created a webpage which runs on P* and
> looks nicer:
>
> http://www.p-star.org/
Disclaimer: I'm not a web UI designer (QA is what I do for
living), but you want constructive feedback, so I'll say what I
feel like and what I would do first if it was my page.
I haven't seen the old one, but this one looks OK to me, except
that the fonts are about 2x as big as should be. For example,
the main title in header does not even fit my screen
(1280x800), it cuts at "Prog...".
Ironically, while sometimes I do use "zoom in" feature on
various pages from time to time, on your page I actually had to
zoom out to 50% :)
What I'd do in CSS would be leave out the `body { font-size:
25px }` part, use `em` unit for most sizes on the page. Also
I'd consider adding some margins, at least on sides. It's not
the best reading experience for everyone to use the full width
of the medium (monitor): for example it's more likely to
interfere wit the background (read: the real outer world), and
also need to bounce from left to right puts additional stress
on eyes as well as additional effort just to flip to the
(correct) next line. (That's why magazines use columns.)
(Just look at any nicely designed page, www or real page, they
almost always have fairly wide margins. Header/footer on www
can be exceptions since they are marginal. The mentioned
mojolicio.us is really a very nice example.)
Thanks,
aL.
--
Alois Mahdal
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20131118224547.16949...@hugo.daonet.home
[OT] o2online.de: (Was: Re: Install Google Chrome)
On Mon, 18 Nov 2013 22:30:02 +0100 Ralf Mardorf wrote: > I updated to Firefox 25.0.1 and installed Chromium > 31.0.1650.57. For comparison I also launched Opera 12.16, > Midori 0.5.5, QupZilla 1.4.4 and Firefox ESR 17.0.10. > > Only Firefox 25.0 and 25.0.1 cause an unwanted visual effect > when moving the mouse cursor over the page. So this issue > seems not to be related to other issues I experience with > other browsers too. It seems to be a problem for Firefox > 25.0+ only. Same here, with Opera, Chromium and FF. I guess by that effect you mean that "onmouseover" top-left part of the "panel" gets rendered earlier. So the blame is not on Chromium. That said, I must agree it still is a very terribly designed page, especially the middle "panning to infinity but not actually to infinity", "laid out horizontally but not actually horizontally" part. Just because of the design, I feel like running away and never coming back. Maybe if you're their customer, they would value your feedback...? aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131118221140.3d733...@hugo.daonet.home
Re: Install Google Chrome
On Sun, 17 Nov 2013 20:56:59 -0800 David Guntner wrote: > > Offhand, I'd say install the packages it wants. :-) > > Those are all standard repository packages for Debian; start > with the gconf-service package and work up from there. Using > apt-get or aptitude will help you get them all in place. Is there a reason why one would prefer this method over `apt-get install -f`? I believe that if you use the `-f` way, these dependencies will end up marked as automatically installed, making it easy to fully uninstall the Chrome later on. OTOH, if you install them all manually, APT will not uninstall them automatically for you, if you get fed up with Chrome. Am I mising something? aL. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131118050833.09aad...@hugo.daonet.home
Re: Install Google Chrome
On Sun, 17 Nov 2013 23:40:49 -0500 erosenb...@hygeiabiomedical.com wrote: > Dear List - Dear Ethan, > I have done the following - > > Downloaded chrome from site into /opt > > root@meow:/opt# ls > google-chrome-stable_current_i386.deb First, do you really need the Chrome from the site? Chrome is available in Debian repositories under name Chromium, so you should not need to download manually anything, just simple root@meow:~# apt-get install chromium should do. (And `apt-get purge google-chrome-stable` to undo the half-installed Chrome). OK, if you know what you are doing and *really* need their stable version, continue reading. > now I get > > dpkg: dependency problems prevent configuration of > google-chrome-stable: > [...] > dependency problems - leaving unconfigured This is normal and it's because dpkg does not solve the dependencies for you. That's what APT does. > > What should I do? > So *if* the missing dependencies are in Debian repos, all you should need to do is root@meow:~# apt-ger install -f which will try to fix the broken dependencies. Thanks, aL. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131118045453.7967d...@hugo.daonet.home
Re: who uses dual boot? [was: How to start using a free OS]
On Wed, 13 Nov 2013 00:59:02 + José Silva wrote: > He probably has an OEM SLP license, married to a key in the > original machine bios, which won't work in the VM because it > obviously has a different bios that doesn't have the key. > I think there are ways of overcoming this, inserting the > "key" in the VM bios, but I doubt it's legal. You got me, I don't eve n know which license I have (the story is that I bought a NB with Vista bloat and a downgrade CD, which I've used immediately), but anyway I'm sticking with this setup since * I rarely really execute my Windoof scenario (and since some pretty addictive games are there I feel safer if it's little further than arm's reach :D) * and yes, there may be something fishy so it's not very attractive for me to fancy about. However it's good to know that it's possible so big thanks especially to Scott for your detailed steps, they might come in handy. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131113015233.1c253...@hugo.daonet.home
Re: who uses dual boot? [was: How to start using a free OS]
Hi, On Tue, 12 Nov 2013 08:32:55 -0500 Miles Fidelman wrote: > Ralf Mardorf wrote: > > On Tue, 2013-11-12 at 23:01 +1100, Zenaan Harkness wrote: > >>> Install a Linux and call it Windows 2014 - super > >>> professional special admin edition and this kind of user > >>> will have no issue, call it Linux and they will ask you > >>> to remove it and reinstall Windows again. > >> I agree :) It is painful but I can't say it's not true. However, it might change in the future, and given FLOSS licensing principles, (mostly the FL part), future is on our side. > That's a very interesting point, but I wonder if it's true. > There are real-world reasons to run both windows on linux on > the same machine (personal example: running Linux on my > laptop for development and demonstrations; running Windows > for office applications). In my case, it's: * Debian for browsing, occasional gaming (mostly Flash games, Indie games, old ones in DosBox or even some native Linux games), browsing, movies, music, programming, sometimes a little vector or raster graphics, * Windows for occasional gaming (titles too big for DosBox) or (very scarcely, always hoping for no more) working with hardware like Nokia phones and TomTom GPS). > But, having said that, when one really uses two operating > systems on the same machine, I expect it's more common to run > one under virtualization, so you can run both at the same > time - dual booting is a real pain if one is really USING > both operating systems. I have a 5 year old ThinkPad R500 with 1.8GHz dual core and 4G RAM, and while the setup is still really enough for most of what I do, attempting Win XP in VBox for even the most simple tasks was a royal PITA. An I'm not fan of buying new hardware every 2 years just because it allows for more convenient scenario. (Also one reason to keep the Win in the dual-boot manner is that I have a valid OEM license there, which I believe cannot be migrated to VM.) So given that most of people who have hardware do not have later than that and with hardware virtualization support in CPU, I would not expect that to be the most common case. Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20131112213631.2ec07...@hugo.daonet.home
Massive unsolicited OT being lost? What a shame! -- Was: some OT semi-intellectual stuff
On 2013-10-27 19:41, Dmitrii Kashin wrote: Zenaan Harkness writes: Can anyone tell me where my email went - see below - ?? -- Forwarded message -- From: Zenaan Harkness Date: Sat, 31 Aug 2013 12:56:13 +1000 Subject: Re: What's the easiest and/or simplest part of Linux Kernel? To: debian-user@lists.debian.org Friend, it would be better to give us Message-ID, not the whole message. And, btw, why don't you use archives or NNTP to check it yourself? Zeenan, Being the thread * completely OT to this list, * breaking even more rules of ood behavior by not using proper Subject, * in my humble opinion, being less useful than, pardon me, listening to two mind-masturbating intelectuals, I'm surprised that you have enough courage to spam this list even more, asking for help with your lost unsolicited mail. *** Like if anyone should care. *** Please find other platform for your time-wasting conversations. There are people who want to watch this list *because of its purpose*, because they need help or want to help. Not because they are students of psychology, working on a thesis on narcism. Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/23a63b3434b1c16b9fadcd66ca35b...@zxcvb.cz
Re: Security?
On Wed, 11 Sep 2013 20:31:16 -0400 shawn wilson wrote: > [...]- just > immagine all of the juicy info you can get from that. There's > probably more awesome info there than search terms. And since > they're not encrypted and awesome little udp packets of > goodness, my prism would've been just for dns :) Except that from DNS, it's not so easy to get identity info as from HTTP cookies. So unless * your target is using public IP (who has one? and is crazy enough to use it?)[1] * or the IP is shared by few *and* you are willing to somehow smartly couple DNS request data with HTTP and cookies data (say, based on activity during daytime), which in most cases would be useless I don't think you can get much useful info (other than statistical, of course,) from that. [1]: I just realized that I have one for my VPS, and also have set up OpenVPN for access to sensitive sites from coffee shops, although I almost don't really use it... :) aL. -- Alois Mahdal. I prefer to write up this line every time. Really. Two lines now. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130913004013.2311f...@hugo.daonet.home
Re: PuTTY cannot load *server* font after .putty transfer to Jessie
On Fri, 28 Jun 2013 00:20:55 +0200 Alois Mahdal wrote: > [...] > > I'm surprised by the fact that PuTTY fails with this error > now: > > PuTTY: unable to load font "server:terminus-12" I forgot to make it clear that this does not happen everytime, only when launching PuTTY with an older profile, e.g.: me@here:~$ putty -load 'me@there' PuTTY: unable to load font "server:terminus-12" me@here:~$ aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130702000147.224ac...@hugo.daonet.home
PuTTY cannot load *server* font after .putty transfer to Jessie
Hello good people, just yesterday I went on and installed Jessie on my ThinkPad besides Wheezy and started transferring dotfiles from old home to new one (not that I think it's necessary but when else to clean up?). I'm surprised by the fact that PuTTY fails with this error now: PuTTY: unable to load font "server:terminus-12" I haven't yet installed all fonts I used to have in Wheezy, but aren't the "server:" fonts supposed to be downloaded from server (where I'm sure no relevant upgrade took place)? PuTTY is 0.62-11 in Jessie, was 0.62-9 in Wheezy aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130628002055.091a2...@hugo.daonet.home
Re: INSTALLATION OF DEBIAN IN WINDOWS XP
On Wed, 26 Jun 2013 12:43:21 -0400 Doug wrote: > [...] > Boot your PC on the disk you just burned. Shrink your Windows > partition, but leave at least 25% space at the end for more > Windows programs or files. > [...] If I might add, it's good idea to perform following operations in Windows before the shrinking: * clean up as much space as possible (i.e. do your chores now and remove that huge application you haven't used for over a year) * defragment your disk Having more space and less fragmented disk will make it faster and safer to shrink the partition as well as result in less fragmentation afterwards. HTH, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130628001126.3dc0f...@hugo.daonet.home
Re: Sound But No CD Sound - No Solution But Alternatives Work
Hi, On Mon, 10 Jun 2013 10:02:59 -0400 "Thomas H. George" wrote: > On Tue, Jun 04, 2013 at 09:15:35AM -0400, Thomas H. George > wrote: > > My system is Wheezy. > > ogg123 plays ogg files perfectly with sound. > > workbone plays the cd but there is no sound. > > cdir lists the tracks on the cd. > > > A lot of interesting suggestions were made but nothing solved > the problem. Both workbone and cdplay apparently work but no > sound. Alsamixer has no cd slider. are you sure it's a software problem? Last time I saw a CD-ROM on an older desktop, there was this thin 3-vein cable that needed to be connected from the drive to the sound card/motherboard. Besides the IDE and power cable. Forgetting that, I had the same issue: I could rip and list CD tracks, but could not play the CD directly using e.g. cdplayer.exe. (Yes, it was Windows). But some other players could do it---apparently they were using the IDE interface to access the PCM data, not just triggering drive's native play mode. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130610224015.60993...@hugo.daonet.home
Re: Troubleshooting apt-get -f install
On Fri, 24 May 2013 11:02:06 +0300 Andrei POPESCU wrote: > [...] > > What am I missing? How should I help APT here? Does the > > wiki page need update? > > There is also this note: > > If apt-get is not able to install libqtwebkit4:i386 as > dependency, try with aptitude (If you know what are you > doing) > If this is the missing bit [...] Andrei, you hit the nail right on its head. Most aching part of the problem as I see it is that I don't know what is the missing bit. APT does not make it clear what exactly is the missing bit (or bits) and why it cannot fulfill the dependency (no such package? version too high/low? a conflict?). So no, I don't know what I'm doing, but am feeling adventurous, :) so... > [...] you can also do: > > # apt-get install libqtwebkit4:i386 > # apt-get install -f ...i try that and: me@here:~$ sudo apt-get install libqtwebkit4:i386 Reading package lists... Done Building dependency tree Reading state information... Done You might want to run 'apt-get -f install' to correct these: The following packages have unmet dependencies: libqtwebkit4:i386 : Depends: libgstreamer-plugins-base0.10-0:i386 (>= 0.10.31) but it is not going to be installed Depends: libgstreamer0.10-0:i386 (>= 0.10.31) but it is not going to be installed E: Unmet dependencies. Try 'apt-get -f install' with no packages (or specify a solution). me@here:~$ Again, this is what typically makes me dizzy with APT: "it is not going to be installed"--OK, but *why*? me@here:~$ sudo apt-get install -f Reading package lists... Done Building dependency tree Reading state information... Done Correcting dependencies... Done The following packages will be REMOVED: skype:i386 0 upgraded, 0 newly installed, 1 to remove and 55 not upgraded. 1 not fully installed or removed. After this operation, 39.2 MB disk space will be freed. Do you want to continue [Y/n]? n Abort. me@here:~$ Thanks for your help, aL. P.S.: In case I sounded like criticizing APT: No, I know this is about (most probably) broken Skype package and lack of my skill to communicate with APT (mostly stemming from lack of understanding of its internal [resolver] logic and "opinions"). -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130524174426.72aa7...@hugo.daonet.home
Re: Troubleshooting apt-get -f install
On Thu, 23 May 2013 15:36:45 -0500 Hugo Vanwoerkom wrote: > [...] > Yes, that happened to me also. The "solution" that I > found was to do a new install and install skype this way very > early in the process. YMMV. You don't mean complete OS re-install, do you? Give me a break! I did not need it for years with Debian! :D Thanks anyway... aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130523224611.464a0...@hugo.daonet.home
Troubleshooting apt-get -f install
Hi, I'm following instructions[1] from Debian Wiki to install Skype on Wheezy. [1]: http://wiki.debian.org/skype#Debian_7.0_.22Wheezy.22 In a nutshell: # dpkg --add-architecture i386 # apt-get update # wget -O skype-install.deb http://www.skype.com/...snip # dpkg -i skype-install.deb # apt-get -f install Now at the point when I do `apt-get -f install`, only solution APT offers is to remove skype:i386. While I agree with APT that there are much better options than Skype(tm), I need to communicate with people inside the walled garden, so for now I'm going to stick with Skype. me@here:~$ sudo apt-get -f install Reading package lists... Done Building dependency tree Reading state information... Done Correcting dependencies... Done The following packages will be REMOVED: skype:i386 0 upgraded, 0 newly installed, 1 to remove and 55 not upgraded. 1 not fully installed or removed. After this operation, 39.2 MB disk space will be freed. Do you want to continue [Y/n]? What am I missing? How should I help APT here? Does the wiki page need update? (Note that I actually had Skype installed a while ago but purged it as it was somehow "broken": it was missing from Xfce menus and from PATH. So what I'm doing is actually an attempt to "fix" previous installation.) Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130523221838.354a8...@hugo.daonet.home
Re: Unmounting/Ejecting removable media
Hello, On Fri, 10 May 2013 22:47:01 -0400 Curt Howland wrote: > It seems logical to me that there would be a way to say "sync > immediately" or "do not buffer", so that when the drive was > inactive it could be yanked without danger of corruption. IMHO the core of the problem is not in the caching / buffering / syncing, but in the fact that the OS has no way of knowing *when* exactly you are going to yank the device out. So at the very moment when you are pushing your fingers against the thumb and pulling it out, a writing operation spanning few seconds might start and there is no way of ensuring it would not break the integrity of the filesystem or the files on it. By the act of unmounting, ejecting or however you call it, you are telling OS that you are about to do that and by showing a message "it is safe to disconnect...", OS is telling you that, well, it is safe to yank the thing out as OS has already removed access to it from all programs. So even disabling buffering does not solve the problem completely, it only shrinks the time when your device is in unsafe mode to minimum. Extent to how is this successful heavily depends on your use of the drive: e.g. if you use it for office docs that you edit from thumb, the unsafe time will probably shrink from seconds to miliseconds per minute, but if you are running a bittorrent client, it might not help at all. (It could actually do more harm by shortening the drive lifetime.) So if there is a solution, it's in warning the OS. Here we could fancy about body heat detectors that would not work if you pulled the drive using cable attached string, or light detectors, which would not work in dark, et cetera et cetera... aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130511194725.22cb2...@hugo.daonet.home
Re: Debian 7 with two displays.
Hello Antonio and other fellow Debian users! On 2013-05-09 17:06, Antonio Fernández Pérez wrote: Hi everybody, [...] Anybody can give me ideas? Just my CZK0.02: About t month ago I was solving the same in my new job with Xfce. What I found easiest was ARandR (`aptitude show arandr`). It's GUI, but you can save settings that work best for you to a script and have something like `arandr /home/you/.arandr/rc` run on login. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/d6f2693382de4ebd0bc71bc24651f...@zxcvb.cz
Re: are "print server" and "ssh server" needed?
On Tue, 7 May 2013 12:09:58 -0500 Default User wrote: > > Any opinions? Adding to what others posted, one case when SSH might be useful is when bad bug happens to your GUI desktop that prevents you even from switching to terminal to resolve the situation. In such case (provided that you have access to other machine with SSH client one the same network), using SSH to do the killing can be a life-saver. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130508213956.0f1de...@hugo.daonet.home
Re: Upgrading on encrypted partitions from Squeeze to Wheezy
On 2013-04-04 15:23, alois.mahdal.1-ndm...@zxcvb.cz wrote: [...] The point is that he already did successful `apt-get dist-upgrade` and is happily running Wheezy now. Given that, and above, I'm going to do the upgrade as well. I finally finished upgrade and reboot without problems mounting partitions back---just as my colleague. As Jonathan pointed out, we were most probably running dm-crypt in the first place. Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/244b0f0785044e6a3144be96495c3...@zxcvb.cz
Re: Upgrading on encrypted partitions from Squeeze to Wheezy
...in the meantime... On 2013-04-04 14:09, alois.mahdal.1-ndm...@zxcvb.cz wrote: I have quite fresh installation with Squeeze I forgot to mention that one of things I *did* upgrade was kernel: I have switched to 3.2 from Backports without any pain. Since I was installing to crypto volumes (completely new laptop with empty HDD) for first time, I left all installer's choices intact. And I don't remember what it was. I spoke with my colleague who had pretty the same setup: squeeze, BPO kernel, default settings for encrypted partitions in netinst, and the same laptop model. Only exception is slightly different partition layout (I used separate /home, he used single-partition layout, and that I left out first 49GiB as an empty primary partition. You guess right---we are both new employees getting their boxes ready; both "Debianists") The point is that he already did successful `apt-get dist-upgrade` and is happily running Wheezy now. Given that, and above, I'm going to do the upgrade as well. Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/d69c5d5b687f736145e3a65d5...@zxcvb.cz
Upgrading on encrypted partitions from Squeeze to Wheezy
Hello, I have quite fresh installation with Squeeze here and I want to upgrade it to Wheezy. In [upgrade guide][1], I read: Support for cryptoloop has been dropped from the Linux kernel packages included in Debian 7.0. Existing installations using cryptoloop need to be transitioned to dm-crypt before the upgrade. [1]: http://www.debian.org/releases/wheezy/amd64/release-notes/ch-upgrading.en.html#recovery Since I was installing to crypto volumes (completely new laptop with empty HDD) for first time, I left all installer's choices intact. And I don't remember what it was. I'm assuming that Squeeze might have supported (and offered as default) the newer method. So how do I check whether my partitions are encrypted this or other method? And if I need to do the transfer, will this mean that I need to re-encrypt the whole volume again? Can I avoid/postpone that easily? Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/b3a726a521e7306a3cfba3276e392...@zxcvb.cz
Re: I saem to have lost maildrop's xfilter command
Hi, On Fri, 15 Mar 2013 15:59:36 -0400 "Eric d'Halibut" wrote: > Sorry I don't have much more hard data, but my maildrop > scripts, after years of good service, seem all of a sudden > broken. I have no xfilter command, which is required by my > scripts. I don't know maildrop, so this may not be relevant, but last time I've seen commands "magically disappearing" was when I forgot to add proper *shebang* (#!/usr/...) line. .-) (Took *some* time to realize...) aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130323002335.7ed4a...@hugo.daonet.home
Re: apt - missing trustedkeys.gpg to verify source package
Good time of the day! On Fri, 15 Mar 2013 20:12:30 -0600 Bob Proulx wrote: > [..] > Ignore the message about .gnupg/trustedkeys.gpg. It is a > distraction from the real problem. Oh, wait: If I ignored them, I'd never ask and I'd never have my src packages GPG-checked properly! :-) But yes, maybe apt could be a little more helpful with this message and suggest debian-keyring right away...? (Especially if this .gnupg/trustedkeys.gpg cannot be found directly by search on packages.debian.org) Maybe I'll put that on wishlist... wait, I already did[1]! [1]: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=703364 > $ apt-cache show debian-archive-keyring |grep Size: > Installed-Size: 72 > Size: 26218 > $ apt-cache show debian-keyring | grep Size: > Installed-Size: 34988 > Size: 31071928 Funny that "Installed-Size:" is in KiB[2] while "Size:" is apparently in B. Took me a *while* to realize :) [2]: http://www.debian.org/doc/debian-policy/ch-controlfields.html#s-f-Installed-Size Thank you, Bob! aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130318212037.67012...@hugo.daonet.home
Re: apt - missing trustedkeys.gpg to verify source package
Hi, On Fri, 15 Mar 2013 21:32:40 +0100 Sven Joachim wrote: > > when downloading source packages, I see error regarding > > trustedkeys.gpg: > [..] > Install the debian-keyring package. Thanks, worked! APT does not complain with the same... > Note that you still may get the warning if the .dsc file > has been signed with a key that is no longer in the Debian > keyring. Oh, so APT uses different key(ring) to verify sources than to verify .deb packages? I find it quite confusing as well as the fact that while keys for .deb are present by default, keys for src are not... (Correct me if I'm wrong). Regards, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130316012544.048aa...@hugo.daonet.home
apt - missing trustedkeys.gpg to verify source package
Hello,
when downloading source packages, I see error regarding
trustedkeys.gpg:
$ apt-get source mousepad
[...]
Fetched 454 kB in 0s (1,050 kB/s)
gpgv: keyblock resource `/home/me/.gnupg/trustedkeys.gpg':
file open error gpgv: Signature made Sat 30 Jun 2012
05:49:12 PM CEST using RSA key ID 71EF0BA8 gpgv: Can't
check signature: public key not found dpkg-source: warning:
failed to verify signature on ./mousepad_0.2.16-6.dsc
[...]
What is wrong? I tried to search for the missing file via
packages.debian.org/search but no package seems to contain the
file.
My sources.list contains deb and deb-src for Wheezy and sid,
Pinning being set to 650 and 600. Similarly named file
~/.gnupg/trustdb.gpg ("GPG key trust database version 3" in
`file`'s words) exists.
TIA,
aL.
--
Alois Mahdal
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130315202331.02ec8...@hugo.daonet.home
Kernel stability relation to HW-specific code?
Hello, slightly OT, maybe... (specific to kernel, not Debian). On Wed, 13 Mar 2013 13:03:24 -0300 "Joao Luis Meloni Assirati" wrote: > Squeeze is now more than 2 years old and in this period I > installed and administered dozens of debian computers. I had > the oportunity to see lots of crashes not caused by hardware > failure, almost all of them related to video or wifi drivers, > i.e. still related to hardware. I don't recall to find any > "purely algorithmic" kernel crash or panic in the last few > years. Just curious: How would one draw a line between error "related to hardware" and "purely algorithmic"? I'm no kernel expert, but as I understand it, it's basically a layer between applications and hardware. From that point of view, error in kernel not related to hardware seems to me as a rare (if not impossible) thing... I don't mean to bicker, I'm just wondering if it would make any sense to divide kernel code that is more closely related to hardware (like drivers) and more generic code, in attempt to measure stability of these parts separately. Thanks, aL. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130313183843.5b187...@hugo.daonet.home
OT: Windows, a command-line OS (Was: Re: Not for me.)
On Mon, 4 Mar 2013 22:59:33 +0100 ""Morel Bérenger"" wrote: > Le Lun 4 mars 2013 22:26, Joe a écrit : > > On Mon, 04 Mar 2013 14:54:50 -0600 > > Yaro Kasear wrote: > > Mark is wrong, Windows is also a command-line OS. > > Hum... for windows 9x (and Me, and 3.x) ok, since they are > based on MSDOS, but for NT family, there is no way to run it > without graphical layer. In that family, the commandline is > an option, unlike graphical modes, and unlike linux > distributions, where the graphical stuff is optional. I agree. I have done quite a bit of AV software testing, which in some cases involved a little bit of "OS-breaking". I have never seen NT-like Windows without GUI. About the closest I could get was: * a "desktop" (read: single-colored background) with cmd.exe window, which IIRC you can achieve via one of Safe Mode options (torture F8 when booting) Yes, window just as you know it. So maybe explorer.exe is not running, but the window manager is. * visually, BSOD or bootup scandisk session is in text mode, maybe the latter could be interrupted somehow (?) > Well, I might be wrong, I'm not an admin... but if one here > knows how to remove, say, explorer.exe, I would be very happy > to learn how to [...] Maybe a good start: http://en.wikipedia.org/wiki/Windows_shell_replacement > [...] since it would mean I could replace the > buggy (I stopped at windows XP, it may explain my harsh > words... or not.) graphical stuff with something more stable. Maybe I'm wearing pink glasses, maybe bugs are afraid of me, maybe it's kind of like Stockholm syndrome or maybe I'm just a lucky bastard. But I don't remember having real problems with explorer.exe in past like three years. (Win 7 but formerly XP at work...) (Or maybe my memory does a good editing job for me--Except for occasional gaming I have managed to avoid Win for almost three months now.) Also, about the CLI: I have also written quite bunch of cmd.exe scripts, and my experience is that basically what worked in 2000, worked the same all the way up to Win 8. For Windows CLI syntax, there has been some "evolution" (command.com => cmd.exe) which happened to turn the mess into a big mess. It was positive, though: with command.com, it was only spaghetti. With cmd.exe, it was like spaghetti ON STEROIDS. (Read: stronger than you and agressive to your brain.) Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130304233528.1c29a...@hugo.daonet.home
Re: Auto-emptying of trash.
Hello Sharon, bob and everyone!
On Mon, 4 Mar 2013 02:29:09 +
Sharon Kimble wrote:
> [...]
> 'atime' is not installed as its not in the wheezy repos, and
> when i want to install it apt-get comes back at me saying "E:
> Unable to locate package atime"
That is expected. atime is not a package. atime (UNIX-ish
short for "access time") is property of files/directories that
your filesystem takes care of any time a file is accessed--it
updates this field anytime the file is accessed.
You can review atime, mtime and ctime ("change time") using
mentioned `stat` command:
me@here:~$ stat myfile.txt
File: `myfile.txt'
Size: 6 Blocks: 8 IO Block: 4096
regular file
Device: 805h/2053d Inode: 285632 Links: 1
Access: (0644/-rw-r--r--) Uid: ( 1000/ me) Gid:
( 1000/me)
Access: 2013-03-04 04:47:02.929754566 +0100
Modify: 2013-03-04 04:46:22.457204523 +0100
Change: 2013-03-04 04:46:22.457204523 +0100
Birth: -
me@here:~$
However, there is at least one common exception to this
"anytime". As you might know, in UNIX, file system tree is
very often composed of multiple partitions (e.g. different
for /home than for /usr or for /var), often using different
settings on how exactly kernel behaves to them. One common
option is to tell kernel that for particular filesystem (or
"partition", if you prefer), you prefer *not* to have this
atime update done. (Typically people do that for performance
reasons, e.g. on partitions that are designated for files for
which this information is not important.)
One way of knowing if this option is on (not sure if the most
reliable way) is running `mount` command which, if run without
arguments, shows where each filesystem is currently mounted to
which folder and which options are active. You want to look
for "noatime" option:
me@here:~$ mount # shortened for readability
[...]
/dev/disk/by-uuid/f82f7997-1779-4bff-9f06-912e0019b79b on / type ext4
(rw,relatime,errors=remount-ro,user_xattr,barrier=1,data=ordered)
tmpfs on /run/lock type tmpfs (rw,nosuid,nodev,noexec,relatime,size=5120k)
tmpfs on /run/shm type tmpfs (rw,nosuid,nodev,noexec,relatime,size=2049720k)
/dev/sda9 on /var type ext4
(rw,relatime,user_xattr,acl,barrier=1,data=ordered)
/dev/sda5 on /home type ext4
(rw,relatime,user_xattr,acl,barrier=1,data=ordered)
/dev/sda7 on /mnt/pub type fuseblk (rw,nosuid,nodev,noatime,...)
[...]
me@here:~$
Notice the last two lines. Partition /dev/sda5 is mounted
as /home does not have atime turned off (or "noatime" turned
on :)) Partition /dev/sda7, which I use as /mnt/pub *does* have
it (for whatever reason I had in mind when setting this up).
(For sake of correctness: I was using terms "partition" and
"filesystem" kind of interchangeably. It is not the same,
filesystem is not partition, it's something that "lives" on the
partition.)
> You give me the commands and i'll run them.
I do not recommend running any command without understanding it
first! :)
`rm -rf` yo mentioned in other post is *particularly* dangerous.
One typo and you could irreversibly screw up a LOT in no time!
(Remember that there's no such thing as "undelete" on
Linux/UNIX.)
> ls-l didn't show up any mention of trash/wastebin!
(It's typo, right? (I mean, it's `ls -l`).)
Without saying at least *where* (in what dir) you ran the
command, it is impossible to know whether the output is as
expected. If you simply open your terminal, your shell starts
in $HOME, and since Trash is not in that folder, `ls -l` will
not mention it. Correct.
Thanks,
aL.
--
Alois Mahdal
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130304055106.2612d...@hugo.daonet.home
Re: Removing commited versions from git history
On Thu, 28 Feb 2013 04:15:42 +0100 Alois Mahdal wrote: > On Thu, 28 Feb 2013 02:53:31 + (UTC) > T o n g wrote: > > [...] but keep the rest of its revisions. How can I do that? Me sleepy-brain, I mis-read. I'm not sure if it's possible to do with filter-branch. Anyway, I'm sure that rewrite of whole history and all the implied mess is inevitable. It's in fact a feature of git that certain (current) commit cryptographically guarantees literalness of all previous commits. It's like if you went back in time, kill best friend of Joe's grandpa and come back. You'd find out that Joe's a completely different person (if he even exists at all). aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130228043041.71ffc...@hugo.daonet.home
Re: Removing commited versions from git history
On Thu, 28 Feb 2013 02:53:31 + (UTC) T o n g wrote: > > I want to permanently delete an old revision (or revisions) > of single file from Git. Specifically, I want to delete from > my public git server repo my initial published version of one > single file, but keep the rest of its revisions. How can I do > that? Oh, bummer. I believe it's possible and even described in Pro Git Book. The tool is caled filter-branch and it's kind of a nuke: http://git-scm.com/book/ch6-4.html#The-Nuclear-Option:-filter-branch However, since git uses sha1 of previous object as references to them, any change in any (older than the last) revision is going to have to rebuild all future revisions. Provided that your file is soon enough in the history, you are basically going to rebuild your whole repo! If more people have cloned your repo, their current clones including all their local commits are going to be basically lost. Unless you can 100% ensure that all of them commit, push and *delete* the clones (or export their commits separately as patches) BEFORE you even start rewriting the original, I'd think about creating a copy of it, and pushing it to the server as a *new* repo. Then you can explain people that they need to export their commits (anyway) and commit them again against the new one. In any way, you're probably not going to avoid some kind of a very ugly mess. Good luck, you'll need it. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130228041542.5f216...@hugo.daonet.home
Re: Two copies of E-Mail (Re: I wish to advocate linux)
Hi Miles, hi everybody! On Wed, 27 Feb 2013 19:56:40 -0500 Miles Fidelman wrote: > > Sure... thanks! Still, it's a good thing to train folks to > reply appropriately :-) Holy ravioli! Training a troll? Gimme a break! BTW, I'd prefer good old Delete key. I use it a LOT, but it's still working perfectly. My old Delete key and me... aL. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130228032252.281ee...@hugo.daonet.home
Another pointless flame (Was: Re: I wish to advocate linux)
On Tue, 26 Feb 2013 23:21:41 + Joe wrote: > > Large-scale use by people with no idea of even rudimentary > security is what makes malware viable, and good hardware > support will greatly increase the user base. We do worry > about Ubuntu. Brilliant point, indeed.. However... might I propose continuing this thread to OT list? d-community-offto...@lists.alioth.debian.org http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/d-community-offtopic Helping OT get the right direction by few e-mails (as well as exchanging own ones this way) would be a good idea. But I don't see how this is useful anymore. Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130227011926.62197...@hugo.daonet.home
Re: I wish to advocate linux
On Tue, 26 Feb 2013 15:19:17 -0500 "frank ernest" wrote: > I wish to advocate linux and I have one small problem. There is one big problem with advocating Linux. That there are people here that "advocate" Linux by pushing users but that can do much more harm than good. I understand that often this is meant good and it comes from fascination, but don't let that blind you. Linux distros have made a huge progress in past few years, but for "normal desktop use" there definitely *are* problems that won't be overcome easily. (Mainly talking about availability of certain kinds of software.) In my book, best way of advocating it is simply using it. And yes, spreading the word, but generally only if people ask you themselves. Most of people won't ask in million years. I don't dare to bother them. But that's me... However...: > If I were (theoretically speaking,) to go up to Miss Jones > and say "Why not try linux?" she would respond "Which > distro should I try for there are many? There is a great channel on YouTube about this. It goes from "Mum tries out * Linux" to "Mum tries out " most desktop OSes I've ever heard. http://www.youtube.com/user/OsFirstTimer?feature=watch If your Miss Jones is like this amazing lady in the channels, you will learn a lot from these videos. In any case, definitely worth watching at least the main distros or just for fun (they cover even crazy things like Microsof Bob!) > I'M NOT ASKING FOR A WHATS WRONG WITH THE COMPITITION > DESCUSSION!! Please, no double exclamation marks. And definitely no caps. It's strictly unacceptable in pretty every form of netiquette I've ever seen. -cut- Good luck, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130226222147.078e1...@hugo.daonet.home
Re: OT: Answering your own question (Was: Clamav update failed)
On Mon, 25 Feb 2013 13:58:32 +1300 Chris Bannister wrote: > On Sun, Feb 24, 2013 at 08:14:04PM +0100, Alois Mahdal wrote: > > I think that even if you actually knew the answer before you > > asked, it's still better for community to "pretend" that you > > ask and answer than to keep it all to yourself. > > Are you serious? > Imagine the thousands of posts like, "What does uname -a > produce" etc, etc, etc. debian-user is primarily a support > ML. I suggest wiki.debian.org for a knowledge collection > place. Oh, I was just trying to counter that mood of "probably not a good form so I should not post". However, I think I know what you mean and I must admit that my post must have sounded a little naïve. I definitely do not think that this should be the front line and I'm not using it like that. TL;DR: However, I sometimes (actually more often than not) happen to experience what IIRC is called "rubber ducky" effect. First I have problem with something, then I start composing question, and since I always try to add as much useful *verified* info as possible and call things correct names, I often find myself actually answering the question myself. Typically at the phase of verification, i.e. before I hit "Send". Now if the reason why I had the issue was really only my tired brain, or a stupid typo, or some form of `man`- or Google- blindness, I refrain from sending anything, no matter how much time I spent on it. And I did not mean to encourage the opposite in any way. However, at times reason might not be so foolish, sometimes it's something that is confusing or wrong in available sources. Then, I do not see reason why not to post. Of course having the answer prepared in advance. I was just trying to say that "probably not a good form" in OP's sense is not a good reason by itself. > And there's the rub ... who is going to moderate?? > If you prefer forums, ... I hate traditional forums. For me, StackExchange network and the likes is something completely different. Mostly thanks to the semantic level and the "wiki-style", allowing maintenance by community. > ...there is ask.debian.net How come I did not know about that one? :) Thanks, I'll definitely have a close look. Thanks, aL -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130225063055.06439...@hugo.daonet.home
OT: Answering your own question (Was: Clamav update failed)
On Sun, 24 Feb 2013 17:52:09 + Sharon Kimble wrote: > It seems to be bad form to reply to your own email but the > answer is No, no, no... it's the opposite. Nice of you to share the solution, Sharon! Thank you! I think that even if you actually knew the answer before you asked, it's still better for community to "pretend" that you ask and answer than to keep it all to yourself. Many of us (or even yourself in the future) could face the same issue; and mailing list archive is good place to keep notes. (BTW, as a PoC example this principle is actually encouraged at StackExchange forums; when posting the question, you have an "answer your question" checkbox that adds Answer field to the form so you can answer your own question right away! As long as it's helpful, readable and on-topic, you get the reputation points as a reward (voting system). I don't see why this should not apply to mailing list.) Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130224201404.09a6f...@hugo.daonet.home
OT: How to become a Windows device? (Was: possible USB virus?)
Hello, On Sat, 23 Feb 2013 17:09:53 -0600 "Mark Allums" wrote: > FYI, a few USB storage things were known to irrevocably > become Windows devices once used on Windows. This is rare > now. However, if you reformatted and it seems to work now, > forget about it. Please can you explain what do you mean by that? What is the thing you call "Windows device"? (Something like a roller blind?) Like they stopped being usable anywhere but Windows? Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130224044711.73400...@hugo.daonet.home
Re: Open new session with same user (in GDM)
On Wed, 20 Feb 2013 18:55:06 -0600
"Mark Allums" wrote:
> > Le 21/02/2013 00:32, Mark Allums a écrit :
> > >> From: Kent West [mailto:we...@acu.edu]
> > >> On 02/20/2013 04:07 PM, Alois Mahdal wrote:
> > >>> Hello to all!
> > >>>
> > >>> How do I create two same-user sessions from GDM?
> > >>>
> > >>> Say I'm logged in in Xfce (which I normally use, therefore have
> > >>> many windows / apps open) and want to quickly report a bug. But
> > >>> before I do that, I want to check if the same happens in GNOME.
> > >>>
> > >>> xfswitch-plugin ("User switching" in Panel item menu), gets me
> > >>> to the GDM, but even if I choose my username and "Gnome classic"
> > >>> session, typing my password only brings me to the old Xfce4
> > >>> session (i.e. the "session" option in GDM is ignored).
> > >>>
> > >>> Thanks,
> > >>> aL.
> > >>>
> > >>> P.S.: While typing this, I realized that 90% of these cases can
> > >>> be solved by creating a new temporary user, which is way more
> > >>> correct way of bug reporting.
Well, I have to correct myself here: it's not that super-correct
since you are changing 2 factors between 2 tests; you might lead
yourself (and poor unsuspecting developers) into fake conclusion
"it does not happen under gnome", while the it's really "it does
not happen under a newly created user".
> How about running two different DMs? Isn't that what that's for?
> Can't have competing DMs like KDM, GDM, and LightDM. But why
> shouldn't LXDE and e17 run at the same time?
Or Xfce4 and Xfce4 and Gnome and Xfce4? (I assume you mean DEs, not
DMs. Or we might as well want to run 2 inits! :D)
> If no competing sessions can exist, then that goes against the spirit
> of Unix.
Exactly that's what I had in mind.
While I understand that the demand might not be very strong, I do
not really see why multiple sessions could not exist. You can have
as many bashes/zshes/kshes/ashes/cshes as you want, and they don't
start shooting each other in their feet. OK, there's none to very
little overlapping in their spaces, but principle is the same. Why
the hell would Gnome touch Xfce4's files?
What about having one local session and one controlled from a remote
box? What's the principal difference between sshd and *dm?
> Would starting a second X session do any good at all, as Kent
> suggested? There would still be only one DBus?
Yep, that brought up gnome-session with no complaints. (I actually
even omitted the "-- :2" part.) Not via GDM, but that was not
really a requirement, it was rather that I was expecting GDM would
support that. Somehow.
~
But finally, the benefit was even smaller than as I commented in OP.
While the gnome session seemed to run OK, the actual *application*
I wanted to run didn't. Well, this makes sense in many apps like
MUA or web browser and it is often possible to address by telling
the app to use other profile/dotfileset. But we are just getting
to the point when simply creating user foo is easier.
Thanks all!
aL.
--
Alois Mahdal
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130221072916.4c3eb...@hugo.daonet.home
Re: Open new session with same user (in GDM)
On Wed, 20 Feb 2013 17:32:40 -0600 "Mark Allums" wrote: > [...] > As far as starting a new session at GDM, wouldn't it be like starting > any other session with GDM? Just choose your window manager+DE with > the pulldown menu? What am I missing? At the moment when you already have a session, it does not matter what you choose from the pulldown menu, you will be simply returned to the running session. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130221061150.5b311...@hugo.daonet.home
Open new session with same user (in GDM)
Hello to all!
How do I create two same-user sessions from GDM?
Say I'm logged in in Xfce (which I normally use, therefore have
many windows / apps open) and want to quickly report a bug. But
before I do that, I want to check if the same happens in GNOME.
xfswitch-plugin ("User switching" in Panel item menu), gets me
to the GDM, but even if I choose my username and "Gnome classic"
session, typing my password only brings me to the old Xfce4
session (i.e. the "session" option in GDM is ignored).
Thanks,
aL.
P.S.: While typing this, I realized that 90% of these cases can
be solved by creating a new temporary user, which is way more
correct way of bug reporting. However, in some cases (hard to
reproduce bugs, missing privileges to create user...) that might
not be the solution.
--
Alois Mahdal
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130220230738.4064a...@hugo.daonet.home
Re: OT - Convert output of byte count to GB count?
On Wed, 13 Feb 2013 14:41:47 +1300 Richard Hector wrote: > On 13/02/13 14:32, Jerry Stuckle wrote: > > On 2/12/2013 4:43 AM, Ralf Mardorf wrote: > > >> Assumed you'll build a fence, 10m long and every 1m there should > >> be 1 fence post, how many fence posts do you need? > >> > > > > 11 :) > > Depends how you interpret the instructions. If it's "you can leave up > to 1m hanging without a post", then you only need 9 :-) According to Jerry's interpretation: 11. According to yours, it's actually 10 posts less, i.e. only 1. (Finally, as we know, there's only 10 kinds of people...) > Richard > > -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130213153501.18456...@hugo.daonet.home
Re: how to launch these images?
On Sat, 9 Feb 2013 20:02:27 +0800 Long Wind wrote: > > I download the file above and mount it > I find vmlinuz and initrd.img and > I create menu entry for them in grub.cfg > but it does not work If you mount ISO, it only remains mounted while the OS is running (might get remounted on the next boot). When only GRUB is running, there's nothing that would keep the ISO mounted. > Can you help me? > or I have to burn it to a CD? So yes. You have to either burn it to CD, or to USB stick. In the latter case, look for the instructions at the RH's pages, or consider using unetbootin. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130209132234.704ac...@hugo.daonet.home
Re: Where do `pr_warning()`s end up?
On Sat, 9 Feb 2013 13:19:31 +1300 Chris Bannister wrote: > [...] > > Start looking here: > > root@tal:~# less /etc/sysctl.conf > [...] > > # Uncomment the following to stop low-level messages on console > #kernel.printk = 3 4 1 3 > [...] Looks interesting, although from what I understood from reading kernel sources (but maybe I'm being lost), pr_warning() is a macro around eprintf() which is wrapper around vfprintf(), which is standard C function (stdio.h, `man vfprintf`). I.e. these calls do not use printk(). Will try altering these, though... > Dang, if I can remember where the info is about the kernel "noise" > levels. Found it in kernel source, ./Documentation/sysctl/kernel.txt: printk: The four values in printk denote: console_loglevel, default_message_loglevel, minimum_console_loglevel and default_console_loglevel respectively. These values influence printk() behavior when printing or logging error messages. See 'man 2 syslog' for more info on the different loglevels. - console_loglevel: messages with a higher priority than this will be printed to the console - default_message_loglevel: messages without an explicit priority will be printed with this priority - minimum_console_loglevel: minimum (highest) value to which console_loglevel can be set - default_console_loglevel: default value for console_logleve Too bad that I won't be able to reproduce the original issue which was specific to password-protected SD cards. Since I don't have that card now. Or I'll just try to play with the values anyway: what bad can happen to me? ;-D Thanks for pointer. aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130209063452.60303...@hugo.daonet.home
Where do `pr_warning()`s end up?
Hello && GTOTD, I'm trying to make sense of some kernel driver code and I noticed some `pr_warning()` calls. From kernel.h I know pr_warning is an `eprintf()` macro: #define pr_warning(fmt, ...) \ eprintf(0, pr_fmt(fmt), ##__VA_ARGS__) while eprintf is defined in debug.c. So Does this appear somewhere during normal (Debian Wheezy) operation? Or is it purely a debugging thing and I would need a specifically compiled kernel? Can I somehow make the driver print them for me without need to recompile kernel? Thanks, aL. -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130208201031.54ea4...@hugo.daonet.home
Re: Client "daemon" for sorting e-mail via IMAP
Hello all,
with embarrassing delay :) I'd like to thank to you again.
Finally I looked at the imapfilter and found out that it was exactly
what I was looking for. Especially I liked how it's versatile and it
does not make assumptions about where you primarily want to store your
mail---you can make queries based on just about anything, and move
messages between as many accounts/boxes/folders as you can talk via
IMAP with.
And since my rules are pretty simple (all based on List-ID:) and few,
and the volume is not expected to be very high (this list plus couple
of much smaller ones), my config.lua did not get large nor complicated,
so need to patch it up from other scripts did not arise.
(Only thing that could be problem to someone is that IIUC, you
need to store your password in the plain text config.lua; so I
would not use it where security is big priority.)
And here's my config.lua:
---
-- Options --
---
options.timeout = 120
options.subscribe = true
options.info = false
-- Accounts --
am = IMAP {
server = 'mail.provider.cz',
username = 'real.n...@example.com',
password = 'foo',
ssl = 'ssl3',
}
am1 = IMAP {
server = 'mail.provider.cz',
username = 'alois punkt mahdal blob 1 laser ndmail at zxcvb
rock cz', password = 'bar',
ssl = 'ssl3',
}
-
-- Mailboxes --
-
-- Get all the messages in the mailbox.
am1i = am1.INBOX:select_all()
-- test --
imft = am1i:contain_field('List-Id', 'imftest')
imft:move_messages(am1.imftest)
-- mailing lists
x = am1i:contain_field('List-Id', 'debian-user.lists.debian.org')
x:move_messages(am1['debian-user'])
x = am1i:contain_field('List-Id', 'dancer-users.dancer.pm')
x:move_messages(am1['dancer-users'])
x = am1i:contain_field('List-Id', 'fprint.lists.freedesktop.org')
x:move_messages(am1['fprint-devel'])
x = am1i:contain_field('List-Id', 'users.httpd.apache.org')
x:move_messages(am1['users-httpd'])
x = am1i:contain_field('List-Id', 'juser.jabber.org')
x:move_messages(am1['jabber-user'])
x = am1i:contain_field('List-Id', 'musicbrainz-users.lists.musicbrainz.org')
x:move_messages(am1['musicbrainz-users'])
x = am1i:contain_field('List-Id',
'security-basics.list-id.securityfocus.com')
x:move_messages(am1['security-basics'])
-- SPAM --
x = am1i:contain_field('X-Spam-Flag', 'YES')
x:move_messages(am1['Junk'])
crontab solves the rest. Note that `options.info = false` is to turn
off output to avoid cronspam (e.g. how many messages were moved).
Thanks,
aL.
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20130208041039.77f4a...@hugo.daonet.home
Re: Client "daemon" for sorting e-mail via IMAP
Thanks everybody for answers. I wonder: is it possible to do the fetchmail/procmail solution without actually *storing* the mail where the daemon runs? I mean, just talk to the remote IMAP server and tell him "move it to that folder there"? I would like to avoid the massive duplication. The VPS has quite a small disk, while on the mailbox side, the storage is free (i.e. claimed "unlimited" for certain price). They just don't help with the sorting/filtering. Also: On 2013-01-22 19:56, Joel Roth wrote: [...] I'll clean up and post the script if anyone is interested. Joel Thanks, Joel, would you be so kind? I'd really like to see your script. aL. -- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/c16d3cd48c3eec5dd4c4912ce3899...@zxcvb.cz
Client "daemon" for sorting e-mail via IMAP
Hello everyone! I wonder if there already is a solution for this: I have several mailboxes in various places that I access using several clients (e.g. other from my laptop, other from my Android and other from a public place). Some of boxes (e.g. the one I use for various mailing lists including debian-user) are quite high-traffic and obviously in need of some filtering, be it based on origin or the spam score. Since I'm using various clients, filtering using rules in MUA is not practical. I would prefer to have all this logic in a single place, (namely my personal VPS box), where for example a script would exist just for purpose of regularly checking new mail and moving the new messages to given folders based on pre-defined rules. I think that this could be nice universal solution for any number of mailboxes (as long as they support IMAP), completely avoiding the question of when I actually use which MUA to read/write stuff. Any ideas how to do this easily? (Of course, the mentioned VPS is Debian Wheezy.) Or, is there a ready solution that could be used right away? -- Alois Mahdal -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20130122184506.09d6e...@hugo.daonet.home
Re: Please help parsing file [sed, awk, fortran, bash]
On Thu, 30 Aug 2012 21:44:04 -0400 "John L. Cunningham" wrote: > Hi Daniel, > > Here's my awk solution: > [snip] Very nice, this should be added as an example to an awk-learning document :) -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20120901211414.1b2eb...@hugo.daonet.home
Re: How to disable TLS in exim4 client part?
Hi, On Tue, 03 Jan 2012 21:59:38 +0100, Chris Davies wrote: [...] I tried adding MAIN_TLS_ENABLE = no to exim4.conf.template[...] This is not a yes/no value, it's a defined / not defined value. So by defining it as "no" you're actually setting it to true. So you should comment out the line, thus: # MAIN_TLS_ENABLE = yes [...] Thanks! I hardly would realize this myself :) However this does not work for me. Actually I believe it was me who added MAIN_TLS_ENABLE line, so I just removed it, but exim4 still uses STARTTLS when talking to that smarthost. Additionally, comments just before ".ifdef MAIN_TLS_ENABLE" already led me to believe this : # TLS/SSL configuration for exim as an SMTP server. # See /usr/share/doc/exim4-base/README.Debian.gz for explanations. And "2.2.1. Exim 4 as TLS/SSL client" in that README.Debian does not speak at all about disabling TLS. "Both exim4-daemon-heavy and exim4-daemon-light support TLS/SSL using the GnuTLS library and STARTTLS. Exim will use TLS via STARTTLS automatically as client if the server Exim connects to offers it." So I guess that only way how to really disable TLS would be compiling exim4 *without support* for TLS, which is waaay too scary for me, and definitely not worth the effort. Especially when there's this nifty[1] -v option to mail, which actually solves the original offer in much better way: It simply shows all the SMTPish information I need to STDERR, as well after STARTTLS as before. Thanks, aL. 1: http://pastebin.com/uN1d9s7v -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.v7ku6laqntc...@aloism.cz.avg.com
Re: Exim4 (mis)configuration for smarthost (Was: Re: How to disable TLS in exim4 client part?)
On Tue, 03 Jan 2012 21:30:22 +0100, Alois Mahdal wrote: It seems that exm4 does not rewrite addresses as I wish according to /etc/exim4/email-addresses: aloism@localhost: correct.sen...@mydomain.cz alo...@gebba.aloism.test.local: correct.sen...@mydomain.cz aloism@gebba: correct.sen...@mydomain.cz aloism: correct.sen...@mydomain.cz Neither of these lines work. But manual says: (sender addresses) "...are rewritten for users that appear to be in the local domain..." and my domain (in resov.conf) is different from gebba.aloism.test.local. Might this be the cause? That "alo...@gebba.aloism.test.local" does not "appear to be in local domain"? Nope, blind silly human. :-D It's /etc/email-addresses Now it works: addresses rewritten, e-mails delivered! :-) (the gate does not seem to care about thing in "EHLO thing") Thanks for patience! :) aL. PS: Well, not SO silly, I guess there's bug in http://wiki.debian.org/GmailAndExim4 ...going to report... -- Alois Mahdal using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.v7i0172n53t...@aloism.cz.avg.com
Exim4 (mis)configuration for smarthost (Was: Re: How to disable TLS in exim4 client part?)
On Tue, 03 Jan 2012 19:16:49 +0100, Joe wrote: "R=smarthost T=remote_smtp_smarthost" I could not find in Exim4 specs what these fields should be. TIA, aL. R is router, T is transport. A router determines where to send a particular email next, a transport is a software means of achieving that. Routers and transports may be user-defined for individual domains or even per recipient, which gives enormous flexibility in shifting mail around. Wow, thanks for insight :-) This entry is only telling you that delivery has been made successfully to your local smarthost, local.gate.cz. The next step is to find out what this server has done with the email, which depends on whether the server is also yours or you have to ask for logs from someone else. If the latter, quote them time and date and the message ID beginning "1Rhhe.." (basically your second line above). The server will definitely have a log entry which is the 'other half' of your entry above, showing this email as being accepted, and the message ID should then appear in one or more additional server log entries where the server tries to send the message to the next SMTP server in the path to your recipient. Unfortunately, the server is not mine, and for some reasons, I can't expect much support. Except when I actually found better solution to the original problem. Generally the SMTP protocol is reliable in the communications sense, in that it will either deliver the mail or it will tell the sender that it can't, and why not. If mail is disappearing without error messages being returned to the sender, an anti-spam system is usually the reason. So probably antispam filter is throwing them away. (Because my client is acting weird... Because I've probably done something silly before, or I'm trying impossible configuration...) Thankfully, I have found that using -v (verbose) option for tests gives me enough insight to communication (about what I expected to find after disabling TLS). And the lines I like the least are: SMTP>> EHLO gebba.aloism.test.local ... SMTP>> MAIL FROM: SIZE=1467 AUTH=alo...@gebba.aloism.test.local ... It seems that exm4 does not rewrite addresses as I wish according to /etc/exim4/email-addresses: aloism@localhost: correct.sen...@mydomain.cz alo...@gebba.aloism.test.local: correct.sen...@mydomain.cz aloism@gebba: correct.sen...@mydomain.cz aloism: correct.sen...@mydomain.cz Neither of these lines work. But manual says: (sender addresses) "...are rewritten for users that appear to be in the local domain..." and my domain (in resov.conf) is different from gebba.aloism.test.local. Might this be the cause? That "alo...@gebba.aloism.test.local" does not "appear to be in local domain"? aL. -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.v7iy8ws3ntc...@aloism.cz.avg.com
Re: How to disable TLS in exim4 client part?
Sorry for delay. On Mon, 02 Jan 2012 09:51:14 +0100, Joe wrote: Is there anything useful in /var/log/exim4/mainlog? That at least should be in plaintext. Lines in Exim4 sound like purring to me: -8<- 2012-01-02 14:09:46 1Rhhe6-00059F-9d <= alo...@gebba.aloism.test.local U=aloism P=local S=442 2012-01-02 14:09:46 1Rhhe6-00059F-9d => correct.addres...@mydomain.cz R=smarthost T=remote_smtp_smarthost H=local.gate.cz [10.1.2.3] X=TLS1.0:RSA_AES_256_CBC_SHA1:32 DN="C=CZ,ST=Czech Republic,L=Town,O=Company,CN=cn.local.gate.cz" 2012-01-02 14:09:46 1Rhhe6-00059F-9d Completed ->8- I have obuscated many fields, the adresses there are correct (I also tried the e-mail address in lowercase), except for one thing I'm not sure about: this part, which is literally: "R=smarthost T=remote_smtp_smarthost" I could not find in Exim4 specs what these fields should be. TIA, aL. -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.v7iqznbnntc...@aloism.cz.avg.com
How to disable TLS in exim4 client part?
Hello! Please, how do I disable TLS in exim4 as a client? I'm trying to send all emails via single SMTP smarthost, but for some reason, server does not deliver them, so I want to look into the trace. But exim4 keeps negotiating TLS, not making life easier for me--the Mallory. (Not that it should... :)) I tried adding MAIN_TLS_ENABLE = no to exim4.conf.template (AND to conf.d/main/03_exim4-config_tlsoptions), but this does not affect client. So what should I do? TIA, aL. -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.v7fiz6lontc...@aloism.cz.avg.com
Best mount point for shared files
#!/usr/bin/perl use warnings; use IIUC; use AFAIK; use Knowledge::Newbie qw( learn ); my $question=<<'EOF'; Hi, I have a server from which I want to share files via multiple protocols--so far it's HTTP and SMB. The files do not reside on the server, they are mounted from other server via NFS. I was wondering: where to put the files? Which is the most "Debian" way? For example, I'm mounting these NFS shares to "serve": storage:/srv/nfs/applications storage:/srv/nfs/samples storage:/srv/nfs/tools I'm planning that all of them will available (to Windows stations connected to "serve") as: http://serve/applications/ http://serve/samples/ http://serve/tools/ \\serve\applications\ \\serve\samples\ \\serve\tools\ (All shares will be read-only. It's an isolated environment for testing, there's no need to have access policy--anyone can read, no-one can write.) A) According to FHS, I should put them in /srv/smb/ for SMB and /srv/http/ for HTTP. But that would mean mounting each FS to two mountpoints. Sounds confusing to me...is it normal? Are there any caveats with having one FS in two mountpoints? B) I have noticed that Apache2 has, by default, DocumentRoot set to /var/www/. Does this mean that Apache expects me to have main storage here? (I know I can change DocumentRoot, but I'm trying to learn from default settings... .-D) Actually I'd expect to store only index.html and such (there will be some of these). C) Or I shoud put all to a different location and use symlinks (I think Apache does not like this)? D) As C), but use "mount -o bind"? Is this how bind is supposed to be used? I'm trying to get an idea of what is "common" or "normal". I welcome any tips, experiences. Thanks! aL. EOF $question .= 'P.S.: Sorry for the typos. I know there are some.'; ask $question; -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.v3aizep2ntc...@aloism.cz.avg.com
Re: [OT] Searching "Squeeze" on Google
On Tue, 04 Oct 2011 18:48:30 +0200, Camaleón wrote: On Tue, 04 Oct 2011 18:19:52 +0200, Alois Mahdal wrote: sorry for OT--this is probably more about Google, but I find this quite puzzling *and* somehow Debian-(popularity)-related :) I wouldn't worry about this. Google's search results varies *a lot* depending on your location, stored, cookies, browser, IP address, rainy or sunny day, etc... ;-) Thanks, I kinda forgot that---trying to see things simple way all day long. :-) Just took a glimpse into HTTP headers and I remembered that there's a whole (beautiful & scary) world between my browser and Google... ;-) Why first Debian-related hit in results after searching "squeeze" on Google is actually Wheezy? Not Squeeze? (...) My first Debian "hit" directs me to Squeeze (current stable). In fact, I see nothing about Wheezy (at least within the first 10 results). ...kind of soothing to know, though... :-D aL. -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.v2t753zuntc...@aloism.cz.avg.com
[OT] Searching "Squeeze" on Google
Hi, sorry for OT--this is probably more about Google, but I find this quite puzzling *and* somehow Debian-(popularity)-related :) Why first Debian-related hit in results after searching "squeeze" on Google is actually Wheezy? Not Squeeze? Actually (not counting "images/videos/news for..."), Debian Wheezy was 4th result, Squeeze was 11th. I always thought that Google gives better score to pages that are actually related, i.e. "most people who were looking for 'squeeze' went there and did not come back". Any ideas? Thanks, aL. -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.v2t4zeopntc...@aloism.cz.avg.com
Re: A courier FAM/Gamin error and an IMAP client
On Thu, 01 Sep 2011 18:45:04 +0200, alois.mahdal.1-ndm...@zxcvb.cz wrote: Thanks to all just for reading up to this point. :) I hope I'll be able to thank you once more for any ideas, pointers... (Well am I so bad at Googling or what? I always start to find reasonable sources *right after* I ask.) There's a debate on this error, though the bug is originally about this message causing login failures: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=578937 Questions remain yet unanswered, though... -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/ac34bfc41aade0c16ab7026c32358...@zxcvb.cz
A courier FAM/Gamin error and an IMAP client
Hi, recently we have installed Debian Squeeze, and set up Postfix + courier for purposes of testing client-side e-mail filter on a Windows client station. The server is supposed to provide about 16 accounts with prepared e-mails while the client application (SeaMonkey on Windows 7 amd64) is configured to download messages from 8 accounts via POP3 and from 8 accounts via IMAP. The amount of e-mails is really huge (about 240k), Client is supposed to download all this non-interactively, say, within 30 hours or so. The issue I'm concerned is happening even without our filter. From my gut feeling, it's more about courier---it should be same issue for any other client on any other OS. For each IMAP folder, first time SeaMonkey accesses this folder within one client session, rather cryptic error is returned: "Filesystem notification initialization error -- contact your mail administrator (check for configuration errors with the FAM/Gamin library)" (Strange thing is that I haven't found anyting in mail.log about this.) The client seems to download list of messages correctly, though, but its behavior kind of hinders the non-interactivity testing: you have to click through all these messages, if you want SeaMonkey to download all e-mails. From what I have found out, one can get rid of this by installation of gamin package. But there's dependency problem: gamin needs libgamin0, but libgamin0 conflicts libfam0. I'd rather know more about it before I mess up anything. Now the questions: - Is it safe, in sense of possible discrepancies (losing any undelivered e-mails, delivering e-mails twice, or hindering performance) to exchange libfam0 for gamin+libgamin0? - From the point of client, what is the real meaning and severity of this message? I can't see a reason why client should bother as long as IMAP works. Isn't it a bug in SeaMonkey that it concerns about this at all? - Wouldn't it be better if courier would be installed with gamin in the first place? Thanks to all just for reading up to this point. :) I hope I'll be able to thank you once more for any ideas, pointers... Alois Mahdal P.S.: OT: I have already realized that for testing, SeaMonkey was definitely not a lucky choice, and decided to find an alternative client. So if you had any tips, I'd appreciate. Yes it's Windows, so it's really an off-topic, but still there's a chance your favorite client is ported :) Keywords are: easily distributable configuration, pop3, imap, smtp, client-side rules, transparent mail storage, good logging and reliability (that's the most important). Also, strong CLI would be a huge advantage. -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/2c5a7380c6fbaf7ae768d43e203a9...@zxcvb.cz
Re: test :-)
On Sat, 18 Sep 2010 14:53:39 +0200, dan wrote: [...] Maybe a change in that point ( #6 ) is required. ...like adding "...except when you are Debian developer? ;-D -- Alois Mahdal, using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.vjvc160v53t...@hugo.lennycz-depot
Re: OT Re: root can't sudo
On Tue, 28 Sep 2010 03:26:42 +0200, T o n g wrote: On Tue, 28 Sep 2010 02:58:05 +0200, Thierry Chatelet wrote: I dont use sudo, but can you explain me,so I will go to bed with more knowledge, why root would need sudo? invoking my scripts embedded with sudo as root I'm, not sure what you mean by "embedded" here, so my guess is that you mean making something like this work for both user and root: u...@machine ~/$ cat > aptupgscript.sh <Alois Mahdal using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.vjqxuati53t...@aloism.cz.avg.com
Re: single click?
Hi, On Sun, 12 Sep 2010 07:58:29 +0200, Doug wrote: [...] Wouldn't you think that Linux would automatically default to that? No, I wouldn't, and I believe it's better that way. At least for accessibility reasons, it's better to keep the more "sturdy", less error-prone alternative default. It's very easy to switch to the other one, but for some people (or in some situations), switching the other way might be somewhere on a scale from hard to impossible. Thx, Al. -- Alois Mahdal using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.vi1dp4o253t...@aloism.cz.avg.com
Re: Dumping terminal contents to a file
On Wed, 08 Sep 2010 22:28:18 +0200, manu revah wrote: Le Mer 8 septembre 2010 18:52, Alois Mahdal a écrit : Hi Debian users, I wonder if there's a simple and universal way to dump output of terminal to a file, particularly for demonstration purposes. not sure if this would do: screendump Yes, Manu! That's exactly what I was looking for! It seems to be kinda smarter alternative to copying /dev/vcs?, which: * ignores garbage characters * ignores spaces at the end of line The fact is that to make it work, you must grant user read rights for /dev/vcs? and /dev/vcsa?, which potentially might be a security issue (as user's process might have access to other tty contents), but for my purpose (testing on VMWares / asking at mailing lists ;D), it's not really a disadvantage. Thanks a lot, guys, now I have a nice new set of toys to play with :-D Al. -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.viqh58cl53t...@hugo.lennycz-depot
Re: Dumping terminal contents to a file
On Wed, 08 Sep 2010 19:18:23 +0200, Frederik Kriewitz wrote: 2010/9/8 Alois Mahdal : I wonder if there's a simple and universal way to dump output of terminal to a file, particularly for demonstration purposes. You might be interested in "script", it's part of the bsdutils package which is probably already installed on your system. Thanks, that will be very useful--especially for posts to lists like this one .-D, but still: * it saves some way more than I need (all bash's Tab completion suggestions, even bells, i haven't checked with colored outputs), usually I need just the final plain text * what if I did not know that I wanted to dump the output and I can't reproduce it again? I'd appreciate possibility to do a "screenshot", rather Thanks anyway, you already helped me a lot :D Al. -- Alois Mahdal using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.vip6u4kx53t...@aloism.cz.avg.com
Re: Straw poll: What browser do you use?
On Tue, 07 Sep 2010 15:16:26 +0200, B. Alexander wrote: So what do others use? --b Opera 10. Oh, it seems that I'm only one that's using Opera as default on my Debian box. On Win, It's been my #1 for years and I'm so used that I simply can't abandon all that so-comfy-comfy design :D Even that it's closed, it doesn't play perfectly with Flash (especially on amd64)--but that's getting way better and it's somehow sluggish onload compared to FF clones, even somehow unstable. I still love it :D Thx, Al. -- Alois Mahdal ta zxcvb tod cz ...using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.vip4lyiz53t...@aloism.cz.avg.com
Dumping terminal contents to a file
Hi Debian users, I wonder if there's a simple and universal way to dump output of terminal to a file, particularly for demonstration purposes. Most of the time when I need to do that, I either use X (or PuTTY, when on Windows) facilities, but often it's not possible, or it may become awkward (due to need to switch between mouse and keyboard). I know there are things like 'screen', but I'm looking for the most universal solution that would work on every system. For example, on SUSE, I can: cat /dev/vcs > example.txt (...though it's catted as a single line with *all* spaces up to $COLS (meaning, e.g. that each--even empty--line would appear as $COLS 0x020-spaces), so it needs some post-processing.) ...but that didn't work on Debian/Lenny for me. Any suggestions? (Extra bonus points for linux-osx-freebsd-unicode-compatible version. :-D) Thanks a lot, Al. -- Alois Mahdal ta zxcvb tod cz ...using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.vip3s3lf53t...@aloism.cz.avg.com
Re: Reklama w serwisie
...I can't speak Polish, but if I can get the point, she is asking about putting paid advertisement on debian.org :-) I guess she might understand Czech at least as I "do" Polish: --- Dobrý den, On Thu, 15 Jul 2010 14:05:39 +0200, Marta Pęczek wrote: Chciałabym się dowiedzieć, czy istnieje możliwość umieszczenia płatnej reklamy w serwisie debian.org. Reklama na stránkách Debianu není možná. Stránky jsou provozovány neziskovou (Organizacja non-profit) organizací SPI Inc. Navíc, tato diskuzní skupina je určena výhradně pro komunikaci v Angličtině. ... "Wysyłaj wszystkie wiadomości w języku angielskim."[2] [1] http://pl.wikipedia.org/wiki/Software_in_the_Public_Interest [2] http://www.debian.org/MailingLists/index.pl.html S pozdravem, Alois Mahdal -- Alois Mahdal using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.vg2tt0ty53t...@aloism.cz.avg.com
Re: Broken dependencies
On Sat, 01 May 2010 19:14:52 +0200, Florian Kulzer wrote: On Sat, May 01, 2010 at 17:20:56 +0200, Alois Mahdal wrote: Hello, because a stupid mistake, I have interrupted apt-get during early stage of dist-upgrade from Lenny to Squeeze. Now I cannot get apt-get working and I don't know how to fix it. [...] dpkg: warning: 'ldconfig' not found on PATH. dpkg: 1 expected program(s) not found on PATH. NB: root's PATH should usually contain /usr/local/sbin, /usr/sbin and /sbin. E: Sub-process /usr/bin/dpkg returned an error code (2) Both "locale" and "ldconfig" were shifted from libc6 to libc-bin going from Lenny and Squeeze; my guess would be that you upgraded to the Squeeze version of libc6, but you did not yet complete the installation of libc-bin, so now these important commands (and others) are missing on your system. (The libc-bin package does not exist for Lenny.) If dpkg still works then you can try to download libc-bin for your architecture from http://packages.debian.org/squeeze/libc-bin and install the .deb with "dpkg -i ...", see here: http://www.debian.org/doc/manuals/reference/ch02.en.html#_rescue_with_the_dpkg_command Thanks, Florian. Unfortunately, dpkg was too broken, so i had to use Ubuntu livecd (because I could not find dpkg at Debian installer CD / rescue mode) to boot and do following: 1. mount -t ext2 /dev/sdaX /media/root # ...and all relevant partitions (/usr, /var, ...) respectively 2. mount --bind /dev /media/root/dev # Since dpkg was complaining about missing Debian-exim group, I had to get rid of statoverride: 3. chroot /media/root 4. dpkg-statoverride --remove /etc/exim4/passwd.client # Later I must not forget to restore /var/lib/dpkg/statoverride from statoverride-old 5. exit && cd 6. wget http://ftp.cz.debian.org/debian/pool/main/e/eglibc/libc-bin_2.10.2-6_amd64.deb 7. wget http://ftp.cz.debian.org/debian/pool/main/e/eglibc/libc6_2.10.2-6_amd64.deb 8. dpkg --root=/media/root -i libc-bin_2.10.2-6_amd64.deb 9. dpkg --root=/media/root -i libc6_2.10.2-6_amd64.deb 10. reboot After reboot and restoration of statoverride, I can continue with apt-get upgrade and apt-get dist-upgrade; dpkg seems to work. Thanks again to you all! -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/op.vb3fsevk53t...@hugo.lennycz-depot
Broken dependencies
Hello, because a stupid mistake, I have interrupted apt-get during early stage of dist-upgrade from Lenny to Squeeze. Now I cannot get apt-get working and I don't know how to fix it. Could anyone of you please help me? I'll be also happy with advice on how to get proper debug info for you, of course. $ apt-get -fyo Debug::pkgProblemResolver=yes upgrade ...returns: --- Reading package lists... Building dependency tree... Reading state information... The following packages have been kept back: abiword abiword-plugin-grammar abiword-plugin-mathview alacarte alsa-utils amule amule-common amule-utils apt apt-utils aptitude at at-spi avahi-daemon bash bind9-host bluetooth bogofilter-bdb brasero bsh bsh-gcj capplets-data cheese cmake cpp cpp-4.3 cron cups cups-bsd cups-client cups-driver-gutenprint cupsddk dasher dasher-data dbus dbus-x11 debian-keyring debianutils default-jre default-jre-headless deskbar-applet devscripts diff djvulibre-desktop dnsutils dpkg-dev einstein ekiga empathy eog epiphany-browser epiphany-browser-data epiphany-extensions epiphany-gecko evince evolution evolution-common evolution-data-server evolution-data-server-common evolution-exchange evolution-plugins evolution-webcal exim4 exim4-base exim4-config exim4-daemon-light exo-utils fast-user-switch-applet file-roller filezilla filezilla-common foo2zjs ftp g++ g++-4.3 gcalctool gcc gcc-4.3 gcc-4.3-base gconf-editor gconf2 gconf2-common gdebi gdebi-core gdm geany gedit gedit-common ghostscript gimp gimp-data gksu gnome gnome-accessibility gnome-accessibility-themes gnome-applets gnome-applets-data gnome-bluetooth gnome-cards-data gnome-control-center gnome-core gnome-desktop-environment gnome-games gnome-games-data gnome-icon-theme gnome-keyring gnome-mag gnome-media gnome-media-common gnome-mount gnome-netstatus-applet gnome-nettool gnome-office gnome-orca gnome-panel gnome-panel-data gnome-power-manager gnome-screensaver gnome-session gnome-settings-daemon gnome-system-monitor gnome-system-tools gnome-terminal gnome-terminal-data gnome-themes gnome-utils gnome-volume-manager gnuchess gnumeric gnumeric-common gnupg gok gparted gpgv grub gstreamer0.10-alsa gstreamer0.10-ffmpeg gstreamer0.10-gnomevfs gstreamer0.10-plugins-bad gstreamer0.10-plugins-base gstreamer0.10-plugins-good gstreamer0.10-plugins-ugly gstreamer0.10-tools gstreamer0.10-x gthumb gthumb-data gtk2-engines gtk2-engines-pixbuf gtk2-engines-xfce gucharmap guile-1.8-libs gvfs gvfs-backends hal hpijs hplip hplip-data ia32-libs ia32-libs-gtk icedove iceweasel iceweasel-l10n-cs imagemagick info inkscape iproute kdelibs4c2a kdiff3 kerneloops latex-xft-fonts lib32asound2 lib32gcc1 lib32ncurses5 lib32nss-mdns lib32stdc++6 lib32z1 libafterimage0 liballegro4.2 liballegro4.2-plugin-jack libaprutil1 libapt-pkg-perl libarchive1 libart2.0-cil libarts1c2a libasound2 libasound2-dev libatspi1.0-0 libavc1394-0 libavformat52 libbonobo2-0 libbonobo2-common libbonobo2-dev libbonoboui2-0 libbonoboui2-common libbonoboui2-dev libboost-date-time-dev libboost-dev libboost-doc libboost-filesystem-dev libboost-graph-dev libboost-iostreams-dev libboost-program-options-dev libboost-python-dev libboost-regex-dev libboost-serialization-dev libboost-signals-dev libboost-test-dev libboost-thread-dev libboost-wave-dev libc6 libc6-dev libc6-i386 libcairo2 libcairo2-dev libcairomm-1.0-1 libcdio-cdda0 libcdio-paranoia0 libclass-accessor-perl libcompress-raw-zlib-perl libcompress-zlib-perl libcrypt-ssleay-perl libcups2 libcupsimage2 libcurl3 libcurl3-gnutls libdbus-glib-1-2 libdc1394-22 libdigest-sha1-perl libdirac-encoder0 libdirectfb-dev libdirectfb-extra libdjvulibre21 libdvdnav4 libebook1.2-9 libecal1.2-7 libedata-book1.2-2 libedata-cal1.2-6 libedataserverui1.2-8 libedit2 libegroupwise1.2-13 libempathy-common libempathy-gtk-common libenchant1c2a libept0 libesd0 libesd0-dev libexchange-storage1.2-3 libexo-0.3-0 libfcgi-perl libfluidsynth1 libfreebob0 libgail-common libgail-dev libgail-gnome-module libgail18 libgcc1 libgcj-bc libgconf2-4 libgconf2-dev libgconf2.0-cil libgdl-1-common libgfortran3 libgimp2.0 libgksu2-0 libgl1-mesa-dev libgl1-mesa-dri libgl1-mesa-glx libglade2-0 libglade2-dev libglade2.0-cil libglib-perl libglib2.0-0 libglib2.0-cil libglib2.0-dev libglibmm-2.4-1c2a libgnome-keyring-dev libgnome-keyring0 libgnome-media0 libgnome-vfs2.0-cil libgnome-window-settings1 libgnome2-0 libgnome2-common libgnome2-dev libgnome2-perl libgnome2-vfs-perl libgnomecanvas2-0 libgnomecanvas2-common libgnomecanvas2-dev libgnomecups1.0-1 libgnomekbd-common libgnomeprint2.2-0 libgnomeprint2.2-data libgnomeprintui2.2-0 libgnomeprintui2.2-common libgnomeui-0 libgnomeui-common libgnomeui-dev libgnomevfs2-0 libgnomevfs2-common libgnomevfs2-dev libgnomevfs2-extra libgomp1 libgphoto2-2 libgphoto2-port0 lib
