Re: solved: `ls` shows file, `bash` says "No such file" ???
On Mon, May 05, 2014 at 08:50:58AM +0900, Joel Rees wrote: > Although, if we wanted to get really crazy, we could say this is a bug in > the spec of ld. But if we go there, we have to acknowledge that the spec of > ld matches the general C/*nix run-time spec, so the bug is in ... (chasing > our semantic tail a bit) ... certain implementation-dependent, but de-facto > spec, low-level elements of the C run-time which make it hard to return > non-scalar error messages from low-level tools. > > (I know that sounds like a string of buzz-words, but making it any simpler > just ends up as a pejorative reference to the 8086's lack of extra address > registers. :-\ ) Indeed. I think my brain just broke. Let's just say the bug report should be against Tom Roche's workplace's network security bods :) Cheers, Tom -- I've got a COUSIN who works in the GARMENT DISTRICT ... signature.asc Description: Digital signature
Re: solved: `ls` shows file, `bash` says "No such file" ???
On Mon, May 05, 2014 at 06:44:30AM +0900, Joel Rees wrote: > Well, okay, we need to start somewhere, and, while we suspect ld, we don't > really know for sure. And we suspect that the actual fix may not end up > being in ld. > > So, what is the name of the package that is trying to load libc6:i386? That > probably isn't where the bug is, but it seems to be the cause of the bug. Well, in this case, Firefox was looking to link in some 32-bit libraries, but those libraries weren't installed, not even a 32-bit linker. The problem was required libraries not being installed, the symptom was a generic "file not found" error which led to all the confusion at the start of the thread. The reason the libraries weren't installed was that Firefox was installed from some external source without the user ensuring the relevant libraries were available to the application. Debian can't be held accountable for that. If the 32-bit Firefox had been able to talk to the 64-bit linker (I'm not sure that's even possible), then it would be up to the linker to return a more meaningful error than "file not found", at least tell us which library or which function was being requested, and so a bug could be reported against libc6 either generically or :amd64 (to be passed upstream). Otherwise there is no bug to be reported against Debian. I'm at best vague on what happens if an application can't find a linker to pull in requested libraries. Should the bug be reported against Firefox? It's an old version, are more recent versions more precise in their error messages in this kind of situation? Cheers, Tom -- Should I start with the time I SWITCHED personalities with a BEATNIK hair stylist or my failure to refer five TEENAGERS to a good OCULIST? signature.asc Description: Digital signature
Re: [multiarch] easy fix for interarchitectural package conflict?
On Sun, May 04, 2014 at 04:04:30PM -0400, Tom Roche wrote: > summary: jessie/sid:amd64 box must install an i386 package which depends on > libgif4:i386, but > > - libgif4:i386 conflicts with libgif4:amd64 > - important apps depend on libgif4:amd64 Is this the same box that you were having the firefox trouble with? At this point, short of getting the problem sorted properly, your best option would be to run this stuff in a 32-bit VM. Cheers, Tom -- Edwin Meese made me wear CORDOVANS!! signature.asc Description: Digital signature
Re: solved: `ls` shows file, `bash` says "No such file" ???
On Sun, May 04, 2014 at 12:03:17PM -0400, Tom H wrote: > But installing libc6:i386 and a few others allows firefox to work: Meaning installing the :i386 versions of the libraries was (part of) the solution, not the problem. Any bug report would have to go against a part of the system that was part of the problem (not withstanding that firefox was installed in a non-standard location and from a non-standard repository). The only real candidates, given that the problem was mainly regarding an insufficient error message about missing libraries, are ld or the kernel. Anything further up than the linker in this scenario can only say "something went wrong, but I don't know what" unless they themselves are given more detailed information. Cheers, Tom -- A man who keeps stealing mopeds is an obvious cycle-path. signature.asc Description: Digital signature
Re: solved: `ls` shows file, `bash` says "No such file" ???
On Sun, May 04, 2014 at 06:23:05PM +0900, Joel Rees wrote: > On Sun, May 4, 2014 at 8:05 AM, Tom Roche wrote: > > > > > summary: solution: install jessie package=libc6:i386 et al > > > > Well, you've actually pinned the problem pretty well. Who, or, rather, > which tool should be responsible for grabbing the global error string at > that point? My first guess would be ld, but I'm not sure that would be the > place to start the bug report. Maybe ia32-libs or even libc6:i386. I don't think your suggestions of ia32-libs or libc6:i386 are good. There is no ia32-libs in wheezy+ and libc6:i386 wasn't installed at the time of the error so can't be responsible. The bug would have to be reported against a package in the execution chain that was at least installed. Cheers, Tom -- If you look good and dress well, you don't need a purpose in life. -- Robert Pante, fashion consultant signature.asc Description: Digital signature
Re: Question about wheezy-backports
On Sat, May 03, 2014 at 11:28:22PM -0600, Paul E Condon wrote: > I have added to my sources.list the following line: > > deb http://ftp.us.debian.org/debian/ wheezy-backports main > > But according to http://backports.debian.org/Instructions/, that won't > make the backport appear in interactive aptitude, or be automatically > 'upgraded' to the backported version. To actually install the > backport, I also have to add the phrase -t wheezy-backports to the > install command. And I have to know that there is a backport for the > particular deb in which I have an interest. I suppose I could try > reading the debian-backports-announce mailinglist, but I am so seldom > bothered with having the latest version that I can't believe I will do > that. Is there also a simple list of backported debs that I can browse > on the web to know what is available, or a wiki pointer? I'd rather > not subscribe to another list and monitor it regularly just for the > very rare occation when I actually need a backport. Probably not quite what you're looking for, but with a backports entry in your sources.list you can search or limit within aptitude with '?version(bpo)'. With a limit view this will only show available packages which have a version in backports (specifically it will limit to any packages with "bpo" in the version string, which all backports do). Cheers, Tom -- : How would you disambiguate these situations? By shooting the person who did the latter. -- Larry Wall in <199710290235.saa02...@wall.org> signature.asc Description: Digital signature
Re: Debian 7.x and desktop environments
On Sat, May 03, 2014 at 12:19:50PM -0700, Robert Holtzman wrote: > > > > tasks-xfce-desktop is a meta package that includes xfce, and also a set > > of additional packages that are typically used on a desktop system, > > like libreoffice and iceweasel. > > Those packages should be part of an OS install, not a DE, or am I wrong? I think you are wrong. Those applications (office-suite, web-browser) are applications on top of your desktop environment, which is a layer on top of the OS. They are certainly useful to most people, but they are not crucial to a functional desktop environment, and they are certainly not crucial to a working operating system. Added to that, intalling one of the -desktop tasks (for the most part) pulls in a particular office suite and a particular browser, which might not be those I want installed. Cheers, Tom -- Did an Italian CRANE OPERATOR just experience uninhibited sensations in a MALIBU HOT TUB? signature.asc Description: Digital signature
Re: sudo in X-environment
On Fri, May 02, 2014 at 02:04:46PM +0200, Slavko wrote: > > On Thu, May 01, 2014 at 07:29:08PM +0100, Tom Furie wrote: > > > Here, have a wrapper script > > > > > > --8x > > > > > > #!/bin/sh > > > > > > xhost + si:localuser:root > > > sudo $@ > > > xhost - si:localuser:root > > > > > > --8x > > And what about gksu(do)? > > ... and preserving environment when acting as a su frontend. It is > useful to menu items or other graphical programs that need to ask a > user's password to run another program as another user. Context is all important. The code you quoted above was a quick response to Артур saying 'xhost +si:localuser:root' was too long to type, which came from me saying that if xhost *must* be used (i.e. there is no other alternative) then at least fully qualify who is allowed to connect. Of course using one of the "graphical sudos" would be the better choice as they are much more complete solutions than the code above. Cheers, Tom -- "Eat, drink, and be merry, for tomorrow you may work." signature.asc Description: Digital signature
Re: sudo in X-environment
On Thu, May 01, 2014 at 05:19:59PM +, Артур Истомин wrote: > On Thu, May 01, 2014 at 05:46:36PM +0100, Tom Furie wrote: > > 'xhost +' allows anyone anywhere access to your X server. If you must use > > xhost in this situation it would be much safer to use 'xhost + > > si:localuser:root', this would allow only root from the local machine to > > connect. Just be sure to run 'xhost - si:localuser:root' as soon as you > > are done. You can run xhost without any arguments to see exactly who is > > alowed to connect to you. > > Exactly :) But I am very lazy. It is very long string for me on my local > machine with only one user =) Here, have a wrapper script --8x #!/bin/sh xhost + si:localuser:root sudo $@ xhost - si:localuser:root --8x Save it as xdo and it's even less typing than 'xhost + ... xhost -' :) Cheers, Tom -- Absence makes the heart grow fonder. -- Sextus Aurelius signature.asc Description: Digital signature
Re: sudo in X-environment
On Thu, May 01, 2014 at 04:30:11PM +, Артур Истомин wrote: > On Thu, May 01, 2014 at 10:51:21AM +0200, Johann Spies wrote: > > For years I have used sudo both in server administration and on the desktop. > > > > Lately I get the following error message and I do not really know what to do > > about it other than logging into the xserver as root - which I do not want > > to do: > > > > $ sudo system-config-printer > > > > No protocol specified > > error: XDG_RUNTIME_DIR not set in the environment. > > Unable to init server: Could not connect: Connection refused > > No protocol specified > > > > xhost + > sudo system-config-printer > xhost - # don't forget to do it after you're done with system-config-printer 'xhost +' allows anyone anywhere access to your X server. If you must use xhost in this situation it would be much safer to use 'xhost + si:localuser:root', this would allow only root from the local machine to connect. Just be sure to run 'xhost - si:localuser:root' as soon as you are done. You can run xhost without any arguments to see exactly who is alowed to connect to you. Cheers, Tom -- If there is a wrong way to do something, then someone will do it. -- Edward A. Murphy Jr. signature.asc Description: Digital signature
Re: [solved] Re: jigdo various questions ( --noask does not work + how to use "/var/cache/apt/archives/*.deb" as source )
On Wed, Apr 30, 2014 at 12:16:32PM +0200, berenger.mo...@neutralite.org wrote:
> Thanks for your replies.
No problem, glad I could help.
> I guess I'll have to submit some reports, for example that the man
> does not says explicitly the order of the arguments. I'll do some
> other tries before to be sure to not being having missed something.
Oh, something else I should probably have mentioned, since you can pass
multiple URIs to jigdo-lite, instead of using 'jigdo-lite
debian-7.5.0-amd64-DVD-$i.jigdo' and looping through $i, launching a new
jigdo job each iteration, you could use 'jigdo-lite
debian-7.5.0-amd64-DVD-{1,2,3}.jigdo' to get the first three DVD images
in one jigdo job (adjust the numbers to suit).
You can use similar bash tricks (probably other shells have similar
functionality) to get some combination of CD or DVD images of linux or
kfreebsd for whichever architectures they are available in a single
jigdo-lite call.
Cheers,
Tom
--
A little pain never hurt anyone.
signature.asc
Description: Digital signature
Re: doc-linux-text Not in Wheezy?
On Wed, Apr 30, 2014 at 10:41:52AM +1000, Scott Ferguson wrote: > On 29/04/14 23:53, Tom Furie wrote: > > What does apt-cache policy doc-linux-text show? > > doc-linux-text: > > Installed: 2008.08-1 > > Candidate: 2008.08-1 > > Version table: > > *** 2008.08-1 0 > > 100 /var/lib/dpkg/status This shows that the package is not available in any of your current repositories. Doc-linux-text was removed from the unstable and testing repositories in April 2012. > cat /etc/apt/*/*.list | grep -v "^#" > deb http://security.debian.org/ wheezy/updates main > deb http://mirror.aarnet.edu.au/pub/debian wheezy main non-free contrib > deb http://mirror.aarnet.edu.au/pub/debian/ wheezy-backports main contrib > non-free > deb http://dl.google.com/linux/chrome/deb/ stable main > deb http://dl.google.com/linux/earth/deb/ stable main > deb http://dl.google.com/linux/talkplugin/deb/ stable main > deb http://mozilla.debian.net/ wheezy-backports iceweasel-release > deb http://www.deb-multimedia.org wheezy main non-free > deb http://download.virtualbox.org/virtualbox/debian/ wheezy contrib > deb http://packages.x2go.org/debian/ wheezy main > > NOTE: clean build from scratch, not upgraded from Squeeze Wherever the package was installed from, it wasn't any of these repositories. Cheers, Tom -- A New York City ordinance prohibits the shooting of rabbits from the rear of a Third Avenue street car -- if the car is in motion. signature.asc Description: Digital signature
Re: jigdo various questions ( --noask does not work + how to use "/var/cache/apt/archives/*.deb" as source )
On Tue, Apr 29, 2014 at 05:00:03PM +0100, Tom Furie wrote: > The URI needs to be the last element of the command. You will find most > of the required files if you mount the iso and pass the mount point with > --scan, passing the iso file finds 0 files. Using --scan causes jigdo to > not ask for further file locations, and only the last instance of --scan > is used. My tests show a better hit rate using the mounted iso than > using the cache. Ideally there would be some way to pass multiple file > locations on the command line so the iso mount point and the cache could > both be checked. Hmm, I think I feel a wishlist bug report coming along. In the meantime, you could create a directory containing links to the apt cache and to the loop mount point(s) and pass that in --scan. Cheers, Tom -- On a paper submitted by a physicist colleague: "This isn't right. This isn't even wrong." -- Wolfgang Pauli signature.asc Description: Digital signature
Re: jigdo various questions ( --noask does not work + how to use "/var/cache/apt/archives/*.deb" as source )
On Tue, Apr 29, 2014 at 04:27:20PM +0200, berenger.mo...@neutralite.org wrote: > Hello. > > I am trying to update some iso images ( I have downloaded ISOs last > week, but it seems that there were an update since, my images are in > 7.4 and last debian stable is 7.5 ) through jigdo, and wanted to > play a little with its scriptable possibilities, since it seems that > there is a --noask option to avoid asking questions to user. > > But when I am doing this: > $i=2;jigdo-lite > "http://cdimage.debian.org/debian-cd/7.5.0/amd64/jigdo-dvd/debian-7.5.0-amd64-DVD-$i.jigdo"; > --scan "debian-7.4.0-amd64-$i.iso" --noask > > jigdo still asks me questions, so I can not send it through a for > loop and a '&'. > > > Also, since jigdo is able to find files from a distant source, I was > wondering if it is or not possible to retrieve those files from a > local directory? > For example, imagine you have the whole collection of debian's DVDs > images ( so, 10 DVDs ) from last stable version ( 7.4.0 ) that you > were often upgrading. The packages are definitely in your > /var/cache/apt/archive, so it would be interesting to not download > them again ( which is, AFAIK the only behavior of jigdo-lite > currently ). > So, am I missing some option, or is this feature simply not > implemented? If not implemented, is there is a reason, or simply > nobody had the idea/time to do it? The URI needs to be the last element of the command. You will find most of the required files if you mount the iso and pass the mount point with --scan, passing the iso file finds 0 files. Using --scan causes jigdo to not ask for further file locations, and only the last instance of --scan is used. My tests show a better hit rate using the mounted iso than using the cache. Ideally there would be some way to pass multiple file locations on the command line so the iso mount point and the cache could both be checked. Hmm, I think I feel a wishlist bug report coming along. Cheers, Tom -- miracle: an extremely outstanding or unusual event, thing, or accomplishment. -- Webster's Dictionary signature.asc Description: Digital signature
Re: doc-linux-text Not in Wheezy?
On Tue, Apr 29, 2014 at 08:29:45PM +1000, Scott Ferguson wrote: > On 29/04/14 20:22, Paul Lane wrote: > > Searched archive and found nothing regarding any discussion of it > > being removed. I did find it available for Squeeze. However, as I > > stated it is not found in the package lists for Wheezy. Has it been > > renamed, removed, or otherwise squashed? > > > > Wheezy main nonfree contrib (with backports enabled), apt-cache search > results:- > linuxdoc-tools-text - Text output facility of LinuxDoc-Tools > doc-linux-text - Linux HOWTOs and FAQs in ASCII format Interesting. Is that an old squeeze package that's been left installed? What does apt-cache policy doc-linux-text show? I have wheezy with main, contrib and non-free enabled on updates and backports but doc-linux-text doesn't appear here, though there *are* -ja-text and -fr-text packages. apt-cache search doc-linux doc-linux-fr-html - Linux docs in French: HOWTOs, MetaFAQs in HTML format doc-linux-fr-text - Linux docs in French: HOWTOs, MetaFAQs in ASCII format doc-linux-hr - Documentation in Croatian / dokumentacija na hrvatskom doc-linux-ja-html - Linux HOWTOs and FAQs in Japanese (HTML format) doc-linux-ja-text - Linux HOWTOs and FAQs in Japanese (TEXT format) doc-linux-pl - Linux docs in Polish: HOWTO - ascii version doc-linux-pl-html - Linux docs in Polish: HOWTO - html version Cheers, Tom -- Bit off more than my mind could chew, Shower or suicide, what do I do? -- Julie Brown, "Will I Make it Through the Eighties?" signature.asc Description: Digital signature
Re: Systemd
On Fri, Apr 25, 2014 at 09:49:18PM +0200, Ralf Mardorf wrote: > On Fri, 2014-04-25 at 15:12 -0400, Steve Litt wrote: > > Kinda seems like the (de) evolution of cars, doesn't it? As a kid, > > I could tune up my beater flat head 6 1959 Plymouth in 20 minutes with > > a 10 inch adjustable and a gapping tool. With today's cars, I'm lucky > > to be able to change the air filter. > > That's a nice analogy :). We could repair our old cars using a hammer > and screwdriver, today we need to be computer experts to repair a car. I know qualified car mechanics whose toolkit for their own cars is a mobile phone and a credit card. Cheers, Tom -- Everybody needs a little love sometime; stop hacking and fall in love! signature.asc Description: Digital signature
Package dependencies
On Wed, Apr 23, 2014 at 10:38:13AM +1000, Chris Angelico wrote: > Those concepts are fine for concrete packages. My MUD client Gypsum, > if I were to package it as a .deb, would Depend on Pike and GTK, would > Recommend the latest Pike (if it's possible to depend on one version > and recommend another), and probably wouldn't have any Suggests. But > with meta-packages, how does that work? "Standard graphical desktop" > definitely requires some things (like a desktop environment, > obviously), but what about all those pre-installed programs? Are they > Recommends or Suggests? I don't think you can have a Depends on one package version and Recommend another, but you can Depend on a package without any version requirement, or with =, <, >, <=, >= version requirements. The dependency system for meta-packages works the same as it does for real packages. In wheezy, looking at the gnome meta-package, the only package I would say is a genuine Depends: would be gnome-core, everything else should be a Recommends: or less. Instead of having dependencies like "libreoffice-writer | abiword" and "libreoffice-calc | gnumeric" there could be an office-suite virtual package that gnome could recommend or suggest. Once you get to things as vague as "Standard graphical desktop" you can use virtual packages, and you could have the desktop meta-packages have a "Provides: graphical-desktop" or some such (I'm not entirely sure how virtual packages work), but then you might have to be careful about potential toolkit issues. Cheers, Tom -- Trying to establish voice contact ... please yell into keyboard. signature.asc Description: Digital signature
Re: Sudo Autoremove wanting to kill my system?
On Wed, Apr 23, 2014 at 09:37:50AM +1000, Chris Angelico wrote: > On Wed, Apr 23, 2014 at 9:32 AM, Tom Furie wrote: > > In your example above, while gnome depends on openoffice, openoffice > > cannot be removed without also removing gnome and thus anything that was > > pulled in by gnome. However, while it is /nice/ to have an office suite > > in your desktop environment it is by no means a requirement, therefore > > if the gnome->openoffice relationship was downgraded to a Recommends: > > the openoffice package could be removed without any further disruption > > to the system. > > Ah! That might be the solution. I haven't dug into it, but if that's > how the base desktop environment metapackages are done now, that would > cure this issue. You install Debian with whichever desktop, then > uninstall all the games, but you still have a perfectly viable system. Unfortunately that is not how the meta-packages are done now. All that has changed is that instead of dependencies of the meta-packages being marked as automatically installed they are now marked as manually installed, which means when you remove the meta-package everything that it pulled in is left behind. To further use the on-going illustration: if you have gnome installed and it depends on openoffice, when you remove openoffice the gnome meta-package must also be removed, but now all of the other packages are left on the system. This at first glance seems like a good idea as apt didn't "try to remove my whole system", however now all those packages that are dependencies of gnome don't have a meta-package covering them and so to remove them you have to track them individually. The intention of the package relationships was supposed to be - Depends: This package won't run unless you have this other package installed. Recommends: This package will run without this other package, but possibly with reduced functionality. Suggests: This package will run fine without this other package, but the other package would be nice to have along with this. This seems to have been lost somewhere along the way. Cheers, Tom -- I am NOT a kludge! I am a computer! -- tts signature.asc Description: Digital signature
Re: Sudo Autoremove wanting to kill my system?
On Wed, Apr 23, 2014 at 09:05:09AM +1000, Chris Angelico wrote: > On Wed, Apr 23, 2014 at 8:56 AM, Tom Furie wrote: > > This seems an odd choice to make. If I installed a meta-package because > > I couldn't be bothered to investigate which individual packages I > > wanted, or just wanted to explore the whole package, then decide I don't > > want that package, so remove the meta-package, nothing now gets removed? > > It's a bit odd either way. Let's suppose you install Debian Squeeze > (again, I don't know where the policy changed, but I know it was since > Squeeze), and then decide that you don't want (say) Open Office. You > uninstall it, but since there was a metapackage "gnome" (if I have the > name right) that depended on it, apt-get removes that package. And > then everything else that was installed from that metapackage is now > free to be removed, right? Along with everything else that was pulled > in by them. This touches on the dependancy relationships discussion that was below the part quoted. If non-core functionality would be a Recommends: relation rather than a Depends: relation then components of the meta-package could be removed painlessly. In your example above, while gnome depends on openoffice, openoffice cannot be removed without also removing gnome and thus anything that was pulled in by gnome. However, while it is /nice/ to have an office suite in your desktop environment it is by no means a requirement, therefore if the gnome->openoffice relationship was downgraded to a Recommends: the openoffice package could be removed without any further disruption to the system. Cheers, Tom -- Well thaaat's okay. signature.asc Description: Digital signature
Re: Sudo Autoremove wanting to kill my system?
On Wed, Apr 23, 2014 at 01:22:40AM +0300, Andrei POPESCU wrote: > On Ma, 22 apr 14, 22:56:27, Lisi Reisz wrote: > Unfortunately this may not have the desired effect with at least Gnome, > due to circular Depends/Recommends of the installed packages. Besides, > recently[1] meta-packages have been getting special treatment, i.e. > their dependencies get marked as manually installed, in order to prevent > the old "I tried to remove a package and now my entire Desktop > Environment was removed". This seems an odd choice to make. If I installed a meta-package because I couldn't be bothered to investigate which individual packages I wanted, or just wanted to explore the whole package, then decide I don't want that package, so remove the meta-package, nothing now gets removed? > I still don't understand why Developers insist on using Depends instead > of Recommends for meta-packages. Indeed. Anything that isn't core functionality should at most be a Recommends relation and much of what is in a meta-package could probably even be a Suggests relation. > > But I don't quite see why you wanted to remove it. There is nothing > > wrong with having 2 desktops available, unless you have a very small > > hard drive. > > Agreed. If there is a package on my systems which I know nobody uses it gets removed, simply because it doesn't need to be there. But then, I come from an age when every byte mattered and think if people were as frugal now as they were then our modern systems would perform even better than they already do. Cheers, Tom -- A closed mouth gathers no foot. signature.asc Description: Digital signature
Re: Logging of commands in a bash script to a file
On Tue, Apr 22, 2014 at 07:06:41PM -0300, Daniel Bareiro wrote: > I'm writing a bash script that runs several routing commands. I would > like these commands, on a part of the script, plus run, are saved to a > log file. I'm having trouble processing this paragraph. Are you saying that you want to run a script and have the commands within the script (and their output) recorded in another file? Cheers, Tom -- Logicians have but ill defined As rational the human kind. Logic, they say, belongs to man, But let them prove it if they can. -- Oliver Goldsmith signature.asc Description: Digital signature
Re: Spam on the list
On Mon, Apr 21, 2014 at 10:06:00PM -0500, c. marlow wrote: > What the heck > Sorry I'm new to the whole group email / NEWSGROUP thing. It was spam. Sometimes it gets through the filters. Best course of action is to not reply to it, and *never* quote it. Cheers, Tom -- Support your local Search and Rescue unit -- get lost. signature.asc Description: Digital signature
Re: install cinnamon
On Mon, Apr 21, 2014 at 02:43:31PM -0500, c. marlow wrote: > On Mon, 2014-04-21 at 19:50 +0100, Brian wrote: > > You have joined your mail to an existing one and your topic has nothing > > to do with the one you have added your mail to. > > > > This doesn't benefit the person who sent the original mail and it > > doesn't benefit you. > > > > Please submit your mail again as a completely separate message. > I was just curious why that question would not be allowed into the > group? Isnt this group for Debian users that are using plain debian > and not a directive as in Ubuntu, mint? > > Isnt this group for all Desktop Environments Xfce, Lxde, Gnome 3? Brian wasn't commenting about the posts suitability for this mailing list but that the poster had hijacked an existing, unrelated thread, and requested that he start a new thread with his question. Some benefits of starting a new thread are that it will not get buried in the existing thread and it will not pollute the existing thread. Potential respondents to the cinnamon question may have seen the original thread, not been interested in it, killed the thread and so would not see the question that they would be able to help with. Cheers, Tom -- Krogt, n. (chemical symbol: Kr): The metallic silver coating found on fast-food game cards. -- Rich Hall, "Sniglets" signature.asc Description: Digital signature
Re: Heartbleed
On Sat, Apr 19, 2014 at 02:33:43PM +1000, Scott Ferguson wrote: > On 19/04/14 07:55, Joe wrote: > > As is the light originating inside peoples' homes and passing out of > > their windows. In which case it is arguable that it is perfectly > > acceptable to collect and record that light with a camera without > > asking the permission of those who own the home and/or who have > > modified the light... > > Most countries don't provide legislative protection from the gaze of > people passing by. For reasons of sanity, and something to do with the > concept of free will (and personal responsibility). There is a very large difference between the gaze of passers-by and actively attempting to see something, especially where recording equipment is involved. Cheers, Tom -- I'm ANN LANDERS!! I can SHOPLIFT!! signature.asc Description: Digital signature
Re: DynDNS no longer free.
On Tue, Apr 08, 2014 at 07:32:37PM -0400, Brad Alexander wrote: > I apologize. I should have vetted these before posting them. Best as I can > tell, ez-ip, penguinpowered, and hn seem to be gone, dhs, ods, easydns are > no longer free, tzo got acquired by dyndns (thus under the 30 days left > clause), and zoneedit (verisign) is also not free. > > dyns claims to have a free service, but requires a minimum of a 5 Euro > donation. > > So again, I apologize. I thought the ez-ipupdate list would be more current. It certainly seems that free dynamic dns is becoming an endangered species. I think I'll end up either using dnsdynamic or dnspark. Cheers, Tom -- A drama critic is a person who surprises a playwright by informing him what he meant. -- Wilson Mizner signature.asc Description: Digital signature
Re: DynDNS no longer free.
On Tue, Apr 08, 2014 at 07:03:26PM -0400, Brad Alexander wrote: > Over lo these many years, I have run ez-ipupdate on my perimeter to keep my > dynamic hostname in sync. So I pulled up the description, which says: > > " Currently supported are: ez-ip (http://www.EZ-IP.Net/), Penguinpowered > (http://www.penguinpowered.com/), DHS (http://members.dhs.org/), > dynDNS (http://members.dyndns.org/), ODS (http://www.ods.org/), > TZO (http://www.tzo.com/), EasyDNS (http://members.easydns.com/), > Justlinux (http://www.justlinux.com), Dyns (http://www.dyns.cx), > HN (http://dup.hn.org/), ZoneEdit (http://www.zoneedit.com/) and > Hurricane Electric's IPv6 Tunnel Broker (http://ipv6tb.he.net/)." I've been using ddclient for so long I didn't even think to look for alternative options in the Debian repos (shame on me). Thank you for the pointers, I shall go exploring :) Cheers, Tom -- We can predict everything, except the future. signature.asc Description: Digital signature
Re: DynDNS no longer free.
On Tue, Apr 08, 2014 at 03:51:01PM -0500, Hugo Vanwoerkom wrote: > DynDNS just announced that their free hostname program in 30 days > will no longer be gratis. > I use that with ddclient to update the IP address for my blog. > Are there other free alternatives? This news disappointed me too, given that I have been using their service since the available domains were dyndns.org and two others that I don't remember. There are some free alternatives which I am currently investigating. Unfortunately most of those I have looked at seem to require a Windows binary download, or use a web-based interface which would require manual intervention whenever my IP changes (rather defeating the purpose, I think). Cheers, Tom -- But you'll notice Perl has a goto. -- Larry Wall in <199710211624.jaa17...@wall.org> signature.asc Description: Digital signature
Re: missing devel pkgs for emacs build
On Mon, Apr 07, 2014 at 09:19:25PM +0100, Tom Furie wrote: > Ah, I tested with wheezy. I'll load up a jessie vm and test again. In > theory it shouldn't make any difference, but that's the difference > between theory and practice :) With a fresh install of Jessie plus xorg-dev, libgif-dev, libtiff-dev and ncurses-dev (along with anything else they pull in) I can't get the build to fail. The build also completes if I use gtk2 or 3. If you're still having problems I'd suggest a back to basics approach. Can you get emacs to build with 'configure --without-all --without-x'? Cheers, Tom -- Real Programmers don't write in PL/I. PL/I is for programmers who can't decide whether to write in COBOL or FORTRAN. signature.asc Description: Digital signature
Re: missing devel pkgs for emacs build
On Mon, Apr 07, 2014 at 03:25:57PM -0400, Harry Putnam wrote: > Tom Furie writes: > > What version of Debian are you on? > > Much of what I mentioned about falling behind has been corrected in > the course of this problem. > > jessie Ah, I tested with wheezy. I'll load up a jessie vm and test again. In theory it shouldn't make any difference, but that's the difference between theory and practice :) > > Shouldn't be relevant, but is that --prefix path really where you want > > to install emacs to? Also, while it usually works it's not normally > > recommended to build in a sub-dir of the source tree, a better option > > would be for example if your source tree is at src/emacs/trunk to build > > in src/emacs/build. > > Yes that is the path I've set. > > My setup looks like: > > /usr/local/src/vcs/bzr/emacs/trunk > /bzr/test/bin > ... /test/libexec > ... /test/share > ... /test/var > Okay, it just seemed such an unusual install path I thought I'd check. > When I created the bzr setup the download does not include such a > scheme as you suggest. I've always run the build commands from the > level with the INSTALL file. In my bzr download that is located in > 'trunk' > > It appears to be where devs expect one to build from. It has INSTALL, > INSTALL.BZR, autogen.sh, configure etc etc. > > You aren't suggesting the setup I have is the cause of my troubles > right? It may have been a typo in your previous post but your configure invocation began with ../configure which would mean to run configure from the parent of the current directory. If it was a typo and you actually run ./configure... that shouldn't be a problem. If you are building from the source root but running ../configure then there's something very odd going on. > Can you say what is better about the scheme you suggest? The main advantage of building in a tree parallel to the source root over building in the source root itself is that it allows you to have multiple builds with different configuration options or to build for multiple architectures without modifying the source tree. For example, you might want to build one emacs with the gtk toolkit and one with the motif toolkit. In that case you could have your source in emacs/trunk, have build directories at emacs/build-gtk and emacs/build-motif and run ../trunk/configure... from the build dirs with appropriate toolkit options without them impacting on each other. Cheers, Tom -- Where do your SOCKS go when you lose them in th' WASHER? signature.asc Description: Digital signature
Re: missing devel pkgs for emacs build
On Mon, Apr 07, 2014 at 12:35:39PM -0400, Harry Putnam wrote: > One thing should be mentioned... I've allowed my debian system to fall > badly behind in updates and general maintenance. Now face a bit of a > job getting back on track. But doubt that is the cause of my troubles. What version of Debian are you on? > Yes, an X system (Lxde ) > > I followed the usual drill of: > > sh autogen.sh (did all the reconf sort of stuff) > > ../configure --with-xft --with-x-toolkit=gtk >--prefix=/usr/local/src/vcs/bzr/test Shouldn't be relevant, but is that --prefix path really where you want to install emacs to? Also, while it usually works it's not normally recommended to build in a sub-dir of the source tree, a better option would be for example if your source tree is at src/emacs/trunk to build in src/emacs/build. Cheers, Tom -- Humor in the Court: Q: Are you qualified to give a urine sample? A: Yes, I have been since early childhood. signature.asc Description: Digital signature
Re: missing devel pkgs for emacs build
On Sun, Apr 06, 2014 at 10:04:31PM -0400, Harry Putnam wrote: > Tom Furie writes: > > > On Sun, Apr 06, 2014 at 05:12:00PM -0400, Harry Putnam wrote: > > > >> ./configure returns the error below concerning not finding certain > >> pkgs installed concerning X. > >> > >> Tail of output: > >> [...] > >> checking whether gcc understands -MMD -MF... yes > >> checking for long file names... yes > >> checking for X... no > >> checking for X... true > >> configure: error: You seem to be running X, but no X development > >> libraries > >> were found. You should install the relevant development files for X > >> and for the toolkit you want, such as Gtk+ or Motif. Also make > >> sure you have development files for image handling, i.e. > >> tiff, gif, jpeg, png and xpm. > >> [...] > > > > As indicated by the error, you require at a minimum the X development > > packages. Install xorg-dev. You will also need -dev packages for > > whatever toolkit you're using. > > In a previous post... I said it built fine. > > Fine is a bit strong since I am still having problems getting some of > the emacsclient code to compile but does not appear to be related > to xorg at all. It builds without trouble here. Are you building on an X system? Are you passing any arguments to configure? What commands are you using before you get to configure? Cheers, Tom -- /* dbmrefcnt--; */ /* doesn't work, rats */ -- Larry Wall in hash.c from the perl source code signature.asc Description: Digital signature
Re: missing devel pkgs for emacs build
On Sun, Apr 06, 2014 at 05:12:00PM -0400, Harry Putnam wrote: > ./configure returns the error below concerning not finding certain > pkgs installed concerning X. > > Tail of output: > [...] > checking whether gcc understands -MMD -MF... yes > checking for long file names... yes > checking for X... no > checking for X... true > configure: error: You seem to be running X, but no X development libraries > were found. You should install the relevant development files for X > and for the toolkit you want, such as Gtk+ or Motif. Also make > sure you have development files for image handling, i.e. > tiff, gif, jpeg, png and xpm. > [...] As indicated by the error, you require at a minimum the X development packages. Install xorg-dev. You will also need -dev packages for whatever toolkit you're using. Cheers, Tom -- The Wright Bothers weren't the first to fly. They were just the first not to crash. signature.asc Description: Digital signature
Re: How to Switch Between Gnome and KDE
On Sat, Apr 05, 2014 at 08:43:51PM -0700, ray wrote: > Great, thank you. I can now switch. I was hoping to see a change. > For example, I understand the Gnome has the date and menu at the top > and KDE at the bottom of the screen. After switching to KDE and > rebooting, I see a new log in window, but when I get to the desk top, > nothing is different from Gnome. The menus are also the same. That sounds like you have switched your display manager but not your desktop environment. What did you do to change to KDE? When you select an alternative environment to log in to the change should be apparent as soon as you log in, you shouldn't need to reboot. Cheers, Tom -- Debian Hint #34: If you want to track Debian sid and have a small download quota or a really slow connection, check out the debdelta package. signature.asc Description: Digital signature
Re: no eth0 connection-redux
On Sat, Apr 05, 2014 at 08:05:09PM -0400, Stephen Powell wrote: > First of all, you've broken the thread. The thread was broken once > already by the March / April forced break at the end of the month. > You can't help that. But this time, you've broken it yourself by > starting a new thread. Thread wasn't broken here. The References: header in Robert's mail indicates that it was indeed a reply (to his original post on this thread). I see that you have an In-Reply-To: header, but no References: header. I also see that you are using Zimbra, perhaps Zimbra needs to be taught about References: headers. Cheers, Tom -- The Second Law of Thermodynamics: If you think things are in a mess now, just wait! -- Jim Warner signature.asc Description: Digital signature
Re: aptitude search says "package will be installed"
On Sun, Apr 06, 2014 at 12:20:34AM +0300, Dalios wrote:
> Hi all,
>
> A few weeks back while I was still following Stable (wheezy) I used
> gdebi to install a package (minitube if that matters). Later I
> upgraded to Testing (jessie) and now I used the command "apt-get
> purge minitube" to uninstall the package but when I use the command
> "aptitude search minitube" the return i get is:
>
> pi minitube - Native YouTube client
>
> ...
>
> Is this a problem because of using gdebi? Is this because I used
> apt-get to uninstall and aptitude didn't get the memo?
There probably has been some confusion of information between the
combination of package management you've used. I think the easiest way
to clear the problem would be to run aptitude in visual mode, search for
minitube ('/') and set its status to purge ('_'), then use 'g' to
process the changes.
At this point you may find there are other packages that aptitude wants
to add or remove, you can set them to the desired status in the preview
screen.
Cheers,
Tom
--
BOFH excuse #140:
LBNC (luser brain not connected)
signature.asc
Description: Digital signature
Re: How to Switch Between Gnome and KDE
On Sat, Apr 05, 2014 at 02:12:15PM -0700, ray wrote: > > Which desktop manager are you using? Once we know, one of us will be > > able to tell you how to switch. > > I'm sure that you don't need to edit anything. > I appologize as I am new to Linux I may need some help in the > terminology. Please help me understand 'which desktop manager'. > When I said I have been running Gnome on Debian 7.4, I thought Gnome > was the desktop manager. Is there something else that is a desktop > manager? I may be wrong, but I think that Lisi meant to ask what *display manager* you are using. There are several of these available in Debian, but given that you are new to Linux and using Gnome (probably a default install) I would expect the display manager you are using is gdm3. Some other display managers that are available are xdm, kdm, lightdm for example. Cheers, Tom -- An authority is a person who can tell you more about something than you really care to know. signature.asc Description: Digital signature
Re: The quest for a multimonitor screensaver
On Fri, Apr 04, 2014 at 07:05:28PM -0400, Ric Moore wrote: > Uberto Lauri has been a source of great help, And so far I've > gotten this far towards a multiscreen screensaver > by running one directly like this: > ric@iam:/usr/lib/xscreensaver$ ./glschool -geometry 5440x1024+1 > > ...for four monitors. But, it doesn't jump to fullscreen number 1 > monitor (leftmost) , although I can drag the window to the first > monitor and pull on the sides and bottom to fill all the monitors. > It works, so that is sort of a proof of concept. Anyone know how to > achieve the effect desired via command line so I can use something > like xautolock to initiate a full spread across multi-monitors and > have it start in the leftmost monitor? That seems to be the Holy > Grail for the lack of such a feature using Linux on the desktop. > Windows has it, which makes me crazy. :) Ric Where does the glschool appear? I would expect that to give a window 5440 pixels wide, 1024 pixels tall, with the left edge offset by one pixel from the left side of the display area. There is no indication as to where the top of the window would be. A -geometry of $WIDTHx$HEIGHT+0+0 would give a window anchored in the top left of the display area. Since you give a height of 1024 I'll assume you are running screens of 1280x1024, in which case '-geometry 5120x1024+0+0' should give what I think you're looking for. Cheers, Tom -- You mean you don't want to watch WRESTLING from ATLANTA? signature.asc Description: Digital signature
Re: how to change default pager for tab completions from 'more' to 'less'
On Wed, Apr 02, 2014 at 10:40:13PM +0100, Tom Furie wrote: > There are a few ways to achieve your goal, depending on exactly what > your goal is. As far as I'm aware 'less' has a higher priority than > 'more' in the alternatives system, so... Oops. No there aren't. That should teach me to read more carefully before responding, but it probably won't. -- Every word is like an unnecessary stain on silence and nothingness. -- Beckett signature.asc Description: Digital signature
Re: how to change default pager for tab completions from 'more' to 'less'
On Wed, Apr 02, 2014 at 01:34:54PM -0700, tom arnall wrote: > if i use tab completion, and there are a lot of possibilities, 'more' > is used as the default pager to show the list. I want to see the list > with the 'less' pager. There are a few ways to achieve your goal, depending on exactly what your goal is. As far as I'm aware 'less' has a higher priority than 'more' in the alternatives system, so... Step 0: Confirm that less is installed. From here, it depends if you want to make this change just for yourself or system-wide. If you want to make this change local to your account you can set $PAGER to be $(which less) in your login scripts. If you want the change to be system-wide then running 'update-alternatives --config pager' as root will set $PAGER to be whatever you select there wherever it isn't otherwise over-ridden. Cheers, Tom -- Kilroe hic erat! signature.asc Description: Digital signature
loop.de and alice-dsl.de mail problems suggestion
Aimed primarily at Ralf and Hans, but may be of interest to other parties. MX lookups for loop.de and alice-dsl.de both resolve to megamailservers.eu, I would suggest that your mail problems lie there. Cheers, Tom -- Finally, Zippy drives his 1958 RAMBLER METROPOLITAN into the faculty dining room. signature.asc Description: Digital signature
Re: Disabling services
On Tue, Apr 01, 2014 at 09:50:49AM -0400, Roman Gelfand wrote: > Is there a tool that would take down and disable a service based on a > configurable criteria? Depending what your "configurable criteria" are, there are probably several tools that could be used. Cheers, Tom -- Hand, n.: A singular instrument worn at the end of a human arm and commonly thrust into somebody's pocket. -- Ambrose Bierce, "The Devil's Dictionary" signature.asc Description: Digital signature
Re: Hardware failover webserver cluster
On Fri, Mar 28, 2014 at 09:29:23AM +0100, basti wrote: > SRV1 -> Node 1 > client -->|(shared IP) \/ > | /\ > SRV2 -> Node 2 > : > Node n > > Can I use a VPS for SRV1/SRV2 or should I use dedicated hardware? SRV1 & 2 need to be separate hardware, otherwise if the hardware fails you've achieved nothing. Cheers, Tom -- You have the capacity to learn from mistakes. You'll learn a lot today. signature.asc Description: Digital signature
Re: cdimage.debian.org how-to? what gives?
On Thu, Mar 20, 2014 at 11:32:44AM -0500, John Hasler wrote: > pecondon writes: > > I tried to access cdimages.debian.org on it using FireFox, and could > > not. > There is no such site. Try https://www.debian.org/CD/ There is no such site as cdimages.debian.org, but there *is* cdimage.debian.org. However, it is only a redirect to the page linked above. Cheers, Tom -- Boling's postulate: If you're feeling good, don't worry. You'll get over it. signature.asc Description: Digital signature
Re: Great Debian experience
On Thu, Mar 20, 2014 at 09:51:25PM +0700, Ken Heard wrote: > Since reading your post I discovered that the latest kernel now > available is 3.12-0.bpo.1-amd64 which I will now install. There are > other kernels mentioned in wheezy-backports labelled "pae". Since I > don't know what that means I will avoid them. Thanks for the tip. PAE stands for "Physical Address Extension", it's a 32-bit extension to allow access to memory beyond 4gig. Since you are running 64-bit you don't need it anyway. Cheers, Tom -- * Simunye is on a oc3->oc12 simmy: bite me. :) daemon: okay :) signature.asc Description: Digital signature
Re: textstudio installs texlive in jessie
On Mon, Mar 17, 2014 at 02:58:01PM +0100, Oliver Kranz wrote: > Last question is there a way to set this option by default or should i just > set an alias? That's all covered in the apt man pages. Cheers, Tom -- Courage is your greatest present need. signature.asc Description: Digital signature
Re: textstudio installs texlive in jessie
On Mon, Mar 17, 2014 at 01:43:59PM +, Tom Furie wrote: > On Mon, Mar 17, 2014 at 02:21:52PM +0100, Oliver Kranz wrote: > > Why does texstudio as editor need to install latex packages? > Since texstudio is a LaTeX editor it makes sense that it would pull in I forgot to mention the --no-install-recommends option to apt-get if you want to not install anything that the package recommends. Cheers, Tom signature.asc Description: Digital signature
Re: textstudio installs texlive in jessie
On Mon, Mar 17, 2014 at 02:21:52PM +0100, Oliver Kranz wrote: > Why does texstudio as editor need to install latex packages? Since texstudio is a LaTeX editor it makes sense that it would pull in latex packages. The package doesn't 'depend' on latex though, it's only a 'recommends' relationship, so you can install just texstudio without latex. The easiest way to do this is using aptitude's visual mode, just deselect the latex packages in the preview screen. Cheers, Tom -- The only thing cheaper than hardware is talk. signature.asc Description: Digital signature
Re: [Fwd: Re: systemd - boot messages]
On Sat, Mar 15, 2014 at 05:35:18PM +0100, Ralf Mardorf wrote: > I could write what ever I want to send to this list, but as soon > "systemd" is part of the subject, my mails are delayed or won't come > through the list. It isn't only you this happens to. *All* messages with systemd in the subject get delayed by the system. Remember the recent threads? Cheers, Tom -- Rebellion Postponement: The tendency in one's youth to avoid traditionally youthful activities and artistic experiences in order to obtain serious career experience. Sometimes results in the mourning for lost youth at about age thirty, followed by silly haircuts and expensive joke-inducing wardrobes. -- Douglas Coupland, "Generation X: Tales for an Accelerated Culture" signature.asc Description: Digital signature
Re: When fogetting assigned login name rather than password
On Sat, Mar 15, 2014 at 11:25:41AM -0400, Steve Litt of Troubleshooters.Com wrote: > On Sat, 15 Mar 2014 11:01:15 +0000 Tom Furie wrote: > > The classic approach to this problem is to pass 'init=/bin/sh' to > > the kernel. > Do you think this is going to continue working when we switch to > systemd? Absolutely. systemd will replace the current init system but will still require a /sbin/init. By default, once the kernel loads it looks for an executable located at /sbin/init, this can be anything that is executable by the system. By passing 'init=...' to the kernel you override this behaviour and the kernel loads whatever you pass in the parameter. Cheers, Tom -- Don't quit now, we might just as well lock the door and throw away the key. signature.asc Description: Digital signature
Re: When fogetting assigned login name rather than password
On Sat, Mar 15, 2014 at 05:45:14AM -0500, Richard Owlett wrote: > The complete hard drive is wiped at least once a month. I > consistently use my name as the login on the first install of any > series. The login of any subsequent install will be a mnemonic > associated associated with the current experiment. > > I just tried to bring up an install without a GUI. I had forgotten > the login name I had assigned a week ago. I solved the immediate > problem by rebooting to the install on sda1 and using it to examine > /etc/passwd on sda7. On further thinking about this I have some questions. You say "I just tried to bring up an install...", by this do you mean you *have* installed and are trying to log in, or that you are now doing a text mode install but have forgotten the previous username you used so don't know what this username should be? Cheers, Tom -- Without love intelligence is dangerous; without intelligence love is not enough. -- Ashley Montagu signature.asc Description: Digital signature
Re: When fogetting assigned login name rather than password
On Sat, Mar 15, 2014 at 12:21:27PM +, Tom Furie wrote: > It does seem much more likely that Scott is correct and root logins are > only disabled at the graphical login - as Scott says, that is the > default configuration - in which case Richard should be able to > ctrl-alt-Fn to a virtual console, ctrl-alt-bspace to kill the X server > (though I think that's disabled since the last few releases), or boot > into recovery mode. Having gone back to re-read Richard's original post, he does state that he was bringing up an install without GUI. Which poses the question "why not just log in as root to get the user name?" unless root logins are disabled, which leads us back to how he has a root password in that case. Without further input from Richard we have no way of knowing for certain what the exact situation is. Cheers, Tom -- Once upon a time there was a kingdom ruled by a great bear. The peasants were not very rich, and one of the few ways to become at all wealthy was to become a Royal Knight. This required an interview with the bear. If the bear liked you, you were knighted on the spot. If not, the bear would just as likely remove your head with one swat of a paw. However, the family of these unfortunate would-be knights was compensated with a beautiful sheepdog from the royal kennels, which was itself a fairly valuable possession. And the moral of the story is: The mourning after a terrible knight, nothing beats the dog of the bear that hit you. signature.asc Description: Digital signature
Re: When fogetting assigned login name rather than password
On Sat, Mar 15, 2014 at 12:58:59PM +0100, Ralf Mardorf wrote: > Tom is smarter than we are, it's likely that his guess is correct. The > OP confused the term for I am far from it, and Lisi and Scott have both made excellent points that illustrate that. > no root account, but the first user has got sudo admin super cow powers, > with a pure and clean enabled root account. With the information supplied by Richard, I jumped to the conclusion that he had disabled root logins - failing to consider that if such was the case then root wouldn't have a password for him to know. It does seem much more likely that Scott is correct and root logins are only disabled at the graphical login - as Scott says, that is the default configuration - in which case Richard should be able to ctrl-alt-Fn to a virtual console, ctrl-alt-bspace to kill the X server (though I think that's disabled since the last few releases), or boot into recovery mode. Without further input from Richard we have no way of knowing for certain what the exact situation is. Cheers, Tom -- All the men on my staff can type. -- Bella Abzug signature.asc Description: Digital signature
Re: When fogetting assigned login name rather than password
On Sat, Mar 15, 2014 at 12:22:10PM +0100, Ralf Mardorf wrote: > On Sat, 2014-03-15 at 05:45 -0500, Richard Owlett wrote: > > If another OS had not been available but I knew the root > > password, is there some way I could have gained access as root? > > If you remember the root password, than I don't understand your problem. My first instinct would be to suppose that he's disabled root logins. Cheers, Tom -- For every action, there is an equal and opposite criticism. -- Harrison signature.asc Description: Digital signature
Re: When fogetting assigned login name rather than password
On Sat, Mar 15, 2014 at 10:09:33PM +1100, Scott Ferguson wrote: > On 15/03/14 21:45, Richard Owlett wrote: > > If another OS had not been available but I knew the root password, is > > there some way I could have gained access as root? > # passwd `grep 1000 /etc/passwd | cut -d : -f1` That doesn't help him until he's logged into the system. Cheers, Tom -- Love is staying up all night with a sick child, or a healthy adult. signature.asc Description: Digital signature
Re: When fogetting assigned login name rather than password
On Sat, Mar 15, 2014 at 05:45:14AM -0500, Richard Owlett wrote: > If another OS had not been available but I knew the root password, > is there some way I could have gained access as root? The classic approach to this problem is to pass 'init=/bin/sh' to the kernel. The method for doing so depends on which boot manager you might be using. Cheers, Tom -- To be a kind of moral Unix, he touched the hem of Nature's shift. -- Shelley signature.asc Description: Digital signature
Re: Apache default directories
On Fri, Mar 14, 2014 at 10:44:14PM -0400, Jerry Stuckle wrote: > >>On 3/14/2014 9:20 PM, Peter Michaux wrote: > >>>I would expect they go somewhere under /usr/share/. ^ > /usr/lib is not a subdirectory of /usr/lib/cgi-bin. A web user can > access anything in the directory and any subdirectories (based on > system permissions, of course). But the web user cannot access > anything in higher directories. I think you missed an important word I've highlighted above. Cheers, Tom -- Kites rise highest against the wind -- not with it. -- Winston Churchill signature.asc Description: Digital signature
Re: SASL auth failure dovecot/postfix
On Tue, Mar 11, 2014 at 09:43:41PM -0400, Tony Baldwin wrote: > This is what we ended up doing, rolling back to an older version of > mariadb, libmysqlclient18, etc. > > I posted the relevant instructions (where to get the older pkgs, etc.) > earlier. > Both Taz' server and the other are both sending mail again (I'm sending > this message from the office server. Taz rents a VPS on another server > of mine, and we co-admin some sites). Yeah, I saw your response with much more definitive information had arrived between me starting my reply and sending it. Good to know you got the problem solved and what the fix was. Cheers, Tom -- Not only is UNIX dead, it's starting to smell really bad. -- Rob Pike signature.asc Description: Digital signature
Re: A question about Aptitude interactive mode [Solved]
On Tue, Mar 11, 2014 at 07:02:46PM -0600, Paul E Condon wrote: > On 20140311_205250, Tom Furie wrote: > > how often do you see that purple when you aren't in aptitude? If it ^^^ > Every single time I do whatever makes it happen, I cannot look at the > screen after a keystroke and verify my work. Single keystroke errors > are, for me, a serious problem. But outside of aptitude how often do you use the purple? It might be worth just changing the purple to another colour. Cheers, Tom -- Never pay a compliment as if expecting a receipt. signature.asc Description: Digital signature
Re: SASL auth failure dovecot/postfix
On Tue, Mar 11, 2014 at 07:36:55PM +0100, Tazman Deville wrote: > There are few users on either server, and all I've tested are unable to > send mail. > In both servers, there is 1 mail DB for both dovecot and postfix, yes. If both dovecot and postfix are using the same authentication mechanism it's very odd that only one of them works. > tail history.log for one of the servers: > > Start-Date: 2014-03-04 15:30:45 > Upgrade: mariadb-server:i386 (5.5.35+maria-1~wheezy, While exploring this I've discovered that there are no Debian packages of mariadb except in sid. Which repository are you using for these packages, Arch? My best guess about the timing of the failure would be that while mysql/mariadb was upgraded on the 4th, the service wasn't actually restarted until you were having the load issues. > Ah! Perhaps this will be useful. > I just logged in with mutt, received mail, and tried to send one > message, and get this from tail auth.log: > > Mar 11 19:33:32 myownsite postfix/smtpd[32642]: sql plugin Parse the > username t...@liberame.org > Mar 11 19:33:32 myownsite postfix/smtpd[32642]: sql plugin try and > connect to a host > Mar 11 19:33:32 myownsite postfix/smtpd[32642]: sql plugin trying to > open db 'mail' on host '127.0.0.1' > Mar 11 19:33:32 myownsite saslauthd[1850]: PAM unable to > dlopen(pam_mysql.so): /lib/security/pam_mysql.so: symbol > make_scrambled_password, version libmysqlclient_18 not defined in file > libmysqlclient.so.18 with link time reference > Mar 11 19:33:32 myownsite saslauthd[1850]: PAM adding faulty module: > pam_mysql.so > Mar 11 19:33:32 myownsite saslauthd[1850]: DEBUG: auth_pam: > pam_authenticate failed: Module is unknown > Mar 11 19:33:32 myownsite saslauthd[1850]: do_auth : auth > failure: [user=t...@liberame.org] [service=smtp] [realm=liberame.org] > [mech=pam] [reason=PAM auth error] > Mar 11 19:33:37 myownsite mutt: DIGEST-MD5 common mech free > Mar 11 19:33:40 myownsite sudo: tazman : TTY=pts/0 ; PWD=/var/log ; > USER=root ; COMMAND=/usr/bin/tail auth.log > Mar 11 19:33:40 myownsite sudo: pam_unix(sudo:session): session opened > for user root by tazman(uid=0) If pam_mysql support has been dropped in the latest version of mariadb, you could try running an older version which does have support. Alternatively you could find out what mechanisms the new version supports and change your system accordingly. Downgrading packages can be tricky, and since I don't know the dependency tree around mariadb I won't be able to help there. One option would be to remove mariadb and anything that has a hard dependency on it, get the required previous versions from the archives (you probably still have them in /var/cache/apt/), install them by dpkg and set them to hold status. Make sure you have good backups before doing any of that. Cheers, Tom -- You will always get the greatest recognition for the job you least like. signature.asc Description: Digital signature
Re: A question about Aptitude interactive mode [Solved]
On Tue, Mar 11, 2014 at 11:01:27AM -0600, Paul E Condon wrote: > Information in manpage convinced me that the problem is a bug in > Aptitude, and search of bug reports shows that it is already > reported. In bug reports, what I called 'interactive', is referred to > as 'visual'. I'm sure it will be fixed soonish. In the meantime, I'll > use deselect, or apt-get in situations where I can't see important > details because of inappropriate visual rendering in Aptitude. Agreed, it's a bug in aptitude (it seems to work okay if you start a terminal in reverse video mode rather than changing the colours), but how often do you see that purple when you aren't in aptitude? If it isn't often another alternative would be just to change that to something that contrasts better against the black. Cheers, Tom signature.asc Description: Digital signature
Re: SASL auth failure dovecot/postfix
On Tue, Mar 11, 2014 at 04:22:45PM +0100, Tazman Deville wrote:
> They both use mysql (well, mariadb) auth for both dovecot and postfix.
> I have confirmed that I can connect to the DB on either server with the
> mail admin account configured in postfix.
> Yet, I can not send mail.
Do they connect to the same database for authentication against both
dovecot and postfix? Can you connect to the database as a normal user?
How many users are having the problem?
> As far as I recall, postfix was not updated yesterday.
> Nor was mariadb, dovecot, or anything else I can imagine is significant
> here. Aside from mail, the servers have slightly different software
> installed. One received 1 update, the other 3.
> I do not believe they were the same at all (as in the 1 update one the
> first server was not the same as any of the 3 on the other).
It might help to know exactly which packages were updated -
/var/log/apt/{history,term}.log and /var/log/dpkg.log
should have the information.
> Still, they're having apparently identical problems now, and the only
> thing that has changed on either since they were working was running the
> aptitude updates.
> All the logs seem to be telling me is that SASL auth is failing,
> which I know. They do not tell me why or wherein lies the failure.
Does /var/log/auth.log have any further details?
Cheers,
Tom
signature.asc
Description: Digital signature
Re: SASL auth failure dovecot/postfix
On Tue, Mar 11, 2014 at 02:49:34PM +0100, Tazman Deville wrote: > On Tue, Mar 11, 2014 at 02:41:02PM +0100, Tazman Deville wrote: > > both servers show similar in mail.warn: > > Mar 11 13:52:16 myownsite postfix/smtpd[16685]: warning: SASL > > authentication failure: Password verification failed > > Mar 11 13:52:16 myownsite postfix/smtpd[16685]: warning: > > myownsite.me[178.238.226.185]: SASL PLAIN authentication failed: > > authentication failure > Perhaps this is relevant: > telnet localhost 25 > Trying 127.0.0.1... > Connected to localhost.localdomain. > Escape character is '^]'. > 220 myownsite.me ESMTP Postfix (Debian/GNU) > ehlo localhost > 250-myownsite.me > 250-PIPELINING > 250-SIZE 3072 > 250-VRFY > 250-ETRN > 250-STARTTLS > 250-AUTH PLAIN LOGIN > 250-AUTH=PLAIN LOGIN > 250-ENHANCEDSTATUSCODES > 250-8BITMIME > 250 DSN > quit > 221 2.0.0 Bye > > looks okay there, yes? That looks okay in as much as you are able to connect to the SMTP server, however you haven't tried to authenticate with it which is where you are having problems. You were having trouble connecting to dovecot on the evening of the 10th due to the database being overloaded, but that problem resolved itself when the server load came down to manageable levels, and you have said you are having no problem receiving mail. Dovecot is not part of the equation. The log segment above indicates a failure to authenticate with the SMTP server. What backend are you using for that authentication, the same mySQL database? Are both servers using the same backend? Cheers, Tom -- Experience varies directly with equipment ruined. signature.asc Description: Digital signature
Re: feature request for this mailing list
On Mon, Mar 10, 2014 at 05:11:05PM +, Jonathan Dowland wrote:
> On Thu, Mar 06, 2014 at 07:24:36PM +0000, Tom Furie wrote:
> > This is the snippet from my .procmailrc for handling Debian mailing
> > lists. Each list gets sorted into it's own directory.
> >
> > :0
> > * ^List-id:
> > * ^List-id: > $MAILDIR/.Debian.$MATCH/
>
> The only problem with that is that anyone can send you a mail with a
> list-id header containing some encoded chars that you would not want
> procmail to put verbatim into a file or folder name. A slightly safer
> solution is (from http://jmtd.net/log/list_filtering/)
>
> :0
> * ^List-Id:.*debian.*lists\.debian\.org
> {
> :0
> subst=| echo "$MATCH" | sed 's/\([A-Za-z0-9-]\+\)\..*/\1/'
>
> :0
> $MAILDIR/$subst/
> }
>
> (note that my variation also removes dots, you may wish to tweak).
That rule is good, but would result in $subst being debian-user for this
list or debian-security-announce for the DSA list.
My version also removes dots, but allows me to serve by imap all the
debian lists within a sub-heirarchy of Debian without a redundant
duplication of 'debian-'
The first List-id: matches
debian-= 1>.lists.debian.org.
The second matches
debian-= 1 until a '.' is found>
so $MATCH becomes 'user' in debian-user.lists.debian.org, or
'security-announce' in debian-security-announce.lists.debian.org.
The message is then moved into my maildir structure. The dots in the
destination are there to keep my IMAP server happy.
You make a good point about potentially unwanted characters in the
List-id: header, I've now limited the second pattern to alphanumerics
and -.
Cheers,
Tom
--
My opinions may have changed, but not the fact that I am right.
signature.asc
Description: Digital signature
Re: Wifi
On Sat, Mar 08, 2014 at 09:51:52PM -0600, Stan Hoeppner wrote: > On 3/8/2014 2:18 PM, Patrick Alouidor wrote: > > Hello all. I'm not sure if it me but I have a fresh install of Debian 7 > > on laptop Toshiba C-55A5310. and For some reason I cannot enable my wifi > > switch. I have been pressing the F keys but no luck. please This is my > > first Laptop ever and I wanted to put something stable on it and now I > > cannot get my wifi to turn on. My I please get some form of assistance > > on wifi. > You mention a "wifi switch". There is no such thing. The laptop has a > "wireless ethernet adapter" usually of the 802.11 a/b/g/n standard. It > will "connect" to a "wireless router" or "wireless access point". Given the context I would surmise that "wifi switch" means a switch on the laptop to enable/disable the wireless adapter, whether that be an actual switch, button, or key-combo. Cheers, Tom -- "All over the place, from the popular culture to the propaganda system, there is constant pressure to make people feel that they are helpless, that the only role they can have is to ratify decisions and to consume." -- Noam Chomsky signature.asc Description: Digital signature
Re: Can't parse interface line '#015'
On Fri, Mar 07, 2014 at 02:44:42PM +0800, lina wrote: > I have never realized that until you pointed out. Thanks, No problem, I see my good deed for today is done :) BTW, did you delete line 12, the 'allow-hotplug eth0' line? If so, you'll probably want to add it back in, or replace it with an 'auto eth0' line. Otherwise eth0 won't come up automatically (unless you're using Network Manager, then it's anybody's guess though I hear it's better behaved than it once was). Cheers, Tom -- Coach: Can I draw you a beer, Norm? Norm: No, I know what they look like. Just pour me one. -- Cheers, No Help Wanted signature.asc Description: Digital signature
Re: Can't parse interface line '#015'
On Fri, Mar 07, 2014 at 11:53:24AM +1100, Scott Ferguson wrote: > On 07/03/14 01:59, lina wrote: > > Hi, > > > >>From syslog, it shows me: > > > > Mar 6 10:47:05 debian NetworkManager[6729]: Error: Can't parse > > interface line '#015' > > > > # cat -n interfaces > > 1 # This file describes the network interfaces available on your > > system > > 2 # and how to activate them. For more information, see > > interfaces(5). > > 3 > > # nano -c /etc/network/interfaces > will show line numbers for you That's exactly what 'cat -n' does. > I'm guessing you have blank lines in your interfaces file. So open that > file for editing with the above command. You should find you can move > the cursor to line 15 *even* though there's nothing written on it. If so > use the backspace key to delete those lines - until the cursor reaches > the end of line 13 (behind the "p" in "dhcp"). Then press "F2" to save, > followed by "y" to save changes. By standard practice #015 doesn't represent decimal 15, but octal 15, which equates to decimal 13, the very line that was causing the problem. Cheers, Tom -- The more the merrier. -- John Heywood signature.asc Description: Digital signature
Re: Test
On Thu, Mar 06, 2014 at 08:31:23PM +, Tom Furie wrote: > I honestly thought I was joking when I mentioned them being intercepted > en-route. Yep, there's the delay. My confidence is shaken, and my curiosity is piqued. Cheers, Tom -- Leave no stone unturned. -- Euripides signature.asc Description: Digital signature
Re: Test (impact of systemd)
On Thu, Mar 06, 2014 at 08:06:47PM +, Brian wrote: > There is a thread at present on -user: > >https://lists.debian.org/debian-user/2014/03/msg00279.html > > Its Subject: is "Read-only rootfs on systemd" Hmm... I now suspect that any message with systemd in the subject is being delayed by the system. There are messages in "that" thread (not the one you link above) that were not subject to the delay, the difference being their subject header didn't contain systemd. Let's see if this one gets delayed. I honestly thought I was joking when I mentioned them being intercepted en-route. Cheers, Tom -- VMS Beer: Requires minimal user interaction, except for popping the top and sipping. However cans have been known on occasion to explode, or contain extremely un-beer-like contents. signature.asc Description: Digital signature
Re: feature request for this mailing list
On Thu, Mar 06, 2014 at 11:08:03AM -0800, David Guntner wrote: > I use Procmail to do my mail filtering for me. The recipe I use is: > > # Debian list processing > # Look for the list address here and put them in their own file > :0: > * ^TO_ .*@lists.debian.org > $MAILDIR/debian/ This is the snippet from my .procmailrc for handling Debian mailing lists. Each list gets sorted into it's own directory. :0 * ^List-id: * ^List-id: signature.asc Description: Digital signature
Re: Firmware stuff - was systemd troll
On Tue, Mar 04, 2014 at 04:29:52PM +0800, Bret Busby wrote: > The device had available, a software suite, named the Samsung > Unified Print Driver. > > That worked with Debian Linux 5. > > It apparently does not work with Debian Linux 6. > > Something changed from Debian Linux 5, to Debian Linux 6, that > prevented that software that had worked on Debian Linux 5, from > working on Debian Linux 6. Most likely the problem is that the software was linked against a particular version of one or more libraries which are not in versions of Debian greater than 5. If the source was available it could be rebuilt, linking to the later versions of those libraries, most likely working fine possibly with some minor tweaks required. Cheers, Tom -- Hey! now! Come hoy now! Whither do you wander? Up, down, near or far, here, there or yonder? Sharp-ears, Wise-nose, Swish-tail and Bumpkin, White-socks my little lad, and old Fatty Lumpkin! -- J. R. R. Tolkien signature.asc Description: Digital signature
Re: Debian init choices
On Mon, Mar 03, 2014 at 02:53:35AM +0200, Andrei POPESCU wrote: > Actually, you just need to pass the proper 'init=' parameter to the > kernel. That works too, but by default the kernel will look for /sbin/init. For testing purposes passing the parameter would obviously be the safer choice. Cheers, Tom -- The Golden Rule is of no use to you whatever unless you realize it is your move. -- Frank Crane signature.asc Description: Digital signature
Re: Debian init choices
On Sun, Mar 02, 2014 at 06:44:33PM -0500, Steve Litt wrote: > I was just wondering something. How much effort would it take for me, > personally, just me, to make my Debian Stable start all its processes > with DJB's Daemontools. I know Daemontools, I understand it, I know how > to work with it and how to troubleshoot it. What I don't know is: > > 1) How to get the kernel to pass off control to Daemontools? > 2) How to know which processes get run after boot? > 3) Can one start drivers with Daemontools? > 4) Would there be a performance or stability cost to doing this, other >than boot taking longer, which I don't care about that much? You can put any executable binary in /sbin/init and the kernel will run it when it's finished loading. Cheers, Tom -- Love thy neighbor, tune thy piano. signature.asc Description: Digital signature
Re: Test
On Sun, Mar 02, 2014 at 12:24:26PM +, Brian wrote: > The interesting questions involve what is happening on bendel after a > mail is accepted. > > 1. Why is any mail delayed for 15 minutes before onward transmission? Could be any of many reasons. System load, SMTP transmission failure, routing problems... the list goes on. > 2. Why do some users and not others experience this 15 minute delay? I haven't done any analysis, but I'd expect most if not all of us experience delays from time to time. For the case in question, it's not all of Ralf's messages that are being delayed, only those in the systemd thread. In fact, it looks as if *all* messages in that thread were delayed by roughly the same amount of time. My guess is extra processing for particularly suspicious looking messages. Or for the conspiracists out there, perhaps they were intercepted en route... ;) Cheers, Tom -- BOFH excuse #306: CPU-angle has to be adjusted because of vibrations coming from the nearby road signature.asc Description: Digital signature
Re: Test
On Sun, Mar 02, 2014 at 10:06:17AM +0100, Ralf Mardorf wrote: > PS: I'll set up msmtp ASAP and then replace Evolutions SMTP thingy, to > see if I get information about the issue. If you look at the headers on your posts (or anyone else's for that matter), you can track the progress from origin to destination. In this instance the messages were received by the Debian mail system almost immediately you sent them, they then came back out (in my case) fifteen minutes later. Other people may have received their copies earlier or later depending on several factors. Email is not a real-time communication system, sometimes there are delays, or even failures, at any step along the route. Cheers, Tom -- Two heads are better than one. -- John Heywood signature.asc Description: Digital signature
Re: Test
On Sun, Mar 02, 2014 at 01:41:48AM +0100, Ralf Mardorf wrote: > My apologies, I'm aware that no test mails should be send to the Debian > user mailing list. If I reply (several times) I don't come through the > list anymore. Some time ago I needed to reply 3 or 4 times and then one > reply did came through the list. I wonder if mails will come through the > list, when I don't reply, but send a new thread. > > This is an issue I've got with several mailing list when using Evolution > MUA + using my Alice account. > > No postmaster messages, no error messages by the MUA :(. On further inspection, your first reply on that systemd thread arrived here roughly fifteen minutes after you sent it, at about the same time you were submitting your fourth attempt. Cheers, Tom -- The older a man gets, the farther he had to walk to school as a boy. signature.asc Description: Digital signature
Re: Test
On Sun, Mar 02, 2014 at 01:41:48AM +0100, Ralf Mardorf wrote: > My apologies, I'm aware that no test mails should be send to the Debian > user mailing list. If I reply (several times) I don't come through the > list anymore. Some time ago I needed to reply 3 or 4 times and then one > reply did came through the list. I wonder if mails will come through the > list, when I don't reply, but send a new thread. > > This is an issue I've got with several mailing list when using Evolution > MUA + using my Alice account. > > No postmaster messages, no error messages by the MUA :(. There were four identical replies from you on the systemd thread, all within fifteen minutes. Perhaps you just need to be a little more patient, or maybe there's a problem with your evolution settings or your Alice account. Cheers, Tom signature.asc Description: Digital signature
Re: Jessie almost freezes every several minutes
On Sat, Mar 01, 2014 at 06:11:42PM -0500, Gary Dale wrote: > BTW: there is something more than just KDE since I didn't get a > slowdown when I tried it with a new user without having the raid > array /home. The differences besides the raid array are: > - new user configuration with almost no files > - no nfs shares mounted (usually have a dozen shares mounted) > - no icedove accounts (usually have a half-dozen e-mail accounts > checked every 10 minutes) > - no kontact calendar, etc. (but kontact was running). From what you've said so far my suspicions would lie with RAID or NFS. What are the symptoms with a new user on raid /home but no NFS mounts? What are the symptoms with a new user on non-raid /home and NFS mounts? Cheers, Tom -- That does not compute. signature.asc Description: Digital signature
Re: need help on using crontab
On Wed, Feb 19, 2014 at 03:22:08AM -0500, Long Wind wrote: > I find that shutdown can take a time argument > so why do I bother with cron > > Thank Raffaele Morelli and Tom anyway! Cron would be useful if you want to regularly shutdown or reboot the machine on some definable interval without usually having to worry about it. If it's just a one time delayed shutdown, then yes, using the functionality built in to shutdown would be the better approach. Cheers, Tom -- A bachelor is an unaltared male. signature.asc Description: Digital signature
Re: need help on using crontab
On Wed, Feb 19, 2014 at 08:20:50AM +, Dom wrote: > From the original post, Long Wind seems to have used the original > method of creating crontabs: > > crontab > > The usual sequence (on the old Unix systems I used to admin) was: > > crontab -l > mycronfile > vi mycronfile (to edit) > crontab mycronfile > > crontab -e is much easier and safer to use. That's the information I'm trying to get from him (I think Long Wind is male, my apologies to you if you aren't). 'crontab -e' will create a crontab as the user, but crontabs created as a normal user shouldn't have the user field populated. 'crontab -l' will list the contents of the crontab created by 'crontab -e', so if the crontab was created this way (with a username), it will look fine but when cron tries to run the job it will look for a command that is the username with the rest of the line as arguments. If the intention is to run a cronjob as some system user the simplest method is to create a file in /etc/cron.d, which would require the user field to be populated. Cheers, Tom -- BOFH excuse #221: The mainframe needs to rest. It's getting old, you know. signature.asc Description: Digital signature
Re: need help on using crontab
On Wed, Feb 19, 2014 at 12:56:50AM -0500, Long Wind wrote: > I want to shutdown at 5:03 > I check with crontab -l > it seems OK Depending on how you created the file the format may or may not be okay. Did you create the file in /etc/cron.d, or as a user with 'crontab -e'? Given that you say 'crontab -l' seems okay I suspect the latter. Cheers, Tom -- The meek shall inherit the earth -- they are too weak to refuse. signature.asc Description: Digital signature
Re: need help on using crontab
On Wed, Feb 19, 2014 at 12:26:38AM -0500, Long Wind wrote: > I want to shutdown at some time, > so I create a file named cmd with a line below: > > 3 5 * * * root /sbin/shutdown -h now > > I run the command : " crontab cmd" > > but it doesn't shutdown > Why? Where did you create the file? Are you expecting the machine to shutdown when you invoke 'crontab cmd'? Cheers, Tom -- Fakir, n: A psychologist whose charismatic data have inspired almost religious devotion in his followers, even though the sources seem to have shinnied up a rope and vanished. signature.asc Description: Digital signature
Re: PXE install, without internet?
On Tue, Feb 04, 2014 at 05:35:31PM +1100, Scott Ferguson wrote: > >> iface eth0 inet static > >> address 172.16.4.104 > > >> netmask 255.255.248.0 > > Just in case it makes a difference - that's a /29 network > i.e. 192 possible hosts (32 subnets with 6 hosts each) unless that's not Easy mistake to make (I've made it myself more than once), but just for clarification that's a /21 network. Cheers, Tom -- The little pieces of my life I give to you, with love, to make a quilt to keep away the cold. signature.asc Description: Digital signature
Re: permissions: can you force ACL to be effective over unix perms?
On Tue, Jan 14, 2014 at 05:21:18PM -0600, Bob Goldberg wrote: > I have 2 classes of users - SFTP users (customers), and SFTP managers > (company users that manage customer data). > > I want a highly secure and privacy safe SFTP server. But I also want it to > appear to users as simple and easy as possible. All users will access SFTP > only via an SFTP client. > so my wants are: > - sftp access only. (but not to exclude ssh access for linux users). > - sftp users chroot'ed to their home dir, without any added level's of > directory's [beneath home]. > - so users should have "w" access to their home. > - sftp managers should have "w" access to all sftp-users' home dir's. > > what would be the best way to accomplish this? > I don't care how complex the setup/config is - as long as it's as easy, and > idiot-proof for my users as possible. The first thing that springs to mind is to have the home dirs owned by the user, with rwx permission, and group of sftpmanager (for example), with rwx permissions. Have your sftp managers (and only your sftp managers) be members of group sftpmanager. You could add g+s permission so newly created files will have group sftpmanager. You mentioned that the ftp server you are using requires that all directories leading to the home directories be owned by root with no group write permissions. Does that apply even to the user's home itself? Cheers, Tom signature.asc Description: Digital signature
Re: PXE install, without internet?
On Sat, Jan 11, 2014 at 10:40:03AM +, Артур Истомин wrote: > On Sat, Jan 11, 2014 at 01:21:30PM +0530, Anubhav Yadav wrote: > > So is there a way to boot an entire 4gb dvd-iso from a server, so that I > > can install it on PC connected on a network?? > > You can create local mirror of repos. A full mirror might be overkill. Another option could be an apt proxy, such as apt-cacher. Cheers, Tom -- Proper treatment will cure a cold in seven days, but left to itself, a cold will hang on for a week. -- Darrell Huff signature.asc Description: Digital signature
Re: Still bothered by annoying Re: new motherboard now Autorepeat of keys even when not pressed down
On Sun, Jan 05, 2014 at 08:43:23AM -0500, Mitchell Laks wrote: > > I wanted to see if I can rejuvenate this thread or if should I start a new > one. > > My debian stable (now is sid but no new behavior) ps2 keyboards keep > maniacally repeating keys. > > occasionally it seems almost spontaneous but likely i hit one return and it > keeps typing return > until i hit another key then it stops. > > when I try with another ps2 keyboard, same problem > > no problem with a usb keyboard. > > it started when I moved to a new motherboard with 6 core processor. I would first suspect the ps2 port on the motherboard. I was going to suggest trying a USB keyboard through a ps2 adapter, but I'm not sure that would isolate any areas of investigation. Cheers, Tom -- This is not the age of pamphleteers. It is the age of the engineers. The spark-gap is mightier than the pen. Democracy will not be salvaged by men who talk fluently, debate forcefully and quote aptly. -- Lancelot Hogben, Science for the Citizen, 1938 signature.asc Description: Digital signature
Re: [OT] non technical Q: bad or worse. Was: To do the same as Windows safe mode...........
On Thu, Jan 02, 2014 at 12:22:15PM +1100, Scott Ferguson wrote: > Better presupposes good. If something doesn't approach good, e.g. > someone misses the target by 10 metres are they "better" than the person > who missed by 20 metres? Or "less worse". To call the 10 miss "better" > is a version of newspeak. Surely the correct term, rather than "less worse" would be "less bad"? Cheers, Tom -- Stanford women are responsible for the success of many Stanford men: they give them "just one more reason" to stay in and study every night. signature.asc Description: Digital signature
Re: jwm
On Thu, Dec 26, 2013 at 08:24:13AM +0100, Ralf Mardorf wrote: > The package is already installed. (JFTR I installed KDE before I > installed JWM, perhaps it was installed with KDE. The KDE menu is ok.) > > How do I use update-menues and install-menu? I couldn't find a howto. Did you look at /usr/share/doc/menu or the manpages for update-menus and install-menu? Cheers, Tom -- If a man stay away from his wife for seven years, the law presumes the separation to have killed him; yet according to our daily experience, it might well prolong his life. -- Charles Darling, "Scintillae Juris", 1877 signature.asc Description: Digital signature
Re: Error in apt-get install in libstdc++ package
On Sat, Sep 07, 2013 at 04:04:37PM +0530, Balamurugan wrote: > Do you mean there is no such package like libstdc++? That is correct, there is no package "libstdc++" > I have just ran the commands on a fresh install of Debian 7 (Wheezy) 64-bit > system. > > root@debian:/home/user# apt-get install libstdc++ > Reading package lists... Done > Building dependency tree > Reading state information... Done > Note, selecting 'libstdc++6-4.4-doc' for regex 'libstdc+' [ Snipped list of packages ] > Note, selecting 'libstdc++6-4.4-dev' for regex 'libstdc+' As you can see from the above lines, as there is no exact match for "libstdc++", apt-get is using the string as a regex (regular expression, pattern to match against). The result is a list of packages that match the regex. In this case the pattern being matched against is "any string containing libstd, immediately followed by one or more c's". > _libstdc++6 is already the newest version._ Here, apt-get tells you the package libstdc++6, which matches the regex, is already at the latest version. > Some packages could not be installed. This may mean that you have > requested an impossible situation or if you are using the unstable ^ > distribution that some required packages have not yet been created > or been moved out of Incoming. > The following information may help to resolve the situation: > > The following packages have unmet dependencies: > libstdc++6-4.6-dbg : Conflicts: libstdc++6-4.4-dbg but 4.4.7-2 is to be > installed > libstdc++6-4.6-doc : Conflicts: libstdc++6-4.4-doc but 4.4.7-2 is to be > installed > libstdc++6-4.7-dbg : Conflicts: libstdc++6-4.4-dbg but 4.4.7-2 is to be > installed > Conflicts: libstdc++6-4.6-dbg but 4.6.3-14 is to be > installed > libstdc++6-4.7-doc : Conflicts: libstdc++6-4.4-doc but 4.4.7-2 is to be > installed > Conflicts: libstdc++6-4.6-doc but 4.6.3-14 is to be > installed > E: Unable to correct problems, you have held broken packages. As you can see from the above list, you have selected mutually exclusive packages for installation. As apt-get can't resolve the situation automatically, it bails out and requires manual intervention to resolve the conflicts. Cheers, Tom -- Bachelors' wives and old maids' children are always perfect. -- Nicolas Chamfort signature.asc Description: Digital signature
Re: DON'T REPLY TO THIS THREAD: adduser (with crypted/random) password]
On Fri, May 17, 2013 at 04:49:24PM +0200, Ralf Mardorf wrote: > What the f... > > I got this, after sending my reply (to the list only) to the request > from Pol Hallen : I received the same error replying to an unrelated thread, so I don't think this is a problem at Pol's mail host. Cheers, Tom -- DISCLAIMER: Use of this advanced computing technology does not imply an endorsement of Western industrial civilization. signature.asc Description: Digital signature
Re: About installation DVD's
On Fri, May 17, 2013 at 08:06:23PM +0530, Rupesh Reddy wrote: > Sir please explain what's the process going on and why the remaining seven > DVD's are not released yet. If you *really* need DVD images beyond the first three, you can produce them using jigdo. Cheers, Tom -- QOTD: "I'm just a boy named 'su'..." signature.asc Description: Digital signature
Re: Spurious letter appears in OpenOffice printed document
On Tue, May 14, 2013 at 11:03:14AM -0400, George Langford, Sc.D. wrote: > Document prints with a spurious letter in upper left hand corner of image: > > 1. On debian PC running squeeze, only since recent apt-get update. > 2. On WinXPSP3 laptop, the same document prints perfectly with the > same printer. > > Debian PC prints through CUPS interface via hardwired ethernet; > WinXP laptop prints through wireless ethernet. Printer is HP P1505n > laserjet. What result do you get if you print from Windows via cups? Cheers, Tom -- "As an adolescent I aspired to lasting fame, I craved factual certainty, and I thirsted for a meaningful vision of human life -- so I became a scientist. This is like becoming an archbishop so you can meet girls." -- Matt Cartmill signature.asc Description: Digital signature
Re: Easiest way to get libredline-dev for Lenny
On Fri, May 10, 2013 at 08:27:06AM -0500, Martin McCormick wrote: > Is there an archive that I can simply build a > sources.lst file for which then makes aptitude or apt-get work to > pull in the Lenny files? This would be quicker, easier and less > of a possibility for human error on my part as I try to recreate > the past to see if it solves my immediate problem or not. I believe you are looking for archive.debian.org. Cheers, Tom -- He is now rising from affluence to poverty. -- Mark Twain signature.asc Description: Digital signature
Re: The order of my SATA and PATA are switching all the time
On Wed, May 08, 2013 at 03:28:47AM +, T o n g wrote: > Can I use something like root="ID=ata-IBM-DBCA-203240_HP0HPL43952"? I > remember nothing worked well, so I reverted to the (now troublesome) > "safe" /dev/sdXn. I think the idea is to replace /dev/sdXn with e.g. /dev/disk/by-id/ata-IBM-DBCA-203240_HP0HPL43952-part1 Cheers, Tom -- I find this corpse guilty of carrying a concealed weapon and I fine it $40. -- Judge Roy Bean, finding a pistol and $40 on a man he'd just shot. signature.asc Description: Digital signature
Re: Unexpected results attempting to install Squeeze(6.0.5) to USB flash drive
On Tue, Apr 16, 2013 at 02:46:07PM -0500, Richard Owlett wrote: > *QUESTION:* Is there a way to defeat/disable OS_PROBER(sp?) during > installation? Os-prober is at most a 'recommends' of grub. You can quite easily not install it if you wish. I'm not sure if os-prober can be skipped if it's installed. I would expect so but can't offer advice as I don't currently have os-prober installed anywhere. Cheers, Tom -- "Hi, I'm Professor Alan Ginsburg... But you can call me... Captain Toke." -- John Lovitz, as ex-Supreme Court nominee Alan Ginsburg, on SNL signature.asc Description: Digital signature
Re: Console showing control characters for input on console keyboard
On Wed, Mar 20, 2013 at 04:43:53PM -0300, francis picabia wrote: > I am beginning to think this system is glitchy when PS/2 is detached > and reattached. That's entirely possible. PS/2 was never designed to be hot-swappable. Cheers, Tom signature.asc Description: Digital signature
Re: Moving from a proprietary OS - unnecessarily inful experience -- was [Re: I wish to advocate linux]
On Sat, Mar 02, 2013 at 07:38:41PM +1300, Chris Bannister wrote: > These days it is != :) (I think <> was "not equal to", was it?) Technically, it's "less than or greater than", but I suppose it amounts to the same thing :) Cheers, Tom -- I think the world is run by C students. -- Al McGuire signature.asc Description: Digital signature
Re: multiple nic/IP in firewall
On Tue, Jan 22, 2013 at 07:54:25PM -0300, Roberto Scattini wrote: > ~# route -n > Kernel IP routing table > Destination Gateway Genmask Flags Metric RefUse > Iface > XX.220.XX.176 0.0.0.0 255.255.255.255 UH0 00 eth3 > YY.20.YY.0 0.0.0.0 255.255.255.255 UH0 00 eth4 > XX.220.XX.176 0.0.0.0 255.255.255.252 U 0 00 eth3 > 192.168.100.0 0.0.0.0 255.255.255.0 U 0 00 eth2 > YY.20.YY.0 0.0.0.0 255.255.255.0 U 0 00 eth4 > 0.0.0.0 XX.220.XX.177 0.0.0.0 UG0 00 eth3 I just noticed here, that your YY traffic is being routed over your default gateway, that should probably have a next hop specific to that network. You also probably want to remove the YY.20.YY.0 host route. This might not solve the whole problem, but it will get some of it out of the way. Cheers, Tom -- Fess: Well, you must admit there is something innately humorous about a man chasing an invention of his own halfway across the galaxy. Rod:Oh yeah, it's a million yuks, sure. But after all, isn't that the basic difference between robots and humans? Fess: What, the ability to form imaginary constructs? Rod:No, the ability to get hung up on them. -- Christopher Stasheff, "The Warlock in Spite of Himself" signature.asc Description: Digital signature
Re: multiple nic/IP in firewall
On Wed, Jan 23, 2013 at 05:47:02PM -0300, Roberto Scattini wrote: > i also tried a different approach, found somewhere with google, that is > more in line with my understanding of the problem. > basically, it marks the packets so they can be routed back to the same nic > they came in: > > ip route flush table T1 > ip rule del fwmark 101 table T1 > ip route add table T1 default via YY.20.YY.3 > ip rule add fwmark 101 table T1 > ip route flush table T2 > ip rule del fwmark 102 table T2 > ip route add table T2 default via XX.220.XX.178 > ip rule add fwmark 102 table T2 > > # Ensure traffic in one interface goes back out the same interface > iptables -t mangle -F PREROUTING > iptables -t mangle -A PREROUTING -j CONNMARK --restore-mark > iptables -t mangle -A PREROUTING -m mark ! --mark 0 -j ACCEPT > iptables -t mangle -A PREROUTING -i eth4 -m state --state NEW -j MARK > --set-mark 101 > iptables -t mangle -A PREROUTING -i eth3 -m state --state NEW -j MARK > --set-mark 102 Possibly a silly question, but something you might have overloooked - what does your nat table look like? Are you forwarding the traffic from eth4 to your web server? Cheers, Tom -- "Rights" is a fictional abstraction. No one has "Rights", neither machines nor flesh-and-blood. Persons... have opportunities, not rights, which they use or do not use. -- Lazarus Long signature.asc Description: Digital signature
Re: iptables; some IPs are getting through netmasks
On Sun, Dec 23, 2012 at 05:10:45AM -0800, Mark Ford wrote: > I am hoping someone can help show me where I'm going wrong. > I have iptables setup in the following way, basically, I am > using the chain "pests" to drop data from certain IPs. > > > Chain INPUT (policy ACCEPT) > target prot opt source destination > pests tcp -- 0.0.0.0/00.0.0.0/0 > > Chain FORWARD (policy ACCEPT) > target prot opt source destination > > Chain OUTPUT (policy ACCEPT) > target prot opt source destination > > Chain pests (1 references) > target prot opt source destination > DROP tcp -- 1.85.17.0/24 0.0.0.0/0 > DROP tcp -- 67.228.245.0/24 0.0.0.0/0 > > > (in reality, the "pests" chain is much larger) > > It seems to work, except I find emails from certain IPs are > still getting through, and it's always when I have set a /24 netmask. > > For example, in a recent email, the header was... > > Received: from mail10.mydailyflog.com ([67.228.245.121]) Is the above your complete iptables ruleset? Is this ruleset on the mail server in question, or on a seperate box? If on a seperate box, is it acting as a router, are you doing any NAT? Cheers, Tom -- Sic transit gloria Monday! signature.asc Description: Digital signature
Re: Are all files produced by GPL Ghostscript copyrighted by 'Artifex Software, Inc.'?
On Fri, Dec 21, 2012 at 10:09:32PM -0800, Vaibhav Niku wrote: > pdf2ps, which is a frontend to gs, inserts a copyright notice in all PS files > it produces. I am using `GPL Ghostscript 8.71 (2010-02-10)'. Files look like > this: > > %!PS-Adobe-3.0 > ... > %%Creator: GPL Ghostscript 871 (pswrite) > ... > %%BeginProlog > % This copyright applies to everything between here and the %%EndProlog: > % Copyright (C) 2010 Artifex Software, Inc. All rights reserved. > %%BeginResource: procset GS_pswrite_2_0_1001 1.001 0 > ... > %%EndProlog > ... > %%EOF > > (Needless to say, the files are corrupted if you delete the Prolog stuff.) To be fair to Artifex Software, it is only the prolog they are claiming copyright on, none of the actual content. Cheers, Tom -- I brake for chezlogs! signature.asc Description: Digital signature
