Gnome Logout/Shutdown
Hi, I installed the desktop task with tasksel and added the logout applet to the Gnome Desktop. My problem now is that I don't habe the option to shutdown. Any hints which button to press so that I can shutdown from the applet? Thanks Martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] FOSS Marketplace
Hi, On Mon, May 19, 2008 at 5:08 PM, Samuel Bächler [EMAIL PROTECTED] wrote: On the marketplace people could ask: Who can set up a mythTV media center for me? Questions like this aren't asked on lists (e.g. debian-users) which brought myself to the idea that maybe there is a market for such problems. http://rentacoder.com contrary to what the name indicates there are also a couple of jobs there that have exactly the requirements you are suggesting. hth martin (at least it's an example of what it could look like) -- http://www.xing.com/profile/Martin_Marcher You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you.
Re: [OT] sqlite-amalgamation
On Thu, Apr 3, 2008 at 9:19 AM, Amit Uttamchandani [EMAIL PROTECTED] wrote:
Hey everyone,
I am currently evaluating the use of sqlite3 to store application data from
a python application that I am writing.
I am using Debian Etch so some of the packages are out of date and thus I
decided to compile them from source. Python 2.5.2 was fairly straightforward
to compile and install.
From python 2.5 you don't need anything. sqlite is included by
default (note: that has nothing to do with the userspace tools to
access a databasee from the commandline)
~ $ python
Python 2.5.1 (r251:54863, Oct 5 2007, 13:36:32)
[GCC 4.1.3 20070929 (prerelease) (Ubuntu 4.1.2-16ubuntu2)] on linux2
Type help, copyright, credits or license for more information.
import sqlite3 as sqlite
con = sqlite.connect(':memory:')
c = con.cursor()
c.execute(CREATE TABLE test(id INTEGER, message TEXT);)
sqlite3.Cursor object at 0xb7d81f20
c.execute(INSERT INTO test (id, message) VALUES (?, ?), (1, message 1))
sqlite3.Cursor object at 0xb7d81f20
c.execute(INSERT INTO test (id, message) VALUES (?, ?), (1, message 2))
sqlite3.Cursor object at 0xb7d81f20
c.execute(INSERT INTO test (id, message) VALUES (?, ?), (1, message 3))
sqlite3.Cursor object at 0xb7d81f20
result = c.execute(SELECT id, message FROM test)
for row in result:
... print row
...
(1, u'message 1')
(1, u'message 2')
(1, u'message 3')
exit()
--
http://tumblr.marcher.name
https://twitter.com/MartinMarcher
http://www.xing.com/profile/Martin_Marcher
http://www.linkedin.com/in/martinmarcher
You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian is losing its users
On Wed, Mar 26, 2008 at 7:43 AM, Wei Chen [EMAIL PROTECTED] wrote: Hi, The search volume for Debian has been continuously decreasing in the recent years, as shown in the search trend statistics of one of the most famous search engines. This indicates that Debian is losing its users, e.g. about 50% in the last 3 years. Hmm... Subject: Debian considered to have best in-system documtentation The search volume for Debian has been continously decreasing the recent years, as schon in the search trend statistics of one of the most famous search engines. This indicates that the ever increasing quality of documentation leads to more information being available without the need to do extensive searches over the internet. And being able to fix problems with the given documentation without the need to rely on possibly unverified HOWTOs from third parties. regards martin -- http://tumblr.marcher.name https://twitter.com/MartinMarcher http://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
SOLVED: LVM operations block
Hi,
a quick note so that it will get indexed :)
I'm on a plain debian/etch (no backports or other unofficial
packages not originally included in etch)
Whenever I do operations on lvm (pvdisplay, vgdisplay, lvdisplay) it
just sits there and blocks. It does look like it's waiting for some
I/O.
The last lines of strace show this:
stat(/dev/etherd/e0.1, {st_mode=S_IFBLK|0660, st_rdev=makedev(152,
16), ...}) = 0
stat(/dev/etherd/e0.1, {st_mode=S_IFBLK|0660, st_rdev=makedev(152,
16), ...}) = 0
open(/dev/etherd/e0.1, O_RDONLY|O_DIRECT|O_NOATIME) = 3
fstat(3, {st_mode=S_IFBLK|0660, st_rdev=makedev(152, 16), ...}) = 0
ioctl(3, BLKBSZGET, 0x593430) = 0
lseek(3, 0, SEEK_SET) = 0
read(3,
hangs around lazily
/dev/etherd/e0.1 is a deactivated aoe device. It was at some point in
time exported on the server but now it isn't anymore and with plain
etch packages it is not possible to get rid of it.
The way to get around this is to remove such devices from the cache
and use the filter = [] expression in /etc/lvm/lvm.conf
For me the solution was:
in /etc/lvm/lvm.conf:
filter = [ r|/dev/cdrom|, r|/dev/etherd/*| ]
vgscan # to rebuild the cache (which devices should be scanned)
lvdisplay; vgdisplay; pvdisplay
works again as a excpected :)
hope you don't mind that I abuse the list as a note archive for this,
it took me 2 days before I remembered the config file since lvm just
works(tm) under normal circumstances
regards
martin
--
http://tumblr.marcher.name
https://twitter.com/MartinMarcher
http://www.xing.com/profile/Martin_Marcher
http://www.linkedin.com/in/martinmarcher
You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Silent Cron Jobs
Hi, afaik cron (by default) mails all output from a script. If i create a cronjob I usually dump all stdout (just redirect it to /dev/null) But I want to be informed of any errors so I keep stderr. example: # this will get mailed echo My cool cron script # this will not mail stdout, but stderr (see below for an example) echo My cooler cron script /dev/null # this will send you a mail because there was output on stderr, if there was output on stdout it wouldn't mail /nonexistant/echo My failing cron script /dev/null Not this is untested and just a quick writeup but the general rule should apply hth martin On Wed, Mar 26, 2008 at 3:05 PM, Curt Howland [EMAIL PROTECTED] wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi. I tried to whip up a small cron job, I put a short script in /etc/cron.daily thinking that this would work. Well, yes, it works, but I get mail sent to me by cron explaining that the job executed successfully. I'd prefer not to get the mail. I don't get mail for any of the other jobs in cron.daily, and I don't understand enough of bash scripting to see how mine is different from the others. If all else fails I could just add a line to /etc/crontab, but I think the Debian way is so very much better coordinated and elegant. Curt- - -- Treason! http://blog.mises.org/archives/007926.asp -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) iQEVAwUBR+pYQi9Y35yItIgBAQJkcgf/c8y5Sbd4WkNChuA2muNkprTVFy6JkMIs 8gHQReKiEJH6R4QHVTWRtElqjWHDcry15lCV6h2AIN5w+FIPKmvFcViA3rGk5jK3 Jr/NzC3twwtRaxhvUKDNrfr0VHmAjHeVxBdBHt287zejzDc9TCECcPBderco82rO OwmDs7WuNzQrWZSz8VDGFhjxdJrdhUIVzgeSamD0xtt65gNvUj6GN2YxGeUODlTk V9XfO6vxsrK3chBrag7Cz4EA5pPsyK3QoMtr/NrCPSBPbE2dT38hZAtz3dIN4yq1 ve2fmP7HMF24TtKhTKxkmKN6Qm3uOT8B9qiZ7VPrJsvldcSJVAHuAA== =IgSr -END PGP SIGNATURE- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- http://tumblr.marcher.name https://twitter.com/MartinMarcher http://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
[OT] SQL Inline Documentation
Hi, I'm looking for the doxygen equivalent (enhancement) whatever. What it should be able to do is to document: * DDL Statements * Functions * Triggers absolutely cool would be something that generates a graph of dependencies between the tables. Ubercool(TM) if it did an analysis of the triggers and functions and then creates a graph of those also... any suggestions welcome thanks -- http://tumblr.marcher.name https://twitter.com/MartinMarcher http://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: exim/postfix comparisons
Hello, On Mon, Mar 24, 2008 at 2:23 PM, Douglas A. Tutty [EMAIL PROTECTED] wrote: I've heard comments from respected frequent posters on this list both that they use postfix and that debian's decision to have Exim as default was controversial. I also lurk on OpenBSD's misc list. OBSD uses sendmail by default because they have made it secure, it has a BSD license, and it is small so fits nicely in their base install. They didn't use Postfix because of licencing issues but I think the license has changed. Many people who lurk there use postfix on their BSD boxes. They don't use Exim also because of licensing and there have been some disparaging comments on the complexity. hmm I do that too. In fact I'm collecting stuff again after leaving openBSD at 3.5 to get back on track with my knowledge. As for a compelling reason to learn sendmail, the only one I can think of is if you ever want to run a base BSD (I think they all include sendmail in base). Since sendmail was written as the standards for email were being standardized, it can do anything and not just what is current practice. This flexibility makes it very complex; whole (large) books are written on it. In the past, this complexity has left it open to security issues which I think have now all been fixed (until the next time one is found, but then OpenBSD would have to change their front-page brag). Postfix was written by a security guru (I think the person who gave us tcp-wrappers) and is flexible enough to do everything neccessary on a modern network. I guess I'll add myself to the sendmail people then. Can't be that bad and I don't plan to spend vast amounts of time on that, only on a as needed basis :) If it were me and I only knew postfix, I'd find something else to do than learn another MTA. Well sometimes when a client asks you And why exactly do you use postfix you just can't answer Because it's the only tool I know :) -- http://tumblr.marcher.name https://twitter.com/MartinMarcher http://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
exim/postfix comparisons
Hello, I've been a happy user of postfix for a long time but I generally consider it a knowledge lag not to know at least one competing product (which I don't). So it's time to change that and since I use debian I figured it can't be too bad to learn exim, a bit of googling around just brought up comparisons that seem to be very subjective, is anyone aware of some good links that compare feature wise? I think it would be best if it was something like a list of you can't do that with X, do it with Y this way. I'm pretty sure this won't be too long and quite a good jump start to get into exim. Thanks for any pointers. martin PS: if there's a compelling reason to go in the sendmail direction (or any other mta) i'm willing to do that, but I refuse qmail because of licensing issues -- http://tumblr.marcher.name https://twitter.com/MartinMarcher http://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
[OT] searching for crm
Hi, I'm looking for a CRM software, looking around it seems that SugarCRM is the package of choice in this field but it's written in php and I don't do that normally so I'd rather choose a steep learning curve with python than with php. I also know of xrms.sf.net (php again) as an alternative but I recently had to enhance it and the code really seems unmaintainable (mixed PHP/HTML/SQL in a single file), looks like it was created way before anybody thought about MVC or some best practices, yes it does work but I'll probably be enhancing it with some custom stuff where I see fit. my requirements would be: * web based * needs to have workflows (e.g. assign activity to a contact, if no updates within a week - escalate by mail,) * soap/xmlrpc interface (xmlrpc preferred, it just so much nicer to use) * mail gateway (incoming mail should be automagically processed and attached to activities/contacts,) nice to have would be: * written in python (what I sript in most of the time) * xmpp interface (escalation or something, i just like that protocol it seems to be able to fit nearly any use be it human/human messaging, human/app or app/app) * NO: tinyerp.org * NO: erp5.org * I had a look at both of these and they seem to just hide under open source there isn't really that much documentation there (especially erp5.org when i last looked at it), also they seem overkill as I really just need CRM not ERP atm. so does anyone have suggestions about more alternatives to look at? thanks martin -- http://tumblr.marcher.name https://twitter.com/MartinMarcher http://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
[OT maybe]Email Archiving
Hello, I'm looking for a solution for email archiving the priorities I have for it are the following: # reliability # offline backups (that'll probably be scripted) # availability # speed as the least important I guess every decent box that is out there does that, but I'm a bit unsure on the software part- I've been using cyrus in the past but I want this to be as simple as possible. This will really be archiving, so the only operation that will really be done is droping old mails there and searching for them once in a while. First i was thinking about just using cyrus and a few system users (I want as few dependencies as possible) but then again cyrus seems so bloated on the other hand googling for email archiving leads me to appliances with open end pricing which in turn suggests that there's more to it than simply providing storage that is not the main imap server and is just there to search for stuff. I guess I'm totally on the wrong track or just got confused by all the buzzwords certain searches throw at you. I'm open to any suggestion about how you do archiving (buying one of those multi thousand euro appliances is not an option) thanks martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to send mails with attachments for each file in a directory ?
On 2008-01-30 09:26:12, Dan H. wrote: My favorite way is to use find's -printf directive to construct the complete commands and pipe the result to a shell. Has the advantage that you first hack away at your complete find commend and give it a dry run, and if you're happy with what it spits out you just tack the | sh bit to the end. Now, _that_ is a tip i wish I head read earlier, guess it also plays really well with akward input like stuff you need to escape as you can simply look at what it will do before killing a lot of stuff. martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Getting started with Xen -- Xen enabled kernel for Lenny?
Rick Thomas wrote: I sincerely hope that the lack of Xen support in Lenny is a temporary thing that will be fixed before the first Beta release. Does anybody know what's the problem? The problem is that xen is quite behind with kernels, afaik it's a HUGE patch to apply and the most recent version is 18 (or 20 or something, it's been some time) and I gave up following that since the only sane options to get a stable xen host where: a) use distro packages from a stable tree b) use the download from xensource (which I dislike because it pollutes the machine) If I could decide again I'd probably go with VMWare or kvm or vserver which I heard are include in the kernel mainline (well the latter 2 iirc), but xen seems to choose the commercial way and just adds patches based on their business plan. I don't hink it will be included in the mainline kernel any time soon. (Yes bash me, but that is just my personal experience) I may be wrong on vserver, but i read a bit about the kvm stuff and forgot about it since we have no way to switch away from xen due to lack of resources. /martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apache/subversion with ldap
hhding wrote: Thanks for your advice :) Here is the config file, I take your advice and change Require group to Require ldap-group, but it does not work. :: /etc/apache2/mods-enabled/dav_svn.conf :: Location /svn DAV svn SVNPath /var/lib/svn AuthType basic AuthName SVN AuthBasicProvider ldap AuthzLDAPAuthoritative on AuthLDAPURL ldap://localhost:389/dc=my?uid Require ldap-group cn=groupA,dc=my AuthzSVNAccessFile /etc/apache2/dav_svn.authz /Location Looks very much like my config: LocationMatch /(.*) AuthType Basic AuthName Authentication AuthBasicProvider ldap AuthUserFile /dev/null AuthLDAPURL ldap://backend.example.com/dc=openforce,dc=com?uid?sub (objectClass=simplesecurityObject) AuthLDAPRemoteUserIsDn off AuthzLDAPAuthoritative on AuthBasicAuthoritative on AuthLDAPGroupAttribute memberUID AuthLDAPGroupAttributeIsDN off /LocationMatch LocationMatch /maven(.*) Dav On Limit GET HEAD OPTIONS PROPFIND require ldap-group cn=maven,ou=accounts,ou=groups,dc=example,dc=com /Limit LimitExcept GET HEAD OPTIONS PROPFIND require ldap-group cn=developement,ou=accounts,ou=groups,dc=example,dc=com /LimitExcept /LocationMatch the ldap looks fine to me. /martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Getting started with Xen -- Xen enabled kernel for Lenny?
Ted Hilts wrote: Also, I was talking about kernel versions higher than yours (up in the twenties where yours was 18) and 32 bit. But whether the CPU is 32 bit debian stable highest version number is 2.6.18 so there is no way to get a stable distro with a higher version number (stable as in: only packages from the repo) - also I recommend against using packages outside of stable on a server (at least if you don't have a testing lab to verify everything works) on 64bit machines could be a problem, I don't know. Also, the AMD was the only working CPU architecture available on Debian and I don't know no, xen only supports x86, amd isn't the architecture, read below amd64 is for all x86_64 CPUs - afaik it's just a historic naming oddity. I'm running xen as a playground at home too with an intel core 2 duo - no idea about detailed specs just some office pc that was cheap at the time buying it. why that was the case. Many people do not use AMD as their CPU it's just called amd64 you can run it fine on intels 64bit processors. It's historical since the x86_64 was first from amd and only later from intel. Any specific reason you need a newer kernel? (Again) I wouldn't do that on a server machine. Also xen patches are always a couple of versions behind - at least in my experience - so they only apply cleanly to the version stated on xensource architecture. Somewhere, just before etch was declared as STABLE the AMD Xen stuff failed to work properly and this condition was verified by someone (I don't have the name handy) who was doing some kind of liaison between Debian and Xen. That's why I said it did not work on Debian. well stable has always worked for me. sarge didn't have xen so i can't follow the notion of not working installing the original packages from xensource has always worked on any distro (tried it with 3 different ubuntu versions and sarge) This liaison person has already confirmed that and was attempting to find a way around no Debian Xen until the next stable version (which seems to be on its way or is already here). So it seems by my if you really must use xen+debian/oldstable (sarge) I'd go with the official packages from the xen homepage. information the Xen Debian problem occurred on the 2.6 kernel at some point and there were many requests on the debian-user list asking why they could not get the AMD Xen stuff to work. So it will be interesting to see if things have now changed. I don't doubt your set up works and works well but I am willing to bet that the etch stable kernel version etch has been stable since some time now will not work for you. Maybe, with Debian 4.0 the problem has been resolved -- hope so! before etch there was no xen in stable (read: sarge didn't have xen) iirc I did not snip out the rest of the stuff. it already is in the archives, so no reason to -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
pxelinux.0 and bsd.rd
Hi, I'm trying to add the option for installing openBSD to our pxe boot server (debian/etch), I got bsd.rd from a mirror and placed LABEL openbsd_i386_42 KERNEL openbsd/i386/42/bsd.rd in the pxelinux.cfg/default for that all I get is: Invalid or corrupt kernel image ok, next try (found this on a few mailing lists but I have a feeling that's a feature that is not available): LABEL openbsd_i386_42 KERNEL memdiskappend APPEND initrd=openbsd/i386/42/bsd.rd -- I get: Could not find kernel image: memdiskapped (The same is true for memdisk only) Last thing i came up with was: LABEL openbsd_i386_42 KERNEL openbsd/42/i386/floppy42.fs which loads, but then it's just weird colored output. I have a feeling that pxelinux.0 (from debian/etch) is just incapable of booting openbsd kernels. Or is there anything I'm missing in the config? Note I'm not using syslinux I just installed tftp-hpa specified my root dir, put the pxlinux.0 boot.txt pxelinux.cfg/default in it which all works fine for linux installations. I'd like to avoid installing more stuff and keep it simple by just adding the necessary switches to pxelinux.cfg/default if that is possible thanks martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Getting started with Xen -- Xen enabled kernel for Lenny?
Jozef Peterka wrote: Hi all, I might be rushing in to conversation, but I will try to install Debian Etch and make it Dom0 this very weekend. I really look forward to it - although with a little hope to success :) Nevermind, I wanted wish you good luck with xen, and the important is let everybody here know what happened. I will do that after weekend as well, I will post my experience in a short mail to this threat. xen runs fine in etch i have ~15 domUs running on 2 physical machines with ~50LVs attached. not a single problem regarding xen on either of those. hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apache/subversion with ldap
On Thursday 24 January 2008 11:26 hhding wrote: hi, my friends I try to verify access of svn user from apache by openldap. I create userA and userB with schema posixAccount and create groupA with schema posixGroups in openldap and add userA to it. then I add line *Require group cn=groupA,dc=my* to apache's config file. require ldap-gorup cn=. require ldap-user .. below is the version I use : apache2 Version: 2.2.6-3 from what I remember this the debian/etch version right? slapd Version: 2.3.38-1 libapache2-svn 1.4.4dfsg1-1 and attachements are complete config files please don't do attachements. They aren't visible everywhere, for me they just vanished as my primary interface to the ML is gmane.org hope it helps martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apache/subversion with ldap
On Thursday 24 January 2008 21:31 Martin Marcher wrote: On Thursday 24 January 2008 11:26 hhding wrote: and attachements are complete config files please don't do attachements. They aren't visible everywhere, for me they just vanished as my primary interface to the ML is gmane.org now _that_ statemant was absolute b - I was just to dumb to see the attachements. anyway, the ldap-group ldap-user should be the solution, /martin... -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Getting started with Xen -- Xen enabled kernel for Lenny?
On Thursday 24 January 2008 17:13 Ted Hilts wrote: Martin What is the exact kernel version you are running on both machines and what are their CPU designations? The problems in the past have not been with the distribution but with various kernels. Did you compile your systems from source or did you use pre-compiled packages? Did you compile the source and apply a patch for the kernels or did you use kernel binaries with the Xen package already compiled into the kernel binaries? the xen package isn't compiled into the kernel afaik, you still need some userland stuff to start/stop (interface) the domUs. Looking forward to this information. Thanks, Ted aptitude install xen-linx-image-2.6-xen-amd64 no i'm not joking, those with the hypervisor and ioemu and i was set, I had the 2 or 3 minor updates since etch release and all of those kernels worked fine. the dom0 doesn't have anything apart from the official etch repos http://packages.debian.org/etch/linux-image-2.6-xen-amd64 aptitude search ~ixen i A linux-image-2.6-xen-amd64 - Linux kernel 2.6 image on AMD64 i A linux-image-2.6.18-5-xen-amd64 - Linux 2.6.18 image on AMD64 i linux-image-xen-amd64 - Linux kernel image on AMD64 i A linux-modules-2.6.18-5-xen-amd6 - Linux 2.6.18 modules on AMD64 i xen-hypervisor-3.0.3-1-amd64- The Xen Hypervisor on AMD64 i xen-utils-3.0.3-1 - XEN administrative tools i A xen-utils-common- XEN administrative tools - common files First Box (2 cores/1 cpu): cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 39 model name : AMD Opteron(tm) Processor 146 stepping: 1 cpu MHz : 2009.290 cache size : 1024 KB fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu tsc msr pae mce cx8 apic mtrr mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt lm 3dnowext 3dnow pni lahf_lm bogomips: 5024.69 TLB size: 1024 4K pages clflush size: 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: ts fid vid ttp Second Box (4 cores/2 cpus): $ cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 15 model : 65 model name : Dual-Core AMD Opteron(tm) Processor 2212 HE stepping: 2 cpu MHz : 2000.070 cache size : 1024 KB physical id : 0 siblings: 1 core id : 0 cpu cores : 1 fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu tsc msr pae mce cx8 apic mtrr mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm cr8_legacy bogomips: 5002.02 TLB size: 1024 4K pages clflush size: 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: ts fid vid ttp tm stc AGAIN: I didn't install any specific version i think i even used linux-image-xen-amd64 - it just works with etch hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: package list for CLI-only admin/service install
SpamHog wrote: Does anybody keep such a pure Debian CLI tools metapackage or package list or CDD with such a selction of apps? I guess it really depends on your environment, we have this on every host: # more or less standard packages # this is the tasksel standard selection ~pstandard ~prequired ~pimportant # just in case :) less bzip2 # scripting usage python vim # my boss likes it I don't mc # monitoring - very nice zabbix-agent # killall is nice psmisc # networking screen ethtool iproute # we run on xfs xfsprogs xfsdump # misc stuff acl curl subversion # SSH Stuff openssh-client openssh-server i'm also open to suggestions - puppet is next on the list to be added as it can manage all those things centrally hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: find the blocksize of a FS
Ron Johnson wrote:
On 01/13/08 06:37, Martin Marcher wrote:
On Saturday 12 January 2008 23:38 Jan C. Nordholz wrote:
hmm looks like a starting point, I'm trying to get to that info with
python, if all else fails I think the python ctypes module should be able
to get that info thru the C interface.
Well heck, that's easy.
http://docs.python.org/lib/module-statvfs.html
http://docs.python.org/lib/os-file-dir.html#l2h-2700
$ python
Python 2.4.4 (#2, Jan 3 2008, 13:36:28)
[GCC 4.2.3 20071123 (prerelease) (Debian 4.2.2-4)] on linux2
Type help, copyright, credits or license for more information.
import statvfs
import os
os.statvfs('/')[statvfs.F_BSIZE]
4096
os.statvfs('/')[statvfs.F_FRSIZE]
4096
I _really_ hate python for that. I was in the same situation with
shmutil.rmtree which I found after I finished writing my own function for
that.
I think I'm gonna get an ebook that reads the python lib to me every
night :)
martin
--
http://noneisyours.marcher.name
http://feeds.feedburner.com/NoneIsYours
You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: find the blocksize of a FS
On Saturday 12 January 2008 23:38 Jan C. Nordholz wrote:
Hi,
is there a way to get the blocksize of an FS thru /proc or something
else that doesn't rely on the utils (dumpe2fs, xfs_info) to be
installed?
Not that I can see. You can query the disk geometry, but proc is too
low level to know about what filesystems are on the devices.
/proc/filesystems only gives what filesystems the kernel supports.
you could code up a C snippet:
] echo -e '#define PATH_TO_MOUNTED_FS ...
] #include sys/statvfs.h
] #include stdio.h
] int main() { struct statvfs V; statvfs(PATH_TO_MOUNTED_FS, V);
]printf(%u\\n, V.f_bsize); return 0; }' |
] gcc -x c - -o /tmp/fsbsq /tmp/fsbsq
hmm looks like a starting point, I'm trying to get to that info with python,
if all else fails I think the python ctypes module should be able to get
that info thru the C interface.
thanks
--
http://noneisyours.marcher.name
http://feeds.feedburner.com/NoneIsYours
You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Trusted computing [WAS new user question: debian on a Thinkpad T61]
On Saturday 12 January 2008 08:45 David wrote: I'm a member of Al Quaida OMG, everybody RUN! Yes that missquote was on purpose, please read the references before arresting this person...(whoever it may concern...) -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
find the blocksize of a FS
Hi, is there a way to get the blocksize of an FS thru /proc or something else that doesn't rely on the utils (dumpe2fs, xfs_info) to be installed? thanks martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: .bash_profile and .bashrc not executing
On Saturday 12 January 2008 20:50 John Salmon wrote: I'm a new user to Debian Linux. I have the latest version loaded on a dedicated PC with all the default settings. I have added a ~/bin directory to my system. My .bash_profile and .bashrc files were the default files loaded during the install. However, my PATH remains unchanged when I log on even though the .bash_profile file has the lines to add my ~/bin directory. I can make the change manually after I've logged on and can execute files that are in that directory. Also, the aliases set in my .bashrc file don't work. As a check, I've set environment variables in both files and they return null with echo after logging on. I haven't tried re-installing the system from scratch. Any suggestions? you did re-login after the changes did you? (i think bash -l also behaves like a full re-llogin) hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ntpd restart on IP address change the Debian way
Bob wrote:
Sorry to resurrect such an old thread but this is really irritating me,
after reading through the Bug Reports it seems this has been fixed in
version 4.2.4 which is fine for Lenny but I don't want to run Lenny on
my firewall, it's very simple dedicated etch box with nothing other than
main and updates in its /etc/apt/sources.list so I don't want to start
messing around with apt pinning.
Is there a simple way to get the dhcp client to restart ntpd on IP
address change?
Thank you.
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=455717
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=439734
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=231354
I'm on my ubuntu workstation right now, they seem to have incorporated that
thing already (which would work with any demon, i found at my workstation
at home that squid restarts whenever I fire up the VPN to the office):
/etc/dhcp3/dhclient-enter-hooks.d/ntpdate
/etc/dhcp3/dhclient-enter-hooks.d $ cat ntpdate
NTPDATE_CONF=/etc/default/ntpdate
NTPDATE_DHCP_CONF=/etc/default/ntpdate.dhcp
--snip--
ntp_servers_setup_remove() {
rm -f $NTPDATE_DHCP_CONF
}
ntp_servers_setup_add() {
if [ -e $NTPDATE_DHCP_CONF ] [ $new_ntp_servers
= $old_ntp_servers ]; then
return
fi
if [ -z $new_ntp_servers ]; then
ntp_servers_setup_remove
return
fi
tmp=$(mktemp $NTPDATE_DHCP_CONF.XX) || return
chmod --reference=$NTPDATE_CONF $tmp
chown --reference=$NTPDATE_CONF $tmp
(
echo # NTP server entries received from DHCP server
echo NTPSERVERS='$new_ntp_servers'
) $tmp
mv $tmp $NTPDATE_DHCP_CONF
}
ntp_servers_setup() {
case $reason in
BOUND|RENEW|REBIND|REBOOT)
ntp_servers_setup_add
;;
EXPIRE|FAIL)
ntp_servers_setup_remove
;;
esac
}
ntp_servers_setup
--snap--
--
http://noneisyours.marcher.name
http://feeds.feedburner.com/NoneIsYours
You are not free to read this message,
by doing so, you have violated my licence
and are required to urinate publicly. Thank you.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: What is going on in debian-user?
Raquel wrote: I think that this is what bothers me. I'm on at least a dozen lists other than debian-user and debian-isp. However, the 2 Debian lists are the ONLY ones from which come SPAM and viruses. Go figure, debian is supported by volunteers. If you can handle the spam problem better, go ahead sent a mail to the listmasters and offer them your help. martin PS: but I got at least twice as much spam from the OMG there's actually spam threads than I got from the lottery, pr0n and pharmacy guys. I'm thinking about redefining my view on spam... PPS: yes I'm aware of the paradoxon that this mail just adds one more spam message. -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
[OT] Areca 1210 Capacity Expansion
Hello, we own an areca 1210 and currently have a RAID5 with 4x250GB. As disk space is low we'd like to replace it with 4x1000GB. We already found a supported disk, hotpluggable, etc. Now after subsequently replacing the drives how can we expand the RAID set to a net of 3TB (the controller does support sets that large). I scanned the manual and FAQ at areca but couldn't find any pointer on how we would go about that. Anyone been there done that? thank martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] Areca 1210 Capacity Expansion
Rick Thomas wrote: Whatever you decide to do... Do a full backup first! Been there done that. Since that (not so amusing) memory I do have cyclic backups with the option to trigger it whenever needed (like now) :) martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] [SOLVED] Areca 1210 Capacity Expansion
Martin Marcher wrote: we own an areca 1210 and currently have a RAID5 with 4x250GB. As disk space is low we'd like to replace it with 4x1000GB. We already found a supported disk, hotpluggable, etc. Now after subsequently replacing the drives how can we expand the RAID set to a net of 3TB (the controller does support sets that large). I scanned the manual and FAQ at areca but couldn't find any pointer on how we would go about that. Anyone been there done that? To answer myself and for the reference of someone else: You need to update the firmware to at least: V1.43 2007-4-17 Then get the archttp from the areca site (get the newest!) after that you can RESETCAPACITY Raid Set # 00 enter in the Rescue Raid Set screen. It will recognise the new sizes. Proceed as normal (add VolumeSets, expand a volumeset whatever...) hope it saves someone some time... -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Getting System Stats
top free vmstat? you might want to google about monitoring linux in general On 1/9/08, Shane D [EMAIL PROTECTED] wrote: Hey, What would one type in to the shell to get some of the statistics of a machine? I am running asterisk on an old laptop, and I want to try to see if the studder I am recieving is caused by a terrible processor... -- -Shane Blog: http://blind-geek.com/blog/ CoOwner: http://sjtechzone.com AIM: inhaddict Skype: chatter8712 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours You are not free to read this message, by doing so, you have violated my licence and are required to urinate publicly. Thank you. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: galeon and mutt
Hi, On 12/12/07, Gerard Robin [EMAIL PROTECTED] wrote: Hello, with xterm installed in my box, when I clicked on an address e-mail galeon opened mutt in an xterm. Now I have removed xterm and I installed roxterm instead, but when I click on an address e-mail nothing happens. How can I set galeon so that it opens mutt via roxterm ? what does update-alternatives --display x-terminal-emulator you might want to set it to the correct term with update-alternatives --config x-terminal-emulator if you find it points to a wrong term-emulator hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: permissions in general (WAS: Re: permissions in /sbin)
On 12/5/07, Joey Hess [EMAIL PROTECTED] wrote: Martin Marcher wrote: So the user needs to get a precompiled gcc somewhere. Then she would need to get all the header files necessary Then she needs to get the source. Then the quota is full... :) Most systems come with perl. Perl can do anything any non-suid program in /sbin can do. Most systems come with ar, tar, and wget. This can be used to download any .deb and unpack it. The kind of security you're suggesting has hstorically worked miserably, see for example Microsoft Windows, which does not come with a C compiler or many useful programs. /usr/bin/perl /usr/bin/wget /bin/tar exactly my point none of these tools would be accessible in the first place without explicit permission by the sysadmin. And btw. I'm not talking about tools, etc. I see a tendency in systems being more secured with RBAC, MAC, auditing tools, $whatever. But since *nix has a history of being secure because a user/process can't by default destroy any data besides the data one/it owns. Why not take that one further and require explicit permission to even run a program that can potentially destroy data? * Why not take that one further and require explicit permission to run _any_ program? Revoking others access by default does just that. I think my point wasn't clear. -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: permissions in general (WAS: Re: permissions in /sbin)
On 12/5/07, Mike Bird [EMAIL PROTECTED] wrote: I guess it's more a historical reason that others can r+x most of the system but I can see a lot of benefits in denying others by default (of course there's a lot of work involved to migrate from the current permission schema that's at least a serious drawback) There's very little value to blocking read or execute access to executables. A user could compile or download their own executable in their own home directory to do the same job. So the user needs to get a precompiled gcc somewhere. Then she would need to get all the header files necessary Then she needs to get the source. Then the quota is full... :) Instead we control what executables can do, e.g. by limiting which files can be read or written (by any/all executables). are you talking about Role Based access or limiting access to the directories where a user can write to? Why I think it's good to remove others is somewhat the same reasen as why in a firewall ruleset the policy should be drop. You can easily forget to lock down something but if you forget to open it up you can be sure that within an hour users will give you a call (or mail if they can execute the program) and complain... -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
permissions in general (WAS: Re: permissions in /sbin)
Hi, jumping in. On 12/4/07, andy [EMAIL PROTECTED] wrote: ls -l /sbin is all -rwxr-xr-x 1 root root ... I understand this issue. What I don't get is why it seems to be the overall default that others may read and execute files in most cases. To me it would make sense to have something like (very naive right now, hope you get the idea): /bin root:users rwxr-x--- /sbin root:adm rwxr-x--- /usr/bin root:users rwxr-x--- /usr/sbin root:adm rwxr-x--- and so on. Using acl's it would be very easy to add even more groups. I think the explicit adding of others would make a lot of sense and secure the system in a standard way. I guess it's more a historical reason that others can r+x most of the system but I can see a lot of benefits in denying others by default (of course there's a lot of work involved to migrate from the current permission schema that's at least a serious drawback) What do you think? -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: permissions in general (WAS: Re: permissions in /sbin)
Hi, On 12/5/07, Nyizsnyik Ferenc [EMAIL PROTECTED] wrote: On Wed, 5 Dec 2007 16:58:59 +0100 Martin Marcher [EMAIL PROTECTED] wrote: /bin root:users rwxr-x--- /sbin root:adm rwxr-x--- /usr/bin root:users rwxr-x--- /usr/sbin root:adm rwxr-x--- I do get your idea, but have a look at /bin! You will find some very important stuff there, like bash, login and cat, but many more, that every user should be able to use. If a user and or group needs to be able to access stuff from a directory the admin should explicitely allow access. Not rely on that users can do so anyway I also get that you want to enable every user by adding r-x rights to the users group, but there are a few users that are not members of the users group, such as www-data (Apache's user) and postgres. They also need those binaries. While that is true I still think that the added administrational overhead (again: explicit is better then implicit) from man setfacl setfacl -m g:www-data:rx /bin wouldn't that work too? and so on. Using acl's it would be very easy to add even more groups. I think the explicit adding of others would make a lot of sense and secure the system in a standard way. I guess it's more a historical reason that others can r+x most of the system but I can see a lot of benefits in denying others by default (of course there's a lot of work involved to migrate from the current permission schema that's at least a serious drawback) -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: SUDO
2007/12/3, Henning Follmann [EMAIL PROTECTED]: Add erik to the sudores file? visudo is a wrapper around vi to edit the /etc/sudoers file. well not exactly vi. It uses $EDITOR if that can't be found /usr/bin/editor (iirc) which on debian uses the update alternatives system. So there's no need to be afraid of vi/vim when you don't want to use it. Note: I use vi/vim but there may be people that live on the dark side :) -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to mount ext3 so the files belong to a specific user?
2007/11/14, Kent West [EMAIL PROTECTED]: I have a partition that I'm mounting in a specific user's home directory, and want that user to be able to read/write to that partition. However, I've been unable to find (google, man, etc) any way to accomplish this; the few hints I have found indicate it works fine with a VFAT partition, or that you can manually chown the perms after the mount, but it seems crazy to me that you'd not be able to set ownership at mount time. AFAIK, you can't do that with ext2/3/xfs/reiser/... because they store the permissions inside the file system. that means if you mount the filesystem all files belong to specific uid/gid that may or may not match to your system (e.g I'm pretty sure if I chown -R 65434:65434 /mnt/harddisk and give that to you you will just see the number and not any user name) hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: what's your favourite FLOSS?
2007/11/7, John Masters [EMAIL PROTECTED]: Sorry but here's where I unsubscribe for a couple of months while the same stuff is rehashed as it was a few months ago. Does this totally unscientific poll have any real use? Apart that is from reigniting the antediluvian arguments like mutt v emacs, bash v korn etc. It does have a purpose, unregularly just scrolling thru the results may point you to something you may find usefull. How would I use the results from $SCIENTIFIC_POLL - I don't even now where I could get one for free -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Help with Algorythm to test if a progrem is secure,
2007/11/12, Jabka Atu [EMAIL PROTECTED]: -BEGIN PGP SIGNED MESSAGE- 1.Google / IRC for the program name || Security || Trojan || hacks etc .. if found the don't use that program. 2.If possible read source code or give it to some body for finding stuff. 3.Install it on a Virtual Machine (qemu) on a pc without network connection (physical). Test for new open ports and Google for them. Test for pending connection (netstat -a 127.0.0.1) 4.Connect the Virtual machine to a small LAN (with a LOT of domains ) search for iptables or other related strange logs. make some testes (Are there strange DNS quarries ? ) 5.install and connect the pc to a real network. Why do you use a blacklist approach? 1. Find an arbitrary program 2. Go to the (hopefully) limited number of sources you trust 3. Ask/Query them about the program 4. If it's whitelisted (known to be good) use it 5. otherwise drop it... If you have the skills and time: 1. see above 2. examine source 3. if good: package it send it to other skilled people for examination (with source/link of course) 4. add to your whitelist so that other people can refer to you -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mutt and IMAP accounts
and now for the rest of the world :) -- Forwarded message -- From: Martin Marcher [EMAIL PROTECTED] Date: 29.10.2007 08:47 Subject: Re: mutt and IMAP accounts To: Kevin Mark [EMAIL PROTECTED], debian-user@lists.debian.org 2007/10/29, Kevin Mark [EMAIL PROTECTED]: On Sun, Oct 28, 2007 at 09:34:14PM -0700, Richard Otte wrote: I am trying to set up mutt to access IMAP accounts. I can access them, but I want to be able to move to the available mailboxes when I type c (and it says to type '?' to see the mailboxes and then select a mailbox. I can see the mailboxes, but if I choose one, I get a line that says: 1 IMAP and if I click on that, I simply get the list of mailboxes again; I can never open a mailbox. If I eventually type 'y', I go back to the main mailbox. I can move to the mailboxes if I type 'c' and then type = name of the mailbox. see the check_subscribed option in the manual (imap_check_subscribed, imap_list_subscribed may also be what you want) less /usr/share/doc/mutt/manual.txt.gz hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mutt and IMAP accounts
2007/10/29, Kevin Mark [EMAIL PROTECTED]: On Sun, Oct 28, 2007 at 09:34:14PM -0700, Richard Otte wrote: I am trying to set up mutt to access IMAP accounts. I can access them, but I want to be able to move to the available mailboxes when I type c (and it says to type '?' to see the mailboxes and then select a mailbox. I can see the mailboxes, but if I choose one, I get a line that says: 1 IMAP and if I click on that, I simply get the list of mailboxes again; I can never open a mailbox. If I eventually type 'y', I go back to the main mailbox. I can move to the mailboxes if I type 'c' and then type = name of the mailbox. see the check_subscribed option in the manual (imap_check_subscribed, imap_list_subscribed may also be what you want) less /usr/share/doc/mutt/manual.txt.gz hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Etch and Audio CDs
2007/10/23, Michelle Konzack [EMAIL PROTECTED]: Am 2007-10-23 16:41:27, schrieb Martin Marcher: Maybe you can bug your vendor to get the money back? No, since in Germany and in France they must write ON THE CD cover that the CD is copy protected. I found that a lot of people either don't know about that or just want to keep customers. So TRUE you don't have any legal reason that allows you get your money back. But on the other hand bugging your vendor enough with questions and explanations will make him think twice about loosing a customer (I tend to avoid large stores and go to a vendor I trust (in terms of that her recommendations are good, not in terms of privacy in the first place) martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: wireles /wpa_supplicant - where to start?
2007/10/25, Tyler Smith [EMAIL PROTECTED]: Hi, I'm trying to set myself up on my University's wireless network. I can successfully connect to my home wireless using: in /etc/network/interfaces my eth0 looks like this: iface eth0 inet dhcp wpa-driver wext wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf this make ifup and ifdown work as expected - bring up wpa_supplicant and then the interface. hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: wireles /wpa_supplicant - where to start?
Forgot :)
to me the wpa config file looks ok.
this is what I have there:
# WPA-PSK/TKIP
ctrl_interface=/var/run/wpa_supplicant
network={
ssid=the_essid
#key_mgmt=WPA-PSK
#proto=WPA
#pairwise=TKIP
#group=TKIP
scan_ssid=1
psk=geheim
}
--EOF
please note that I have commented out most of the TKIP, AES stuff as
wpa_supplicant usually is capable of finding it by itself, also in
/usr/share/doc/wpa_supplicant/examples there's a catch-all file IIRC
that just tries everything (radius as well as non radius)
2007/10/25, Martin Marcher [EMAIL PROTECTED]:
2007/10/25, Tyler Smith [EMAIL PROTECTED]:
Hi,
I'm trying to set myself up on my University's wireless network. I can
successfully connect to my home wireless using:
in /etc/network/interfaces my eth0 looks like this:
iface eth0 inet dhcp
wpa-driver wext
wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf
this make ifup and ifdown work as expected - bring up wpa_supplicant
and then the interface.
hth
martin
--
http://noneisyours.marcher.name
http://feeds.feedburner.com/NoneIsYours
--
http://noneisyours.marcher.name
http://feeds.feedburner.com/NoneIsYours
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: cobol compiler/gui dev enviroment
2007/10/24, Ron Johnson [EMAIL PROTECTED]: On 10/23/07 22:11, Douglas A. Tutty wrote: I'll bite: why do you need COBOL? You've go C, Fortran, and Ada in Because apps written in COBOL are highly portable across platforms. Are there any sarcasm tags missing? So it is highly portable, yet you have to search around for a compiler I don't get it. martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sharing CD-RW from Linux to Win2k
2007/10/24, [EMAIL PROTECTED] [EMAIL PROTECTED]: I have a CD-RW drive attached to my Linux server and I was wondering if it was at all possible to be able to access this drive (for both reading and writing, though writing is what's important) from my Win2k box. If it is possible, how would I go about doing it? All the information I can find online is either Win-Win or Win-Linux. Any ideas? I know there are a few packages that let you access your writer thru a webinterface. I guess setting up a proper share and limiting the max usage (with quotas) to 800MB should do fine (google cd writer webinterface yielded this as the 3rd link: http://freshmeat.net/projects/webcdwriter/?branch_id=12597release_id=132088) I've never done that but used such a setup at a friend's place. hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Etch and Audio CDs
2007/10/23, Michelle Konzack [EMAIL PROTECTED]: Am 2007-10-19 10:36:02, schrieb Todd A. Jacobs: I'm having a problem under Etch reading the last track on some CDs. I've googled around enough to know I'm not the only one having the problem, but can't find a solution. This is NOT AN ERROR on your CD(s). Don't know about your situation but here in .at a CD is only allowed to carry the CD logo if and only if it actually is a CD. Interestingly enough there are quite a couple of high quality CD players that will choke on those discs (which actually aren't a CD because of that copy protection). Maybe you can bug your vendor to get the money back? /martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: XEN: xen-create-image with more than one disk
2007/10/23, Klaus Rödel [EMAIL PROTECTED]: Hello, i'm a xen newby and want to create a xen image with more than one disk but i found nothing about this? is it actually possible? Here's an example config from a multi disk xen image with the partitions coming from LVM: --snip-- kernel = /kernelImage memory = 512 name = com.example.shares vif = ['mac=00:16:3e:01:00:17,bridge=lanbr0', 'mac=00:16:3e:01:00:18,bridge=xenbr0'] disk = ['phy:vg01/com.example.shares.root,sda1,w', \ 'phy:vg01/com.example.shares.usr,sda2,w', \ 'phy:vg01/com.example.shares.var,sda3,w', \ 'phy:vg01/com.example.shares.home,sda4,w', \ 'phy:vg01/com.example.shares.scratch,sda5,w', \ 'phy:vg01/com.example.shares.backup,sda6,w', \ 'phy:vg01/com.example.shares.data,sda7,w', \ 'phy:vg01/com.example.shares.holdingdisk,sda8,w'] #'phy:vg01/com.example.shares.music,sda9,w'] root = /dev/sda1 ro --snap-- so yes it can be done -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours
Re: raid1 mdadm v's lvm
Hi, 2007/10/18, Andy Smith [EMAIL PROTECTED]: Hi Alex, On Wed, Oct 17, 2007 at 08:53:09PM +1000, Alex Samad wrote: Interesting, I have a habit though of keeping root out of LVM, very easy to get access to root in emergency when its a raid1 parition Agree. Only partly, with a somewhat recent boot CD you won't have any problems mounting LVM on RAID or doing the necessary recovery tasks. The debian Etch (even netinst iirc) has all the necessary tools. However, I personally use a much smaller root, say 1G or less, and then have /usr, /var (and possibly some others depending on the purpose of the machine) inside LVM. For that reason i keep as much as possible in LVM. Maintenance just is easier when you find you ran out of space on some partition and simply can lvresize it. But I guess that is just personal preference I have avoided LVM mirroring because as far as I am aware the machine would not come up entirely without human intervention if a drive would be lost - please correct me if I am wrong there.. IMHO lvm mirroring is useless. if I do LVM (on servers) I have multiple drives and i tend to trust mdadm more in this field than LVM (don't ask me why - I couldn't logically say why). martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: raid1 mdadm v's lvm
2007/10/18, Alex Samad [EMAIL PROTECTED]: On Thu, Oct 18, 2007 at 01:39:21PM +0200, Martin Marcher wrote: IMHO lvm mirroring is useless. if I do LVM (on servers) I have multiple drives and i tend to trust mdadm more in this field than LVM (don't ask me why - I couldn't logically say why). it would be good to here from somebody that uses lvm mirroring. I would like to access some of that extra space that would be made available if I did mirroring at the lvm level instead of the md level. wouldn't that defeat the purpose of mirror? I think if I do mirror my data I _want_ it to be redundantent (how do you spell that?). I wouldn't want to risk something by gaining a few % of space, rather I'd like to be assured that if a disk dies I can simply replace it and LVM will not even take care about that (hotplug with mdadm) - on the other hand I never looked into mirroring LVM because mdadm is just made for that. Correct me if I'm wrong. martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: raid1 mdadm v's lvm
Hi, 2007/10/17, Alex Samad [EMAIL PROTECTED]: 4 partitions to make up 4 md's md0 /boot ( ~ 500M - I like have space here) sda1 + sdb1 md1 / (~10G) sda2 + sdb2 md2 swap sda3 + sdb3 md3 LVM PV sda4 + sdb4 from the PV I make a VG and then lots of LV's I have used this sort of set for a while, but what I would like to do is not make md3 and instead use sda4 + sdb4 as 2 PV and then I get the chance to decide which pv is made up of mirrored space and which is not. personally I'd either use option A md0 /boot md1 sda + sdb (LVM on top of that) option B (data and system separately) md0 /boot md1 sda5 + sdb5 (logical partitions here, I just like them better) md2 sda6 + sdb6 (again) (for testing) md3 sda7 + sdb7 (about 100 - 500 MB to test raid expansion, network block devices, ata over ethernet, iscsi. - you could do that with option A too) md1 -- vg /system md2 -- vg /data md3 -- vg /playground I use system and data seperately because If I'm ever going to grow the RAID or add another disk i can (hopefully) move/backup the data LVs to the LVs on system for intermediate backup (that is if I don't prepare things and have an offline media at hand) Pro: you can decide how much mirroring, etc, yadda, yadda you want for the data VG (either thru LVM or mdadm or some other means) also you can later on add - as said above - nbd's iscsi devices and some such to /playground for experiments (which I found no matter how secure/stable I want a server to be happens in the lifetime - I'm unfortunately not in the financial position to have redundant systems available) hth martin PS: I wouldn't use LVM mirroring - stay for the partion layout with LVM. stay for data security with mdadm. AND DO MAKE BACKUPS :) -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: SSH Question
Hi, 2007/10/11, Telly Williams [EMAIL PROTECTED]: Hi, I see some of you talking about SSHing into your computer from another. What if the computer you're using isn't Linux/Unix? if you talk about the target computer being a windows host then cygwin has an ssh daemon, personally I use rdesktop for windows as the shell is pretty useless on windows imho. if you talk about another machine being windows and your home machine - being remote and the target google for putty hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mdadm, partitioned array
Hi, 2007/10/9, [EMAIL PROTECTED] [EMAIL PROTECTED]: Adding some more info: Interesting results. I wonder if anyone else has a partitioned raid 5 array and have normal speed results? I do. Well sort of. I don't partition RAIDs but I put LVM on top of it. I couldn't tell any difference between the logical volumes in term of r/w speed. That has always worked for me. Out of interested: Why are you partitioning your RAID and not using LVM? thanks martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to set iceweasel as the application to open http hyper links in icedove?
Hi, 2007/10/6, Michael Yang [EMAIL PROTECTED]: When I click a hyper link in icedove, I'd like one browser, for example, iceweasel, to open this link. I didn't find any where within icedove to set this option. Could anyone point it out? That is not an option in iceweasel it is related to the alternatives system in debian. To do it properly have a look at man update-alternatives The easy way would be to just let /etc/alternatives/x-www-browser (something like that) point to the correct browser (your iceweasel binary) hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
PXE booting over VPN
Hello, I'm going to deploy several soekris boxes on different locations. Now those nifty things can PXE boot and I was thinking about how to update them. I think the easiest would be to PXE boot them from some server located remotely and trigger something that let's the box reboot when I need to update it, the problem now is how I can establish a VPN tunnel etc to do that. Now I have now idea how to actually do that. I need a VPN tunnel to reach the boot server but for that the machine has to have VPN connectivity which is only available after it's booted. Chicken egg problem :/. any recipes how to achieve that with debian are welcome. thanks martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Installing Debian from a Lan
Hi, 2007/10/3, [EMAIL PROTECTED] [EMAIL PROTECTED]: I am hoping to install Debian over a lan. I know this is possible but I haven't found much information on it. Suggestions? Could I set up the server to be 64bit but provide the packages for the 386 install? That shouldn't be much of a problem. Is the install manual on debian.org[1] unclear? Also have a look ant section 4.6 (it's linked from the page i gave you [1] http://www.debian.org/releases/stable/i386/ch05s01.html.en#boot-tftp -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: soekris net5501 and vpn14[01]1
Apologies, what I meant to say was I'm not building a standard router. I can't go with distros like pebble, shorewall or ipcop. They don't have package management as debian has it or upgrade mechanisms... The routing i'm trying to achieve will be more like streamlined multiplexing of traffic (don't have the correct terminology yet, it will be something that audits traffic and logs it to different target, imho quite specialized so I need to have a full distro as I probably need to create some interfaces for easy maintenance, nagios, snmp, webinterface for configuration etc) 2007/9/29, Martin Marcher [EMAIL PROTECTED]: 2007/9/29, Kevin Mark [EMAIL PROTECTED]: On Sat, Sep 29, 2007 at 12:31:06AM +0200, Martin Marcher wrote: Hello, Is the mentioned vpn card[1] ready to use in the stock debian kernel, and (this is something in found in several openbsd mailing lists) will openssl/gnutls use it's capabilities or would that be lost money? thanks for your input [1] http://www.soekris.com/vpn1401.htm http://pyramid.metrix.net/trac/wiki/FrequentlyAskedQuestions Interesting distribution, but neither the soekris net5501 nor any of the vpn cards are mentioned there. Also the task of this distribution isn't in any way related to what I'd like to build. I'm not working on a router, or borderline box for a network. Which of the FAQ point should be related to my questions? thanks martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: soekris net5501 and vpn14[01]1
2007/9/29, Kevin Mark [EMAIL PROTECTED]: On Sat, Sep 29, 2007 at 12:31:06AM +0200, Martin Marcher wrote: Hello, Is the mentioned vpn card[1] ready to use in the stock debian kernel, and (this is something in found in several openbsd mailing lists) will openssl/gnutls use it's capabilities or would that be lost money? thanks for your input [1] http://www.soekris.com/vpn1401.htm http://pyramid.metrix.net/trac/wiki/FrequentlyAskedQuestions Interesting distribution, but neither the soekris net5501 nor any of the vpn cards are mentioned there. Also the task of this distribution isn't in any way related to what I'd like to build. I'm not working on a router, or borderline box for a network. Which of the FAQ point should be related to my questions? thanks martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: syslog recommendations?
2007/9/29, s. keeling [EMAIL PROTECTED]: Michael Shuler [EMAIL PROTECTED]: On 09/28/2007 11:29 PM, Rogelio Bastardo wrote: I'm looking for a few F/OSS syslog programs -- one easy to use (sort of like Kiwi syslog) and another that's much more scalable and would let me, say, aggregate logs from lots of different boxes and maybe even do other sorts of cool things (find patterns, etc). rsyslog may be what you are searching for -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
soekris net5501 and vpn14[01]1
Hello, I'm looking for a router, VPN server box. The soekris hardware[] seems to be exactly what I want. Seems it would work fine even withoug the vpn card. Initial googling showed me a vast amount of working *bsd installations but no clear answer on the linux part. I'd rather have debian on this box as I'm more used to the OS and it's configuration quirks (some are always there :)). Is the mentioned vpn card[1] ready to use in the stock debian kernel, and (this is something in found in several openbsd mailing lists) will openssl/gnutls use it's capabilities or would that be lost money? thanks for your input [0] http://www.soekris.com/index.htm [1] http://www.soekris.com/vpn1401.htm -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debian may lose a user
Hello,
I'm interested in the job offer you posted on
[EMAIL PROTECTED] I have several years of experience in
Desktop and Server systems with debian and other linux distributions.
I charge by the hour, every started hour is normally EUR 50. Contact
me privately if you are interested.
I also do have some offers if you expect a larger volume of support
calls. If you'd like to receive more info on that just add a note.
Best regards
martin
2007/9/25, Mike McCarty [EMAIL PROTECTED]:
I have some feedback about my GF who uses Debian at my suggestion.
I have no irons in the fire on this one, as I don't use Debian,
though I do administer her machine for her. So, please don't take
this as a complaint from me, as it isn't. I'm simply informing
the Debian forum of a situation.
She's had four problems with using Debian on her machine,
and support response from this forum has been somewhat less
than she had hoped for. Of the four problems, one I was able
to fix up somewhat by cooking up a printer description file
for her new printer, which now works in a limited sense. One
of them we have a work around, though it isn't pleasant, and
requires me to do some physical recabling of the machine.
The other two remain completely unfixed.
I used the official reporting tool on one of the problems,
and we were not even accorded the courtesy of a response
indicating that the report had been received and was going
to be acted upon. The tool did confirm that a report had
been made, but that was all. I've seen no indication from
Debian that any progress has been made.
At one point, another fellow contacted me stating that
one of the unsolved problems had also bitten him, and
wanted to know what progress or solution eventually came
out. I regretfully responded that there was, AFAIK, no
solution, and she simply lives with the fact that Debian
cannot do what she wants at all.
Anyway, she bought a copy of Windows XP a few weeks ago, and I'm
pretty sure she intends to install this weekend, since she
sent me an e-mail showing that she purchased a copy of F-Prot
for Windows. This would be a heads up for me, indicating
what might be on the honey do list for this weekend.
I provide this only to let you know that it looks like Debian
is going to lose a user to Windows shortly, due to perceived
lack of concern over user's difficulties shown by those who do
support for Debian. I have gently nudged her in the direction of
sticking with it a little longer, and so due to my reluctance to
kill Debian she has. But things seemingly have just gone on too
long.
Anyway, that's it, FWIW.
Mike
--
p=p=%c%s%c;main(){printf(p,34,p,34);};main(){printf(p,34,p,34);}
Oppose globalization and One World Governments like the UN.
This message made from 100% recycled bits.
You have found the bank of Larn.
I can explain it for you, but I can't understand it for you.
I speak only for myself, and I am unanimous in that!
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
--
http://noneisyours.marcher.name
http://feeds.feedburner.com/NoneIsYours
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How do I know debian has detected all my hardware?
2007/9/25, Amit Uttamchandani [EMAIL PROTECTED]: Thank you for the suggestion. This is definitely the tool that I was looking for! I am still curious, however, if this actually shows all my hardware? I guess the best comparison would be to Windows Device Manager where it shows you all hardware associated with your system and a question mark on those that it does not have drivers for. funny i do it just the other way around. If I'm forced to install windows and something doesn't work I check back with a live cd. apart from that I'd say open up the manual that came with your hardware and check if all the devices that are listed work. If so than you know that everything works. hth martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: debian-user-digest Digest V2007 #2427
2007/9/21, David Balazic [EMAIL PROTECTED]: From: Martin Marcher [mailto:[EMAIL PROTECTED] 2007/9/21, Tim [EMAIL PROTECTED]: please take me off of your email list. List-Id: debian-user.lists.debian.org List-Post: mailto:debian-user@lists.debian.org List-Help: mailto:[EMAIL PROTECTED] List-Subscribe: mailto:[EMAIL PROTECTED] List-Unsubscribe: mailto:[EMAIL PROTECTED] Precedence: list do these hints help you somehow? They are contained in every message you receive from the list They are also invisible to most users. (except hacker, but they know that info already) ;-) true but i remember that I got 3 messages when subscribing telling me how to unsubscribe and *to keep the mail for future reference* - also a lot of people have a hint how to unsubscribe in their sig. I guess a simple search for unsubscribe (since he uses the digest list) should have brought something up at least he probably knows that he should really inspect mails from mailing lists in the future and he got a little knwoledge more about mailing lists (I like to believe that, plese spare me :) martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: can't access localhost
2007/9/21, Liam O'Toole [EMAIL PROTECTED]: On Fri, 21 Sep 2007 08:32:21 +0100 John O Laoi [EMAIL PROTECTED] wrote: # cat /etc/hosts 127.0.0.1 localhost mc 127.0.1.1 mc.home.m mc I see that you solved the CUPS problem, but I should point out that the hosts file above is likely to cause problems elsewhere. You should remove 'mc' from the first line. why should that cause problems? hosts(5): For each host a single line should be present with the following information: IP_address canonical_hostname [aliases...] I can't see anything here that suggests the above entry is wrong. I'd even say that if something should change that John should _add_ localhost.localdomain to the 127.0.0.1 line -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: can't access localhost
2007/9/21, Liam O'Toole [EMAIL PROTECTED]: The issue is that 'mc' resolves to 127.0.0.1 (the first match), whereas mc.home.m resolves to 127.0.1.1 (the only match). Thus the canonical host name and its alias resolve to different IP addresses. uh oh, right, reading in context actually helps :) -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Fwd: Basic Tomcatt5.5 Admin question
Chris i just mistakenly posted that to your address and got an answer from bluebottle.com could you at least not blacklist me in case we have to get in touch by some incident in the future so we won't wonder why mail isn't arriving. oh and I f* hate those verification tools... -- Forwarded message -- From: Martin Marcher [EMAIL PROTECTED] Date: 21.09.2007 19:13 Subject: Re: Basic Tomcatt5.5 Admin question To: Nyizsnyik Ferenc [EMAIL PROTECTED] 2007/9/21, Nyizsnyik Ferenc [EMAIL PROTECTED]: On Fri, 21 Sep 2007 07:42:59 -0700 Dancing Fingers [EMAIL PROTECTED] wrote: Hi guys, I finally got Tomcat5.5 going on Etch AMD and I start studying the JSP developers guide but I get stuck in the beginning. The book says to create a new directory under ROOT but I don't have permission in my default account. The last install I did a chgrp / user/share/tomcat5.5-webapps/ROOT www-data but I screwed the package up so bad that I re-installed the system. Could anyone tell me how the is typically done? when learning tomcat/jboss just download the tarball unpack it to some dir and use startup.sh (tomcat) or run.sh (jboss) to run it. both can be a ressource hog and i think tomcat is by default configured to use 256MB Ram for doing nothing, nada, zip, null, zero. you then also have the opportunity to do fast testing of libraries that require certain tomcat versions but will otherwise ease up your life or simply deploy your playground to a tomcat6 just to see how it behaves (or even jboss for that matter) i think to learn the jsp side this is the better approach (the same is true for eclipse imho, which if you do serious development in J2EE will end up anyway having multiple times on disk even in the same version because 2 plugins just won't play together and you don't have the time to fix it but rather work, so you download eclipse 3.2 the 3rd time set up a worskpace and configure your plugins for the project using $PLUGIN there (hope someone can still follow what I mean) martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Password file with over 3000 users.
2007/9/21, Thomas Goirand [EMAIL PROTECTED]: But the reference for Debian is sh, not bash... (at least when writing packages). why is it then that /bin/sh is /bin/bash after a plain install of about every stable version i had by now? ok that makes only 3 but still (potato sarge etch iirc for that matter) martin -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: debian-user-digest Digest V2007 #2427
2007/9/21, Tim [EMAIL PROTECTED]: please take me off of your email list. List-Id: debian-user.lists.debian.org List-Post: mailto:debian-user@lists.debian.org List-Help: mailto:[EMAIL PROTECTED] List-Subscribe: mailto:[EMAIL PROTECTED] List-Unsubscribe: mailto:[EMAIL PROTECTED] Precedence: list do these hints help you somehow? They are contained in every message you receive from the list -- http://noneisyours.marcher.name http://feeds.feedburner.com/NoneIsYours -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Up-to-date Gnome versions?
2007/9/20, Gabriel Parrondo [EMAIL PROTECTED]: El jue, 20-09-2007 a las 08:56 -0700, Dr. Jennifer Nussbaum escribió: I see that Gnome 2.20 was just released. Im running Debian Etch, which still seems to be stuck on Gnome 2.14, even though 2.16 was released about a year ago and 2.18 since then. Since Etch is now the stable distribution, it is frozen, which means that no newer apps will enter it (except for security fixes). If you want more up-to-date apps you'd rather use Lenny, which is the actual testing distribution. It has the advantage of being up-to-date while keeping a good level of stability. It's meant for final users (unlike stable, which is meant for servers) no testing is not meant for end users, it's ment for testing! Testing means that things may break and you are on your own if things break. It also means that by installing testing by definition you declare that you to some extent know what you are doing and are able to fix this yourself or wait until someone fixes it for you (probably with the next aptitude full-upgrade - which is btw. afaik the recommended way of the former apt-get dist-upgrade iirc). It's not bad to use but you are asking for trouble if you use testing and aren't aware of the above. martin
Re: what's problem with router configuration
Hi, 2007/9/19, Serena Cantor [EMAIL PROTECTED]: I used to connect Internet thru cable modem. Now I buy a wireless router and connect thru router. Actually I use wired part of the wireless router. IP is auto configed as 192.168.1.100, and I can visit admin page of router, but can't visit the rest of Internet. there are a couple of providers out there that restrict to the mac address your computer has (read: a single mac address registered at the provider). call their support and tell them the mac of your router if that is the case. other things: * does the router get a public IP from the cable modem? * does the route have connection to the rest of the world? * is the router supported by your isp (if not just thell them your network card changed and don't tell them about the router until you're absolutely sure that it for some reason doesn't work - i found that if you have unsupported hardware support get's picky about giving you any info at all even if you are allowed to have a route just happens to be a model that is unsupported...) apart from that I assume that you have a Linksys WRT54GL ver 3.1 with 2MB of flash and flashed that with openwrt. you can always connect to the router with ssh and see if that has a connection and check the logs on your router (/var/log) to find hints about problems. hth martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
[OT?] etch + java 1.5.12 + jboss 4.2
Hello, just debootstrapped etch (like so often, about 25 servers in xen running here so something wrong here is unlikely, still open to hints). Got * jdk1.5.0_12 * jboss-4.2.0.GA from their original download websites. extracted to /opt/jboss-4.2.0.GA and /opt/jdk1.5.0_12 I'm open to any hints, I just can't see what I've done wrong, memory seems to be enough, even tried with 512MB swap added no change in behaviour. I just don't have any clue as the general jboss setup itself isn't a problem normally and neither version changes (tried 4.0 as well, and another java version) helped and all threw the same error I have to overlook something... here's my shell session (really _any_ hint is welcome, apart from installing deb's I'm bound to use the official downloads): jboss-b:/opt# pwd /opt jboss-b:/opt# ls -l total 8 drwxr-xr-x 7 root root 4096 2007-09-19 21:33 jboss-4.2.0.GA drwxr-xr-x 9 root root 4096 2007-09-19 21:34 jdk1.5.0_12 jboss-b:/opt/# free -m total used free sharedbuffers cached Mem: 768154613 0 0104 -/+ buffers/cache: 50717 Swap:0 0 0 jboss-b:/opt# export JAVA_HOME=/opt/jdk1.5.0_12 jboss-b:/opt# $JAVA_HOME/bin/java -version java version 1.5.0_12 Java(TM) 2 Runtime Environment, Standard Edition (build 1.5.0_12-b04) Java HotSpot(TM) 64-Bit Server VM (build 1.5.0_12-b04, mixed mode) jboss-b:/opt# cd jboss-4.2.0.GA/bin/ jboss-b:/opt/jboss-4.2.0.GA/bin# ./run.sh -c default = JBoss Bootstrap Environment JBOSS_HOME: /opt/jboss-4.2.0.GA JAVA: /opt/jdk1.5.0_12/bin/java JAVA_OPTS: -Dprogram.name=run.sh -server -Xms128m -Xmx512m -Dsun.rmi.dgc.client.gcInterval=360 -Dsun.rmi.dgc.server.gcInterval=360 -Djava.net.preferIPv4Stack=true CLASSPATH: /opt/jboss-4.2.0.GA/bin/run.jar:/opt/jdk1.5.0_12/lib/tools.jar = 21:51:36,145 INFO [Server] Starting JBoss (MX MicroKernel)... 21:51:36,146 INFO [Server] Release ID: JBoss [Trinity] 4.2.0.GA (build: SVNTag=JBoss_4_2_0_GA date=200705111440) 21:51:36,147 INFO [Server] Home Dir: /opt/jboss-4.2.0.GA 21:51:36,147 INFO [Server] Home URL: file:/opt/jboss-4.2.0.GA/ 21:51:36,148 INFO [Server] Patch URL: null 21:51:36,149 INFO [Server] Server Name: default 21:51:36,149 INFO [Server] Server Home Dir: /opt/jboss-4.2.0.GA/server/default 21:51:36,149 INFO [Server] Server Home URL: file:/opt/jboss-4.2.0.GA/server/default/ 21:51:36,149 INFO [Server] Server Log Dir: /opt/jboss-4.2.0.GA/server/default/log 21:51:36,149 INFO [Server] Server Temp Dir: /opt/jboss-4.2.0.GA/server/default/tmp 21:51:36,150 INFO [Server] Root Deployment Filename: jboss-service.xml 21:51:36,597 INFO [ServerInfo] Java version: 1.5.0_12,Sun Microsystems Inc. 21:51:36,597 INFO [ServerInfo] Java VM: Java HotSpot(TM) 64-Bit Server VM 1.5.0_12-b04,Sun Microsystems Inc. 21:51:36,597 INFO [ServerInfo] OS-System: Linux 2.6.18-4-xen-amd64,amd64 21:51:37,539 INFO [Server] Core system initialized 21:51:41,050 WARN [BasicMBeanRegistry] javax.management.MBeanRegistrationException: preRegister() failed: [ObjectName='jboss.remoting:service=NetworkRegistry', Class=org.jboss.remoting.network.NetworkRegistry ([EMAIL PROTECTED])] 21:51:41,101 ERROR [MainDeployer] Could not create deployment: file:/opt/jboss-4.2.0.GA/server/default/conf/jboss-service.xml org.jboss.deployment.DeploymentException: - nested throwable: (java.lang.reflect.InvocationTargetException) at org.jboss.system.ServiceConfigurator.install(ServiceConfigurator.java:196) at org.jboss.system.ServiceController.install(ServiceController.java:226) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25) at java.lang.reflect.Method.invoke(Method.java:585) at org.jboss.mx.interceptor.ReflectedDispatcher.invoke(ReflectedDispatcher.java:155) at org.jboss.mx.server.Invocation.dispatch(Invocation.java:94) at org.jboss.mx.server.Invocation.invoke(Invocation.java:86) at org.jboss.mx.server.AbstractMBeanInvoker.invoke(AbstractMBeanInvoker.java:264) at org.jboss.mx.server.MBeanServerImpl.invoke(MBeanServerImpl.java:659) at org.jboss.mx.util.MBeanProxyExt.invoke(MBeanProxyExt.java:210) at $Proxy4.install(Unknown Source) at org.jboss.deployment.SARDeployer.create(SARDeployer.java:249) at org.jboss.deployment.MainDeployer.create(MainDeployer.java:969) at org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:818) at org.jboss.deployment.MainDeployer.deploy(MainDeployer.java:782) at
Re: [OT?] etch + java 1.5.12 + jboss 4.2
Found it, and it probably will hit others you _need_ to have the identity jboss wants to create in /etc/hosts - it doesn't seem to be enough that the FQDN is resolveable in dns hope it helps someone at least 2007/9/19, Martin Marcher [EMAIL PROTECTED]: Hello, [lengthy info about jboss not starting] Caused by: java.lang.RuntimeException: Exception creating identity: jboss-b: jboss-b at org.jboss.remoting.ident.Identity.get(Identity.java:211) at org.jboss.remoting.network.NetworkRegistry.preRegister(NetworkRegistry.java:268) at org.jboss.mx.server.AbstractMBeanInvoker.invokePreRegister(AbstractMBeanInvoker.java:966) at org.jboss.mx.modelmbean.ModelMBeanInvoker.invokePreRegister(ModelMBeanInvoker.java:489) at org.jboss.mx.server.AbstractMBeanInvoker.preRegister(AbstractMBeanInvoker.java:654) at org.jboss.mx.server.registry.BasicMBeanRegistry.invokePreRegister(BasicMBeanRegistry.java:697) ... 56 more 21:51:41,267 INFO [Server] Runtime shutdown hook called, forceHalt: true 21:51:41,267 INFO [Server] JBoss SHUTDOWN: Undeploying all packages 21:51:41,273 INFO [Server] Shutdown complete Shutdown complete Halting VM -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: is my raid1 array really working??
Hi, 2007/9/18, harland christofferson [EMAIL PROTECTED]: To me, it looks like HDA and HDC are in good health and configured properly ... but what do I really know. Can someone tell me how I can really know if my configuration is working? Is it REALLY this simple? Disclaimer - read the manpage again before doing that! # mdadm --detail --scan # cat /proc/mdstat # mdadm /dev/md0 --fail /dev/hdc # mdadm --detail --scan # cat /proc/mdstat # mdadm /dev/md0 --remove /dev/hdc # dd if=/dev/random of=/tmp/testFile bs=1M count=100 # cat /proc/mdstat # mdadm /dev/md0 --add /dev/hdc # cat /proc/mdstat do that to see what happens if a drive fails and is replaced. of course if you are somewhat confident that your raid work (and boots from any drive) you can also: * fail a drive and reboot * fail a drive and remove it from the RAID, then reboot /proc/mdstat will always tell you... hth martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Making use of the default printer
Hi, 2007/9/16, - Tong - [EMAIL PROTECTED]: Hi What's the Default Printer actually means? I've setup one of my printers to be Default Printer, http://localhost:631/printers/ seems you refer to CUPS... and I unset the LPDEST environment variable, in hoping the consequent lpr will be using the Default Printer, but this is what I get: just use lpoptions -d foo to set this the default for your user (if you want a special printer to be a default for a certain user) otherwise set the default printer from the CUPS page and simply use lp as the wrapper from the command line -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Making use of the default printer
Hi, 2007/9/16, - Tong - [EMAIL PROTECTED]: On Sun, 16 Sep 2007 18:37:43 +0200, Martin Marcher wrote: hmm, same error: lp test.ps lp: Error - LPDEST environment variable names non-existent destination ! Anything wrong with my system? there could be something that set's the LP environment variable # echo $LP #env |grep LP #set |grep LP and see ich LP is set to something possibly weird/wornd as an intermediate solution if the variable ist set # unset LP of course for a permanent solution you should try to find the program/daemon that is setting LP (if you don't want it which i suspect) does # lp -d yourPrinter testfile.pdf # yes lp handles pdf files just right also ps work as you'd expect it (with LP set and LP unset) if so then you really just need to get rid of the automatically created environment variable LP... And Btw. please provide a proper name on the list gmail shows you as - in conversations which not only looks quite bad but iirc it is also netiquette to provide your real name. $ type lp lp is hashed (/usr/bin/lp) $ type lpr lpr is /usr/bin/lpr $ dir /usr/bin/lp /usr/bin/lpr -rwxr-xr-x 1 root root 14024 2007-02-02 /usr/bin/lp* -rwxr-xr-x 1 root root 9888 2007-02-02 /usr/bin/lpr* My cupsys packages: cupsys_1.2.7-4 cupsys-bsd_1.2.7-4 cupsys-client_1.2.7-4 cupsys-common_1.2.7-4 looks all fine to me... -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Best way to install second version of Debian?
Sorry for top posting (to short to care about), you might be interested in this: http://www.debian.org/releases/stable/amd64/apds03.html.en 2007/9/16, Anthony Campbell [EMAIL PROTECTED]: On 15 Sep 2007, Joe wrote: Anthony Campbell wrote: I want to install a second version of Debian on a spare partition of my HDD (can't have too much of a good thing ...). Currently I have /dev/sda1 as Sidux, /dev/sda3 as Ubuntu. I want to put Debian on /dev/sda8. The step I'm worried about is the final one: i.e. how to manage the boot subsequently without losing access to the existing partitions. Any pointers to documentation, or advice please? You will be asked if you want to install a boot manager. Don't, then at the end of the installation, boot into the 'master' OS, the one which installed the current grub bootloader, then edit /boot/grub/menu.lst to add the new OS. It should be reasonably obvious what is needed, but menu.lst will contain the list of documentation, which may not be the same in all installations. It is fairly well commented. Generally use the same options as the existing OS entries, unless you know otherwise, and avoid any section which is auto-generated. Thanks; this is what I had guessed would be the best way to do it but I didn't want any unpleasant surprises! Anthony -- Anthony Campbell - [EMAIL PROTECTED] Microsoft-free zone - Using Linux Gnu-Debian http://www.acampbell.org.uk (blog, book reviews, on-line books and sceptical articles) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: amd64 vs i386
Hello, 2007/9/15, pietia [EMAIL PROTECTED]: hi Do you have any experiences in that competition: Debian amd64 vs Debian i386 ? yes :) Is i386 still faster than amd64 ? the question isn't wether it's faster or not it's more like the need of 64bit. You can easily go beyond 4GB of RAM (also possible with i386 but you need a kernel option for that which is afaik enabled anyway in standard debian kernels) And what with missing packages like flash plugin ? if you want a desktop PC stay with i386 since there isn't anything available that makes flash work out of the box with amd64. It can however be done (look at the various posts on this list which suggest in essence a chroot and a full 32bit installation of firefox/iceweasel) Personally I don't care (for desktops) about 64 vs 32 bit. The reason for that is that I disable flash anyway since it's useseless for me. For servers I keep with 64bit if (and only if) I expect to need more than 4GB of RAM (which is the usual use case). If I have to decide actively wether I use 32 or 64 bit on a desktop with linux i choose 32bit because most of the software will just work. On the other hand of course if you don't mind some plumbing the 64 bit marketing lie (faster) won't be much of trouble for you. Just my experience... -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: amd64 vs i386
2007/9/15, Ron Johnson [EMAIL PROTECTED]: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 09/15/07 03:57, pietia wrote: hi Do you have any experiences in that competition: Debian amd64 vs Debian i386 ? Is i386 still faster than amd64 ? Since when was i386 *ever* faster than AMD64? since when was 64bit *ever* faster than 32bit? (speaking of speed here not the actual pro of having more address space) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: postgresql help
Hi, 2007/9/15, Chris Bannister [EMAIL PROTECTED]: On Mon, Sep 10, 2007 at 08:36:59PM -0400, Tom Allison wrote: But somewhere I screwed up my postgresql-8.2 database installation. I meant to set up the data directory on the RAID disk array Unfortunately I deleted the /var/lib/postgresql/ directory and can't or don't know what to do to fix it. You mean you have lost your data? If you are just talking about creating an empty cluster in a new location The debian wrappers for this are described in: pg_createcluster, pg_ctlcluster, pg_dropcluster, pg_lsclusters, pg_maintenance if i remember correctly. If you want to stay with postgres builtin methods have a read at: http://www.postgresql.org/docs/8.2/static/creating-cluster.html Have you a recent backup? if you really lost your data by accidentally having deleted the files a recent backup is your only chance of getting back to where you were hth martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
[OT] File Storage Server
Hello, I've been looking around at the dell site (dell preferred since it's our standard vendor) for server with a lot of disks. I'm thinking of some box with 2U - 4U (maybe even 5) that will be a simple fileserver so that I can put debian on it make a SoftRAID (which is what I prefer since we don't have the budget to order backup RAID cards if it breaks) put a LVM on top of that and configure the shares to be available. Now dell has quite a bunch of SAN servers with 9 - 15 disks but those are all prepared SANs which means afaik that they have some OS preinstalled that will serve iSCSI block devices where I also need HBA cards in the clients. In essence all I would like would be a standard server where I can hotplug a lot of disks and be done with it (RAID resizing etc could be done from debian then). Something like 2GB RAM a decent CPU (not too much since it'll be dedicated to file services and RAID) and the option to add disks. Any hints on what to get? (open to suggestions apart from dell but I need the option of a service contract in case of hardware failure... thanks martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] File Storage Server
Hello, 2007/9/14, Miles Fidelman [EMAIL PROTECTED]: Martin Marcher wrote: In essence all I would like would be a standard server where I can hotplug a lot of disks and be done with it (RAID resizing etc could be done from debian then). Something like 2GB RAM a decent CPU (not too much since it'll be dedicated to file services and RAID) and the option to add disks. Any hints on what to get? (open to suggestions apart from dell but I need the option of a service contract in case of hardware failure... I've had really good luck with a couple of boxes I purchased from Rackmounts, Etc. - a lot more horsepower and gigabytes per dollar than the Dell stuff. They come with a 3-year warranty. hmm they look good. Just one problem they only ship within the US and I'm in AT so they are out of question :/ /martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: mondoarchive
2007/9/13, debian user [EMAIL PROTECTED]: I am running Sarge. I apt-get(ted) the stable mondo tools and am trying to use mondoarchive. etch is stable i suggest you get the oldstable (which is sarge) hth martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: make menuconfig problem
Hello, 2007/9/12, Thomas H. George [EMAIL PROTECTED]: I know there is another package which must be installed before make menuconfig will work but I have forgotten what it is. i think that were build-essential and libcurses-dev (something with curses essentially) hth martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Why doin't I have StringDefs.h in X11 include? (etch)
Hi, 2007/9/12, Martin Waller [EMAIL PROTECTED]: A find cannot locate it anywhere on my system - have I missed some important X!! dev package or something? Other includes appear to be present in /usr/include/X11... apt-file search StringDefs.h after aptitude install apt-file -s has helped me a lot of times when searching for stuff like this hth martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: two protocols on one port.
Hi, 2007/9/12, Mumia W.. [EMAIL PROTECTED]: On 09/12/2007 10:51 AM, PETER EASTHOPE wrote: 12.140.16.4 is June. Now you would reconfigure openvpn on May to listen on port 4122. Connections to port 22 on May would be redirected to port 4122 on May only if they come from 12.140.16.4. Otherwise, the connection would go directly May's port 22 which should be running sshd. I think what he's asking for is a protocol multiplexer so that the multiplexer will be listening on say port 22 and depending on which protocol the client is using it will be delegated (not using NATed intentionally) to the correct daemonwhich is a piece of software that would be useful and something i don't think exists until yet... regards martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Multiple MTAs?
Hi, On 8/21/07, Strake [EMAIL PROTECTED] wrote: Hi, I have a Debian server, and was about to install an MTA, when I was stricken by indecision as to which MTA actually to install. I realized that it would probably be best to use Postfix to send/receive mail to/from the Internet (because of its security) and Exim on my internal network (because of its flexibility). I'd suggest you stay with a single package - much easier to maintain when you dig into the stuff deeper than you initially wanted to, just choose the one you are most comfortable with. For me it's postfix if overriding parameters in master.cf isn't flexible enough for you have a look at http://advosys.ca/papers/postfix-instance.html it shows how to create multiple instances of postfix with totally independent configurations. hth martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LVM problems: pvmove errors
Hi, are you absolutely sure that the other PVs have enough space available to remove the desired one? (this bugged me everytime i had to remove a PV) /martin On 8/16/07, michael bane [EMAIL PROTECTED] wrote: I'm trying to reduce a logical vol so that I can remove a HD from it. However, I get errors with pvmove but I don't know why: [EMAIL PROTECTED]:~$ sudo /usr/sbin/pvdisplay /dev/sdb1 Password: --- Physical volume --- PV Name /dev/sdb1 VG Name VolGroup00 PV Size 698.64 GB / not usable 10.34 MB Allocatable yes PE Size (KByte) 32768 Total PE 22356 Free PE 19156 Allocated PE 3200 PV UUID xDWUQB-wdEx-AtEs-YFkh-BKBR-7Jgd-9ik3p0 [EMAIL PROTECTED]:~$ sudo /usr/sbin/pvdisplay /dev/sda2 --- Physical volume --- PV Name /dev/sda2 VG Name VolGroup00 PV Size 465.66 GB / not usable 3.56 MB Allocatable yes PE Size (KByte) 32768 Total PE 14901 Free PE 14839 Allocated PE 62 PV UUID GLGcHf-b3Ga-M61f-okkW-HaAk-Al7P-leVK91 [EMAIL PROTECTED]:~$ sudo /usr/sbin/pvmove /dev/sdb1 Failure parsing mirror status fraction: 1 core Failure parsing mirror status fraction: 1 core ABORTING: Mirror percentage check failed. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Martin Marcher [EMAIL PROTECTED] http://www.mycorners.com https://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher http://www.studivz.net/profile.php?ids=9f83ea8c5996b8ec http://www.amazon.de/gp/registry/wishlist/3KDAGCL2NKOIM/ref=reg_hu-wl_goto-registry/302-4432803-5146435?ie=UTF8sort=date-added -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Help with mod_jk build
Hello, might be A LOT more convienient aptitude search ajp (proxy_ajp something like that) # mind the target protocol RewriteRule ^/foo ajp://the.host.with.tomcat:8009/path/to/webapp hth martin On 8/14/07, Dancing Fingers [EMAIL PROTECTED] wrote: Hi all, Has anyone built mod_jk for sid? I downloaded the source from jakarta but i keep getting stuck. Does anyone know what $APACHE2_HOME is? Thanks. Chris -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Martin Marcher [EMAIL PROTECTED] http://www.mycorners.com https://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher http://www.studivz.net/profile.php?ids=9f83ea8c5996b8ec http://www.amazon.de/gp/registry/wishlist/3KDAGCL2NKOIM/ref=reg_hu-wl_goto-registry/302-4432803-5146435?ie=UTF8sort=date-added -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: bash vs. python scripts - which one is better?
Hi, personally I'd say they both equally powerfull in general (I think both a touring complete which makes them both full grown programming languages - correct me if i'm wrong) depending on the job i use one over the other. If it would be a python script only spawning OS processes it might be a lot easier with bash. on the other hand if there's number/string parsing involved i can do a lot better with python YMMV. /martin On 8/14/07, Vincent Lefevre [EMAIL PROTECTED] wrote: On 2007-08-14 09:32:44 -0400, Steven R. wrote: On Tue, Aug 07, 2007 at 04:03:05PM -0700, Andrew Sackville-West wrote: So what's the right way to do this? I hacked one together the other day: IFS=$'\t\n'; for i in `find . -iname \*m4a`; do faad... blah blah blah and I knew it was a hack because setting $IFS just seems bad... possible unintended consquences, but it worked. I have seen something like the following: find | while read FILE; do echo $FILE done which is almost as bad, as filenames can have \n characters in them. That's why find has -print0... Unfortunately the read builtin doesn't seem to support this feature (or anyone knows how to use its -d option to declare \0 as the delimiter?). -- Vincent Lefèvre [EMAIL PROTECTED] - Web: http://www.vinc17.org/ 100% accessible validated (X)HTML - Blog: http://www.vinc17.org/blog/ Work: CR INRIA - computer arithmetic / Arenaire project (LIP, ENS-Lyon) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Martin Marcher [EMAIL PROTECTED] http://www.mycorners.com https://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher http://www.studivz.net/profile.php?ids=9f83ea8c5996b8ec http://www.amazon.de/gp/registry/wishlist/3KDAGCL2NKOIM/ref=reg_hu-wl_goto-registry/302-4432803-5146435?ie=UTF8sort=date-added
Re: PAM + LDAP and SSH
Hello, On 8/1/07, Bhasker C V [EMAIL PROTECTED] wrote: If you do not want LDAP based authentication then you can edit the nsswitch.conf file (passwd and shadow) to point to appropriate values ? like files, yp for local and NIS auth only. the problem with that is that you then don't have any information about the users available. If you need to have getent passwd list the users in your ldap and don't want them to be able to login just don't mess around with pam, and stick with libnss-ldap only. If, on the other hand you need some users to be able to login a pam_filter is apropriate since you can easily use the host or authorizedService attribute (or any other attribute for that matter) to check account validity for a certain box/host. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
PAM + LDAP and SSH
and now for everyone else too :) -- Forwarded message -- From: Martin Marcher [EMAIL PROTECTED] Date: Aug 1, 2007 1:33 AM Subject: Re: PAM + LDAP and SSH To: Ron Peterson [EMAIL PROTECTED] hello, On 7/31/07, Ron Peterson [EMAIL PROTECTED] wrote: 2007-07-31_12:51:35-0400 Allan Senna Porto [EMAIL PROTECTED]: Hi, Anyone know about problems with /etc/security/access.conf and SSH in Etch? I'm trying to block ldap users to access my servers, but everything that I made don't block user to loggin. If I use this to /etc/pam.d/login and /etc/pam.d/gdm, the user can't loggin. It would help others to help you to see what you actually did. this is how i do it: # ldapsearch -x -LLL -b 'ou=people,dc=example,dc=com'\ '((objectClass=sambaSamAccount))' dn host dn: uid=will.waldo,ou=accounts,ou=people,dc=example,dc=com host: havanna.example.com host: kairo.example.com host: novosibirsk.example.com # grep filter /etc/pam_ldap.conf pam_filter |(host=backend.example.com)(host=\*) hth martin -- Martin Marcher [EMAIL PROTECTED] http://www.mycorners.com https://www.xing.com/profile/Martin_Marcher http://www.linkedin.com/in/martinmarcher http://www.studivz.net/profile.php?ids=9f83ea8c5996b8ec http://www.amazon.de/gp/registry/wishlist/3KDAGCL2NKOIM/ref=reg_hu-wl_goto-registry/302-4432803-5146435?ie=UTF8sort=date-added -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Swap configuration for 16GB of RAM, 8 cores
hello, On 7/31/07, Ron Johnson [EMAIL PROTECTED] wrote: On 07/31/07 15:18, [EMAIL PROTECTED] wrote: What are the current best practices with regards to swap partitions? Is it better to create one big, or several smaller swap partitions? Is the rule of thumb still RAM*2 for the total size? i dropped the rule of thumb advices for swap long ago. usually i watch load (uptime), htop and free as closely as possible (let it mail to you about every 30 minutes the first few weeks) and decide how much swap you need. i know these are really basic tools (especially free only reporting specific points in time) but they are available everywhere and usable everywhere. once you got used to initially calculating the stats by hand it's not that bad at all. monitoring software can be added later on when everything runs stable. Create swap *files* instead. See man mkswap for the technique. Put them on /. personally I stay with LVM. I found it a lot easier to manage (probably just being a habit) as it is the _single_ source of storage for me (as in I try to avoid using other storage techniques as loop mounting, swap files in this case, yadda yadda as much as possible) hth martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: to netatalk+samba+nfs or not
hi, On 7/27/07, Gilles Mocellin [EMAIL PROTECTED] wrote: Le Friday 27 July 2007 18:45:13 Andrew Sackville-West, vous avez écrit: can't answer your question directly, but doesn't apple support nfs? yes And SMB... and yes (somewhat) the reason why i'd like to use netatalk too is that it should provide native browsing capabilities to mac clients (like users have with windows explorer and samba servers). I have to run around each day just to type in the correct URI because people are to lazy to a) either read the FAQ which lists all shares available or b) are just to lazy to enter them martin
Re: IM on a home debian network
hi, On 7/29/07, Jesus Arocho [EMAIL PROTECTED] wrote: I have a home network with 5 boxes, one of which is a server. I run a combination of Ubuntu and Debian desktops and Debian on the server. Is there a program available that will provide instant messaging within the network?# as pointed out, jabber is imho _the_ solution. I prefer wildfire (google: wildfire igniterealtime). It's java but i found it to be the most painless one in upgrading/maintaining and it has a bunch of usefull plugins included. hth martin PS: with that option you'll also have the chance to stay with that account should you ever decide to make it public/use it as a public available service. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
to netatalk+samba+nfs or not
Hello, in our company we have mac/linux/windows clients and I think about adding the native service for each of those. the clients aren't under my control but can be considered as trusted. I know that samba+mac does work (to some extent) but I'm thinking about adding netatalk so that mac users have a more native feeling with all this (automagic share exploration, etc). My worries are wether locking issues could occur. samba+nfs is working seamlessly (nfs opened files are locked and samba das know about that) but does afp also honor kernel oplocks, especially with this combination? thanks martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] Source Based Routing
Hi, On 7/25/07, Bob Proulx [EMAIL PROTECTED] wrote: Martin Marcher wrote: i have a setup where i have a borderline box that has 5 public IP Addresses (this is for the sake of example: 192.0.2.8/29), all is NATed to 10.200.10.0/24. Now the IP the provider uses as gateway is 192.0.2.9 which makes me have 192.0.2.10-14 as a usable range. Uhm... Five public IPs is normal for an 8 IP subnet. But I don't understand your statement about NAT'ing them all to 10.200.10.0/24, especially since the ranges do not add up. I am guessing you are sanitizing addresses (okay) and not quite getting things lined up right. You are right, atm the IPs are 192.0.2.9-14 which are public * 192.0.2.9 is used by our provider * 192.0.2.10-14 are free for us to use 192.10.2.10 is the default outgoing route (where this physical box has aliases for the other IPs too on the same interface) The internal interface on this box has the IP 10.200.10.1 and serves the rest of the range via DHCP where 10.200.10.1 is the gateway address (that's what I meant by the network behind is NATed) The default gateway on my borderline box is 192.10.2.10.10 which makes Hmm... Check that address again. It can't be a 40 bit address. :-) err yes :) It was a bit late yesterday, as said above 192.0.2.10 is the default gateway on the borderline box all traffic from my network look like it came from this address which was fine until now. Do you mean that you have NAT configured so that all clients appears to come from your gateway? exactly Say I want 10.200.10.50 to look like it came from 192.0.2.11 how do I do that? Let me recommend using Shorewall for this. There are a lot of packages that facilitate driving the Linux netfilter. I like the Shorewall one best. YMMV. http://www.shorewall.net/NAT.htm I believe that page documents the configuration that you are wanting to create. If nothing else it should be an additional reference. sounds like a good options, unfortunately I can only use it as a last resort, since the box is an embedded box and not exactly debian - It's embcop but a standard x86 Processor afaik and my boss is picky about it. I'll look into shorewall heard a lot about it and mostly only good stuff. I found http://www.wlug.org.nz/SourceBasedRouting which seems to do what I want the approach also seems easily extensible since I could easily add more addresses to a certain routing table, or maybe even add more routing tables so that I could use the full range as outgoing IP addresses depending on the source IP address. Am I on the right track? Seemingly so far you seem to be on track to me. Excepting the noted confusion about addresses. Ah yea, OOC[1]-Segfault martin [1] Out Of Coffe -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [OT] Source Based Routing
hi, On 7/25/07, Erik Persson [EMAIL PROTECTED] wrote: I have not read all you have written, but will not: iptables -I PREROUTING -t nat -s 10.200.10.50 ! -d 192.0.2.8/29 -j SNAT --to-source 192.0.2.11 do the trick? i guess that's one of the reasons why I prefer debian-user over other lists. I do get information :) Yes that seems to work. I can't test it at the moment as I have to do that on our live system but that sound like a simple solution. gotta look into the difference of source nat and source based routing thou I guess there's a reason why both solutions exist. By the way, if the provider uses 192.0.2.9 as the gateway, and you have 192.0.2.10-192.0.2.14, how do you get the traffic for the ips, other than 192.168.2.10, to your gateway? Are you using multiple ip-addresses on the same interface, or do you use proxy arp? Yes just multiple IPs on the external interface on our borderline host martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: how to ssh to a linux box from an internet cafe
hi, On 7/25/07, Nick Demou [EMAIL PROTECTED] wrote: I'll soon be on vacations without my PC. I believe that internet access from an internet cafe will be my best option. If things go for the worse how can I ssh to my debian server? I suppose that a PC in most internet cafes will be willing to download and run putty.exe but am I right? If not is there any other option? how about ajaxterm? http://antony.lesuisse.org/qweb/trac/wiki/AjaxTerm martin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
