Re: Intel GM965 on Etch
On Sun, Dec 28, 2008 at 10:51:46PM +0530, Nishita Desai wrote: > Thank you for your reply Justin. But i810 does not work. It gives me > an error saying "No matching Device for instance (BusID :0:2:1) found" You might need the line: BusID "PCI:0:2:1" instead of: BusID "PCI:0:2:0" in your /etc/X11/xorg.conf I don't have your graphics card, but I do have the 945GM. $ lspci | grep Graphics 00:02.0 VGA compatible controller: Intel Corporation Mobile 945GM/GMS, 943/940GML Express Integrated Graphics Controller (rev 03) 00:02.1 Display controller: Intel Corporation Mobile 945GM/GMS/GME, 943/940GML Express Integrated Graphics Controller (rev 03) Thus, in my case, BusID "PCI:0:2:0" would be for an external monitor attached with a VGA cable, and BusID "PCI:0:2:1" would be for the laptop screen. Take a look at the output of lspci. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Re: how do i go back to stable? [was: reverting to 'standard' etch installation]
On Thu, Nov 01, 2007 at 08:53:52PM +, michael wrote: > I can give that a try but anybody else have any suggestions? (I've > removed some packages I'd installed under sid so now it's probably > not as clean/obvious as above suggests)... > > I guess I could remove lots of things and then set sources.list to > etch and then install what i need but that sounds like leaving stuff > to chance Not meant to be taken literally, but for inspiration: apt-show-versions | perl -ne 's{/unstable.*|/sid.*}{/etch} and print' \ | xargs apt-get --simulate install -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Easier to Read Fonts
On Sun, Feb 11, 2007 at 01:11:55PM -0500, Winston Smith wrote: > To my chagrin, web pages are actually easier to read indirectly from > my wife's Windows machine (when I connect to it by VNC from my Debian > box) than directly on my Debian box. Her Firefox uses Times New Roman, > which I believe is a true type font. On my Firefox I've tried Serif, > Free Serif (slightly better), etc., but still no contest. Thanks to everyone for suggesting I install msttcorefonts and ttf-*, which I successfully did. I'm using Bitstream Vera Sans at the moment and it is easier to read. Still, doing head on comparisons like I described above with the same nominal fonts, I notice a difference. It's most pronounced with bold and italic letters. For example, my Arial bold capital 'c' looks jagged, whereas my wife's looks smooth. I tried xfstt and 'dpkg-reconfigure x-ttcidfont-conf' with no change. The latter gives you a choice between backends freetype and xtt -- I tried both. I have png screenshots of 8k and 16k but didn't know if it would be useful and permissible to attach them. Thanks again for your help. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Easier to Read Fonts
To my chagrin, web pages are actually easier to read indirectly from my wife's Windows machine (when I connect to it by VNC from my Debian box) than directly on my Debian box. Her Firefox uses Times New Roman, which I believe is a true type font. On my Firefox I've tried Serif, Free Serif (slightly better), etc., but still no contest. An 'apt-cache search' lead me to xfstt. Is that what I need? If so, where do I get the actual fonts, especially Times New Roman? Thanks for your help. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Checking integrity of cached debs
Winston Smith wrote: > Is there an easy way to check the integrity of cached debs and remove > corrupt ones? Also, is there an easy way to check the integrity of the > unchanging parts of installed packages? Jhair Tocancipa Triana wrote: > You can install debsums for that, > > Description: Verify installed package files against MD5 checksums. > debsums can verify the integrity of installed package files against > MD5 checksums installed by the package, or generated from a .deb > archive. Thank you, Jhair. That takes care of the second question beautifully. As for checking the integrity of a possibly un-installed cached deb like /var/cache/apt/archives/akode_4%3a3.3.2-1_i386.deb, this is the best I know so far: Drop the 4%3a ("%3a" is url-ese for ":") from the base to obtain akode_3.3.2-1_i386.deb, which you look for in the "Filename:" field of the entries in the "Packages" files in /var/lib/apt/lists (or downloaded packages.gz files from the mirrors in /etc/apt/sources.list). Take the md5sum from the entry where you find it, and compare it to the actual md5sum of the cached deb. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Checking integrity of cached debs
Does apt-get check the integrity of cached debs before installing them? For example, if I did apt-get --download-only install which downloads the package to the file /var/cache/apt/archives/.version.deb but doesn't install it, and then sometime later, after the above file became corrupt because of a faulty hard drive, did apt-get install (which required no new download because there is no version change) would the corruption be detected? In other words, is the check done after the download, before the installation, or both? Is there an easy way to check the integrity of cached debs and remove corrupt ones? Also, is there an easy way to check the integrity of the unchanging parts of installed packages? Thanks (a lot). Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Help: NO longer boots ;
These are long shots, but since nobody else answered: If the reboot after the power failure was the first reboot since upgrading a package like udev or installing a new kernel, it could be that your grub or lilo stanza is wrong, especially if you have more than one hard drive. It could also be an incorrect /etc/fstab. You can experiment with changing a grub stanza by using the 'e' option. You can check /etc/fstab using a rescue disk. It could be that /dev/.static/dev was deleted because of a file system error due to the unclean shut down or a faulty hard drive. Good luck! Win On Mon, Nov 13, 2006 at 10:13:20AM -0600, helices wrote: > I have a dual Opteron box (Odin) that went down while I was gone, due to > an extended power failure. > > When it comes up, it fails to complete boot. Please, help me get this > system running again. > > uname -a > Linux (none) 2.6.17-2-amd64 #1 SMP ... > > Bootup sequence from obvious problems: > > ... > Success: loaded module raid10. > Done. > device-mapper: 4.6.0-ioctl (2006-02-17) initialised: ... > 1 logical volume(s) in volume group "VG2" now active > 7 logical volume(s) in volume group "VG1" now active > Done. > stdin: error 0 > Begin: Running /scripts/local-premount ... > kinit: name_to_dev_t(/dev/mapper/VG1-swap) = dm-3(253,3) > kinit: trying to resume from /dev/mapper/VG1-swap > Attempting manual resume > kinit: No resume image, doing normal boot ... > Done. > Usage: modprobe ... > ... > > Begin: running /scripts/init-bottom ... > mount: Mounting /root/dev on /dev/.static/dev failed: No such file or > directory > Done. > mount: Mounting /sys on /root/sys failed: ... > mount: Mounting /proc on /root/proc failed: ... > Target filesystem doesn't have /sbin/init > ... > > (initramfs)_ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Trouble browsing secure web sites (scotiaonline.com)
> > you're not alone: > > http://www.oclug.on.ca/archives/linux/2006-July/000920.html > > http://reporter.mozilla.org/app/report/?report_id=RMO11314606252824 One of those pages mentions fiddling with mtu settings. I second that. I had a problem like yours that I fixed by changing the mtu setting on my router from 1500 to 1400. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Video in mozilla/firefox
I had a similar problem that I fixed by installing hal, which I had uninstalled for some reason. I realized it had to do with hal because of error messages that were printed in the terminal when I started firefox on the command line. Otherwise I wouldn't have seen these messages. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: kernel panic with SATA
On Sun, Mar 26, 2006 at 11:24:50PM +0200, oscar wrote: > I am trying to install Debian in a new Dell M70. I first used a CD with the > testing installer, but although it was able to boot from the CD, later it did > not recognize the CD. I have read in other places that it have to do with > SATA controller for the CD and the HD. > Then I tried the stable installer with a kernel 2.4. It installed Debian with > no problem. With aptitude I install a kernel 2.6.8 (still in stable > distribution), and when rebooting I get a kernel panic (no /dev/console can > be found, attempting to kill init). As far as I have discovered with google > help, it have to do again with the SATA harddisk. I have followed several > advices of people with same problem, basically, to change menu.lst of grub > from root=/dev/hda1 to root=/dev/sda1. No result. Still kernel panic. > Moreover, I try testing distribution to see what happens. Now not only with > kernel 2.6 but even with kernel 2.4 i get a kernel panic. This is how I solved a similar problem: http://lists.debian.org/debian-user/2006/03/msg01661.html Note that I had to add the initrd line in addition to changing root=... Hope that helps. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: kernel panic with SATA
On Sun, Mar 26, 2006 at 02:44:23PM -0700, Glenn English wrote: > The installer sets up disk names and labels one way, then the boot process > names them so they don't match. SATAs are hd's at one point, sd's at another. > And something moves them ahead of real SCSIs in /dev. > > Things seem to work fine with IDEs and SCSIs. And the bug site says they know > about this problem and that they'll fix it in the next release. Or maybe in > an installer update. Ubuntu, not too surprisingly, acts the same way. Glenn, Could you post a link to the bug report you're referring to. Thanks a lot. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Can't reinstall missing init.d script
On Tue, Mar 14, 2006 at 08:06:01PM -0300, Henrique de Moraes Holschuh wrote: > On Wed, 15 Mar 2006, Andrei Popescu wrote: > > > Maybe you could extract the file from the .deb manually. > > This is probably the best idea. Thanks. That worked. If I recall correctly, I did: dpkg-deb --extract \ /var/cache/apt/archives/initscripts_2.86.ds1-12_i386.deb tempdir mv tempdir/etc/init.d/checkfs.sh /etc/init.d rm -fr tempdir Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
udev(?) reordered drives causes kernel panic after upgrade
I had some trouble booting the 2.6.15-1-k7 kernel that I installed to satisfy dependencies in an upgrade from sarge to etch, which included an upgrade of udev. I got this message when booting: VFS: Cannot open root device "sda2" or unknown-block(0,0) Please append a correct "root=" boot option Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0) I fixed the problem by changing sda to sdb and adding an initrd line in the grub stanza for the kernel, i.e., I changed the automatically generated stanza (influenced by the line '# kopt=root=/dev/sda2 ro' in my /boot/grub/menu.lst): title Debian GNU/Linux, kernel 2.6.15-1-k7 root(hd0,0) kernel /vmlinuz-2.6.15-1-k7 root=/dev/sda2 ro savedefault boot to: title Debian GNU/Linux, kernel 2.6.15-1-k7 root(hd0,0) kernel /vmlinuz-2.6.15-1-k7 root=/dev/sdb2 ro initrd /initrd.img-2.6.15-1-k7 savedefault boot It seems that the new udev(?) orders the drives (in my case, sata) differently. Also, previous kernel upgrades added an initrd line automatically to the new stanza in /boot/grub/menu.list. I figured out the new order of the drives by typing 'find | more' in the ash(?) shell I was dropped into after rebooting (only having added the initrd line). I paged until I found paths with sda1, sdb2, sdb2, etc., from which I deduced that sda was now the sata drive with only one partition. Should these issues be reported as bugs and if so,in which packages? I found this: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=292038 Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Can't reinstall missing init.d script
Hi. I would like to recover the /etc/init.d/checkfs.sh script that came with it's package, which according to 'dpkg -S' is initscripts. I hid the current checkfs.sh and ran apt-get --reinstall install initscripts seemingly successfully, but no new checkfs.sh was created. I also tried 'dpkg-reconfigure initscripts' but that didn't do anything. Etch Package: initscripts Version: 2.86.ds1-12 Thanks for your help. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Page crashes mozilla: BadAlloc
On Fri, Mar 10, 2006 at 09:39:38AM -0500, Ralph Katz wrote: > Maybe it's #335708: mozilla-firefox: Firefox crashes with BadAlloc > Tags: confirmed; Merged with #330396, #331673, #334654, #345866; 135 > days old. > > Please check the bug reports, sounds very familiar to me. > > Regards. > Thanks. I narrowed it down to a single gif with over 11,000 columns so it's #331673. I did search the mozilla reports, but not firefox, though I may have missed the same thing among the mozilla reports. I'll be working on my bug searching skills. At least I can notify the wunderground.com folks that they have a crazy gif. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Page crashes mozilla: BadAlloc
On Thu, Mar 09, 2006 at 12:02:36PM -0800, Andrew Sackville-West wrote: > ['http://www.wunderground.com/cgi-bin/findweather/getForecast?query=charlottesville+va'] > works fine in firefox 1.5.0.1 on my sid box. > you're running high 60's to low 70's for the next 5 days > with lows around 50, alternating partly cloudy with 30% chances of rain. hth. > > A Thanks for the report, Andrew. I'd be checking in with you every couple of days for updates from now on, but konqueror doesn't crash, so you're off the hook. ;-) I'd still like to know whether (not weather) to file a bug report. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Page crashes mozilla: BadAlloc
Could someone check if this link crashes mozilla and firefox as it does for me? Thanks. Win $ mozilla 'http://www.wunderground.com/cgi-bin/findweather/getForecast?query=charlottesville+va' > The program 'Gecko' received an X Window System error. > This probably reflects a bug in the program. > The error was 'BadAlloc (insufficient resources for operation)'. > (Details: serial 18054 error_code 11 request_code 53 minor_code 0) > (Note to programmers: normally, X errors are reported asynchronously; >that is, you will receive the error a while after causing it. >To debug your program, run it with the --sync command line >option to change this behavior. You can then get a meaningful >backtrace from your debugger if you break on the gdk_x_error() function.) $ mozilla -v > Mozilla 1.7.8, Copyright (c) 2003-2004 mozilla.org, build 2005092713 I think I recall that the results of 'free' are misleading, but anyway: $ free > total used free sharedbuffers cached > Mem: 15566921470808 85884 0 96892 863764 > -/+ buffers/cache: 5101521046540 > Swap: 987956 54060 933896 $ uname -a > Linux localhost 2.6.11-1-k7 #1 Mon Jun 20 21:26:23 MDT 2005 i686 GNU/Linux I'm running sarge. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: GDM screen too big for display
On Sat, Feb 25, 2006 at 08:37:44PM +0100, nikias wrote: > hello, > > gdm starts default as my login manager. > The problem is that it doesn't use the correct proportions of my screen. > Although everything looks sharp, I have to scroll to see the whole screen. > After I log in, Gnome starts and uses the correct size. > > Does anyone have an idea where I have to look to modify it? Maybe you have a wrong resolution in a Modes line of the file /etc/X11/XF86Config-4. For example, with the Mode line below, I think I would get the effect you describe. Leaving out the resolution "1280x960", which is wrong for my monitor, would fix it. SubSection "Display" Depth 24 Modes "1280x960" "1280x1024" "800x600" "640x480" EndSubSection -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Focus in KDE with kwin, New Widows Have Inferiority Complex (solved)
On Sat, Jan 07, 2006 at 10:34:43AM -0500, Winston Smith wrote: > How do I configure KDE and kwin so that the window for a new application > launched from an existing application, an icon, or the panel gets > focus automatically and doesn't appear below other windows? This was the > default behavior in gnome with the sawfish or metacity window managers. ... > I tried using the "configure window behavior/focus" and > "configure window behavior/window-specific settings/preferences" > menu items from the title bar but couldn't get the results I wanted. > Changing "focus stealing prevention level" from "low" to "none" solved the problem. Apparently 'low' is the default. If I remember correctly, this option is under K menu -> configuration editor -> desktop -> window behavior -> advanced. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: DHCP with Static Addresses
On Fri, Nov 25, 2005 at 12:02:47PM -0500, Kenneth Jacker wrote: > am> server. Assuming you're using dhcpd you'll want something > am> similar to this in your dhcpd.conf: > > am> host yourhostname { > am> hardware ethernet 00:11:22:AA:BB:CC; > am> fixed-address 192.168.0.123; > am> } > > Good idea! > > However, 'dhcpd' is running within my LinkSys router, not on one of my > Debian boxen ... > > I spent quite a few minutes going through the router's option pages, > but couldn't see a way to specify this. I have a Linksys wireless router that, as far as I can tell, can't be configured as suggested above (without changing the firmware). Fortunately, you can use /etc/dhclient.conf on the client instead of dhcpd.conf on the router to accomplish something similar. I have this in /etc/dhclient.conf: alias { interface "eth0"; fixed-address 192.168.1.101; option subnet-mask 255.255.255.0; } And this in /etc/network/interfaces: # The loopback network interface auto lo iface lo inet loopback # The primary network interface auto eth0 iface eth0 inet dhcp This setup works with port forwarding for ssh and, as far as I know, everything else but openvpn. To make it work for openvpn I also changed /etc/dhclient-script to promote the fixed-address, which manifests only as an alias, to the primary address. $diff -u /etc/dhclient-script.dist /etc/dhclient-script --- /etc/dhclient-script.dist 2005-11-26 09:01:15.0 -0500 +++ /etc/dhclient-script2005-11-26 09:00:43.0 -0500 @@ -123,6 +123,12 @@ exit_with_hooks 0 fi +if [ x$alias_ip_address != x ] ; then + given_ip_address=$new_ip_address; + new_ip_address=$alias_ip_address; + alias_ip_address=$given_ip_address; +fi + if [ x$reason = xBOUND ] || [ x$reason = xRENEW ] || \ [ x$reason = xREBIND ] || [ x$reason = xREBOOT ]; then if [ x$old_ip_address != x ] && [ x$alias_ip_address != x ] && \ If you're not familiar with context diffs, the lines beginning with a single '+' are what was added by me. Hope that helps. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Orphan Inodes After Power Outage
After a power outage, a Debian box (without a UPS) had orphan inodes on 4 different partitions: /usr, /var, /tmp/, and /home. The /usr partition alone had 74 orphan inodes. Does that mean there are 74 missing files under /usr? There are no files in the lost+found directories for the partitions. The box seems to be OK. Please tell me there is some harmless explanation. The WD hard drive is less than 1 year old, but it is a replacement from the manufacturer (an RMA) and might be more likely to fail. A dmesg excerpt is below. Thanks. Win kjournald starting. Commit interval 5 seconds EXT3 FS on hda8, internal journal ext3_orphan_cleanup: deleting unreferenced inode 1586955 ext3_orphan_cleanup: deleting unreferenced inode 1586782 ext3_orphan_cleanup: deleting unreferenced inode 1586272 EXT3-fs: hda8: 3 orphan inodes deleted EXT3-fs: recovery complete. EXT3-fs: mounted filesystem with ordered data mode. kjournald starting. Commit interval 5 seconds EXT3 FS on hda7, internal journal ext3_orphan_cleanup: deleting unreferenced inode 12232 ext3_orphan_cleanup: deleting unreferenced inode 12231 ext3_orphan_cleanup: deleting unreferenced inode 12230 ext3_orphan_cleanup: deleting unreferenced inode 12229 ext3_orphan_cleanup: deleting unreferenced inode 12226 ext3_orphan_cleanup: deleting unreferenced inode 12225 ext3_orphan_cleanup: deleting unreferenced inode 12224 ext3_orphan_cleanup: deleting unreferenced inode 12223 EXT3-fs: hda7: 8 orphan inodes deleted EXT3-fs: recovery complete. EXT3-fs: mounted filesystem with ordered data mode. kjournald starting. Commit interval 5 seconds EXT3 FS on hda5, internal journal ext3_orphan_cleanup: deleting unreferenced inode 162939 [!!!MANY LINES DELETED] ext3_orphan_cleanup: deleting unreferenced inode 321338 EXT3-fs: hda5: 74 orphan inodes deleted EXT3-fs: recovery complete. EXT3-fs: mounted filesystem with ordered data mode. kjournald starting. Commit interval 5 seconds EXT3 FS on hda6, internal journal ext3_orphan_cleanup: deleting unreferenced inode 16126 EXT3-fs: hda6: 1 orphan inode deleted EXT3-fs: recovery complete. EXT3-fs: mounted filesystem with ordered data mode. /dev/hda5=/usr /dev/hda6=/var /dev/hda7=/tmp /dev/hda8=/home -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Vim question
On Sat, Oct 22, 2005 at 08:16:52PM -0500, Eric P wrote: > > If you wanted to instead replace > > > > line 1 > > line 2 > > line 3 > > > > with > > > > /* > > line 1 > > line 2 > > line 3 > > */ > > > > I'm not sure how you'd do it. Perhaps others do. > > > > Actually, I would like to know how to do this in VIM. > > Any takers? > > Eric I have scripts cos (comment out shell) and coc (comment out C) so that, for example, !}cos comments out a paragraph with '#' at the beginning of each line and !}cos comments out a paragraph with /*...*/. cat ~/bin/cos: #!/usr/bin/perl while (<>) { s/^/#/; print } - cat ~/bin/coc: echo '/*' cat echo '*/' -- Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: excluding files w/ tar (retraction)
On Thu, Jun 02, 2005 at 11:48:00PM -0400, Winston Smith wrote: > One way to test exclude patterns is with > > tar -c --exclude=pattern includedirectorys | tar -t - > > Unfortunately, I didn't run such a test before responding to the > original poster, and now that I have I would like to retract my previous > posts on the subject. Sorry! > > The examples below show that the stripping of the leading / by tar is > not the issue as I thought it was. I should add that the problem of the original poster was caused by the presence of the leading slash in the exclude patterns and its absence in the included files (tar wasn't stripping the leading slash in the exclude patterns because there was no leading slash in the included files). The problem is that the exclude patterns have absolute paths while the included files have relative paths. > $ rm -fr /tmp/{a,b} > $ mkdir -p /tmp/a/b /tmp/b/tmp/a/c #absolute relative $ cd /; tar -c --exclude=/tmp/a tmp/{a,b} | tar -t -f - tmp/a/ tmp/a/b/ tmp/b/ tmp/b/tmp/ tmp/b/tmp/a/ tmp/b/tmp/a/c/ # Nothing is excluded as in the case of the original poster. Changing the exclude patterns to relative paths without using "--anchored" before the exclude patterns can exclude too much: $ cd /; tar -c --exclude=tmp/a tmp/{a,b} | tar -t -f - tmp/b/ tmp/b/tmp/ # Directory tmp/b/tmp/a is excluded. $ cd /; tar -c --anchored --exclude=tmp/a tmp/{a,b} | tar -t -f - tmp/b/ tmp/b/tmp/ tmp/b/tmp/a/ tmp/b/tmp/a/c/ # The desired result. Changing everything to absolute paths also works: $ tar -c --exclude=/tmp/a /tmp/{a,b} | tar -t -f - tar: Removing leading `/' from member names tmp/b/ tmp/b/tmp/ tmp/b/tmp/a/ tmp/b/tmp/a/c/ # The desired result. This also works. $ cd /; tar -c --exclude=./tmp/a ./tmp/{a,b} | tar -t -f - Winston -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: excluding files w/ tar (retraction)
One way to test exclude patterns is with tar -c --exclude=pattern includedirectorys | tar -t - Unfortunately, I didn't run such a test before responding to the original poster, and now that I have I would like to retract my previous posts on the subject. Sorry! The examples below show that the stripping of the leading / by tar is not the issue as I thought it was. $ rm -fr /tmp/{a,b} $ mkdir -p /tmp/a/b /tmp/b/tmp/a/c $ tar -c --exclude=nothing /tmp/{a,b} | tar -t -f - tar: Removing leading `/' from member names tmp/a/ tmp/a/b/ tmp/b/ tmp/b/tmp/ tmp/b/tmp/a/ tmp/b/tmp/a/c/ #everything is listed, as expected $ tar -c --exclude=/tmp/a /tmp/{a,b} | tar -t -f - tar: Removing leading `/' from member names tmp/b/ tmp/b/tmp/ tmp/b/tmp/a/ tmp/b/tmp/a/c/ # the desired result $ tar -c --exclude=tmp/a /tmp/{a,b} | tar -t -f - tar: Removing leading `/' from member names tmp/b/ tmp/b/tmp/ # not the desired result Winston -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: excluding files w/ tar
On Fri, Jun 03, 2005 at 09:48:09AM +0800, Robert Vangel wrote: > `./*' wouldnt include (or exclude) `./.*' files where `.' would. Good catch. Thanks. Winston -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: excluding files w/ tar
On Thu, Jun 02, 2005 at 11:08:15AM -0400, Winston Smith wrote: > would be, for example, "./tmp/", not "/tmp/". > > Try putting ./tmp/* instead of /tmp/* and so on in /tmp/excludes, or, if you My suggestion above assumes you use tar cv -X /tmp/excludes -p --atime-preserve ./* -- | ... ^^^ instead of tar cv -X /tmp/excludes -p --atime-preserve * -- | ... ^ ./* -> . would work too. Winston -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: excluding files w/ tar
On Wed, Jun 01, 2005 at 11:00:18AM -0700, Cameron Matheson wrote: > Hi everyone, > > i'm trying to back up an old suse box so that i can > clear it off and install debian. anyway, i want to > tar up basically everything (except for the obvious > stuff like proc, /tmp/, /dev, etc.) so i've made an > excludes file: > > $ cat /tmp/excludes > /tmp/* > /dev/* > *oldstuff* > /proc > /var/tmp/* > /home/suse > > then i'm using this command to tar it up: > $ tar cv -X /tmp/excludes -p --atime-preserve * -- | > netcat 10.1.22.39 80 > > but all the directories in my excludes file are still > being backed up! where am i going wrong? I think the match is done against the file as it named by tar, which, in the absence of this option of tar: -P, --absolute-names don't strip leading `/'s from file names would be, for example, "./tmp/", not "/tmp/". Try putting ./tmp/* instead of /tmp/* and so on in /tmp/excludes, or, if you never make mistakes, use -P (dangerous when extracting). Winston -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: All browsers crash on button click (java?)
Winston (this poster) writes: > $ ls -l /etc/alternatives/*java* > #edited > java -> /usr/lib/j2re1.5-sun/bin/java > javac -> /usr/bin/jikes-sablevm > javadoc -> /usr/bin/gjdoc > javah -> /usr/bin/javah-cp > javap -> /usr/bin/javap-cp > javaws -> /usr/lib/j2re1.5-sun/bin/javaws > firefox-javaplugin.so -> > /usr/lib/j2re1.5-sun/plugin/i386/ns7/libjavaplugin_oji.so > mozilla-javaplugin.so -> > /usr/lib/j2re1.5-sun/plugin/i386/ns7/libjavaplugin_oji.so > netscape-javaplugin.so -> > /usr/lib/j2re1.5-sun/plugin/i386/ns4/libjavaplugin.so > I installed the sdk (sun-j2sdk1.5_1.5.0+update03_i386.deb) and ran java-alt-setup, so the relavent links above all point to files in the /usr/lib/j2sdk1.5-sun/ directory now. Mozilla and Firefox still crash on the epa link. Winston -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: All browsers crash on button click (java?)
Wulfmann writes: > I'm currently on a WinME machine, Mozilla 1.7.5, Java 1.5. When I go to > the page and click on the "generate map" button I get a page with the > following error: > "Error connecting to the SAS server ... Thanks. That may be relevant. I don't know. Jonathan writes: > I can confirm that your link does crash Firefox. It does NOT crash > Konqueror, however and the map seems to work normally. I can make it > bigger or smaller. I'm using jre-1_5_0_02 that I got from Sun (part of > JDK) on Sarge 2.6.8-2-386 > Hope this helps, Thanks for mentioning Konqueror. I found that Konqueror works for me too. Yeah Konqueror! Also thanks for the crash confirmation. Greg C. Madden writes: > Map generation works here with jre-1_5_0_03-linux-i586.bin, note, one > version off from what you posted, also used the make-jpkg, with Sarge. Thanks. I upgraded from update 1_5_0_02 to '03 motivated by your post and still no go with anything but Konqueror. Roberto writes: > FF 1.0.4 on Sarge with Sun JDK 1.5.0 update 03. Works like a champ for > me. That's great. So it's possible. I'm running testing up to date (same FF). Maybe I should intall the JDK instead of the just the JRE, but that shouldn't matter. Maybe I have something mis-configured. Stupidity on my part is likely. I checked some obvious things: $ ls -l /etc/alternatives/*java* #edited java -> /usr/lib/j2re1.5-sun/bin/java javac -> /usr/bin/jikes-sablevm javadoc -> /usr/bin/gjdoc javah -> /usr/bin/javah-cp javap -> /usr/bin/javap-cp javaws -> /usr/lib/j2re1.5-sun/bin/javaws firefox-javaplugin.so -> /usr/lib/j2re1.5-sun/plugin/i386/ns7/libjavaplugin_oji.so mozilla-javaplugin.so -> /usr/lib/j2re1.5-sun/plugin/i386/ns7/libjavaplugin_oji.so netscape-javaplugin.so -> /usr/lib/j2re1.5-sun/plugin/i386/ns4/libjavaplugin.so I did an strace on mozilla. Towards the end there was mention of "java/lang/System" followed by incomprehensible stuff followed by boring stuff. Thanks to all who responded. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
All browsers crash on button click (java?)
All browsers I've tried including Mozilla, Firfox, and Epiphany crash (all browser windows cease to exist) when I click on the "generate map" button on this page: http://www.epa.gov/air/data/monloc.html?st~VA~Virginia It may have to do with java. I downloaded jre-1_5_0_02-linux-i586.bin from sun and ran /usr/bin/make-jpkg from the java-package package to create sun-j2re1.5_1.5.0+update02_i386.deb which I installed. This java test page works just fine: http://www.bodo.com/javame.htm If someone hase a combination of jre and browser that can handle the epa page, I would really like to know. Thanks. Win -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: select N random lines in a file
On Sun, Aug 22, 2004 at 02:35:14PM -0500, Lance Hoffmeyer wrote: > I would like to write a script that will select N number of > random lines in a file. Any suggestions on how to do this? This program has the advantage that it doesn't read the whole file into memory, which is important if the file is large. Save it as an executable file, randomlines, and then type "randomlines N file" to get N random lines from file (without repetition). The lines will be in the same order they were in the file. Type "randomlines -r N file" to get N random lines in random order. #! /usr/bin/perl -s $N=shift; #first arg is N srand; while(<>){ if(rand($.) < $N){ if(@lines == $N){ # drop one random element splice @lines,int rand $N,1; } if($r){ splice @lines, int rand @lines+1, 0, $_; } else{ push @lines, $_; } } } print $_ for @lines; __END__ The proof that the algorithm is correct is by induction on the number of lines in the file (also, see Knuth reference below). It is based on a program in the perl documentation that returns 1 random line from a file, which I found by typing "perldoc -q 'random line'": How do I select a random line from a file? Here's an algorithm from the Camel Book: srand; rand($.) < 1 && ($line = $_) while <>; This has a significant advantage in space over reading the whole file in. You can find a proof of this method in The Art of Computer Pro- gramming, Volume 2, Section 3.4.2, by Donald E. Knuth. You can use the File::Random module which provides a function for that algorithm: use File::Random qw/random_line/; my $line = random_line($filename); Another way is to use the Tie::File module, which treats the entire file as an array. Simply access a random array element. (END) Winston Smith, [EMAIL PROTECTED] where x=winstonsmith, y=ispwest.com -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]