Re: Firefox problem
On Wed, 03 Feb 2021 05:20:01 +0100, Charles Curley wrote: > On Tue, 2 Feb 2021 23:12:52 - (UTC) > Frank Miles wrote: > >> With the most recent Firefox update, some of the widgets on a >> Bibliocommons (library) website cease working. They still work >> either using the Epiphany browser, or my android phone. >> >> Any recommendations on how I might debug this? > > Sometimes Firefox loses its marbles. To check on this, shut Firefox > down, move ~/.mozilla aside (rename it), then start Firefox up again. > > If that solves the problem, you have some re-installing to do. If that > doesn't solve the problem, reverse the above. That fixes it! As such, I presume it's a matter of piece-by-piece "installing" add-ons and configuration settings, not installing Firefox itself since that "works". Thanks for the tip!! Frank
Firefox problem
With the most recent Firefox update, some of the widgets on a Bibliocommons (library) website cease working. They still work either using the Epiphany browser, or my android phone. Any recommendations on how I might debug this? Running DebianAmd64 'Buster'. Thanks for any suggestions! -Frank
Inreach mini cannot be found by KVM/Win7 guest
Just got an InReach mini (Garmin GPS USB device). It seems ok by 'dmesg'. However [a] when mounted as a USB thumbdrive there doesn't seem to be anything usefully readable; and [b] when I bring up a Win7 guest under KVM/QEMU, win7 can't find it. This is in contrast to my Garmin etrex GPS device, which connects perfectly in both these ways. If I take this same device and plug it into a native Win7 system it works perfectly. This is on a 64-bit 'buster' system. Anyone have any clues about this? Duckduckgo hasn't helped. Thanks--
smbclient not finding share for localhost
My home linux/Debian/buster machine has suddenly made samba 'shares'
invisible to my Win7 virtual machine (running under kvm/qemu).
Using smbclient I get two different responses:
smbclient -L localhost
only gives the error-message response:
Unable to initialize messaging context
Connection to localhost failed (Error NT_STATUS_UNSUCCESSFUL)
But if I given the IP address:
smbclient -L 192.168.1.20
it asks for a password, then tells me about the shares:
Unable to initialize messaging context
Enter WORKGROUP\user's password:
...shares...
server info
workgroup info
If I do the same thing on another computer (running Stretch), I
don't get the messaging-context whining, and after entering my
password get the information whether I've given it 'localhost'
or the IP address as an argument. Unfortunately the sambas
are a different version, so I don't think I can directly compare
their configuration. In case you wonder, testparm's [global]
output is:
[global]
bind interfaces only = Yes
deadtime = 1440
dns proxy = No
domain master = Yes
interfaces = br0 eth0 127.0.0.0/8
keepalive = 5
log file = /var/log/samba/log.%m
map to guest = Bad User
max log size = 1000
name resolve order = wins lmhosts bcast
obey pam restrictions = Yes
pam password change = Yes
panic action = /usr/share/samba/panic-action %d
passwd chat = *Enter\snew\s*\spassword:* %n\n
*Retype\snew\s*\spassword:* %n\n *password\supdated\ssuccessfully* .
passwd program = /usr/bin/passwd %u
server role = standalone server
unix password sync = Yes
username map = /etc/samba/username.map
usershare allow guests = Yes
wins server = 192.168.1.20
idmap config * : backend = tdb
hosts allow = 192.168.0.0/16 127.0.0.1
hosts deny = ALL
Any hints or pointers to the Fine Manual gratefully received!
-Frank
{AFAICR I hadn't changed smb.conf since when it had been working.
And I've tried this with the firewall completely off}
Re: python3.7 disfunctional (was: argparse missing from libpython3.7-stdlib ?) NOW SOLVED
On Sat, 23 Feb 2019 02:30:01 +0100, Frank Miles wrote: > I just upgraded my desktop from stable/stretch to testing/buster. > I'd earlier done the same to a laptop without any problems. > On the desktop, I get the following strange traceback : > > $ python3 Python 3.7.2+ (default, Feb 2 2019, 14:31:48) > [GCC 8.2.0] on linux Type "help", "copyright", "credits" or "license" > for more information. >>>> help() > Traceback (most recent call last): > File "", line 1, in > File "/usr/lib/python3.7/_sitebuiltins.py", line 102, in __call__ > import pydoc > File "/usr/lib/python3.7/pydoc.py", line 66, in > import inspect > File "/usr/lib/python3.7/inspect.py", line 40, in > import linecache > File "/usr/lib/python3.7/linecache.py", line 11, in > import tokenize > File "/usr/lib/python3.7/tokenize.py", line 33, in > import re > File "/usr/lib/python3.7/re.py", line 143, in > class RegexFlag(enum.IntFlag): > AttributeError: module 'enum' has no attribute 'IntFlag' >>>> >>>> > I've tried reinstalling python3.7 and its library packages without > effect. > > Any hints on how I might find what's causing this pathological behavior? > TIA! >-Frank I posted a similar query on the python newsgroup - and Chris Angelico provided a good diagnostic method and correctly speculated about the source of the problem. What had happened some time ago was I'd introduced a python3 enum system, which happened to reside at /usr/local/python3.7/dist-packages. This was earlier in the path than the proper enum module, so got imported rather than the one in /usr/lib/python3.7/ . Renaming things to get it out of the way has restored the proper import, allowing python3 to work normally. Thanks all- Frank
python3.7 disfunctional (was: argparse missing from libpython3.7-stdlib ?)
I just upgraded my desktop from stable/stretch to testing/buster. I'd earlier done the same to a laptop without any problems. On the desktop, I get the following strange traceback : $ python3 Python 3.7.2+ (default, Feb 2 2019, 14:31:48) [GCC 8.2.0] on linux Type "help", "copyright", "credits" or "license" for more information. >>> help() Traceback (most recent call last): File "", line 1, in File "/usr/lib/python3.7/_sitebuiltins.py", line 102, in __call__ import pydoc File "/usr/lib/python3.7/pydoc.py", line 66, in import inspect File "/usr/lib/python3.7/inspect.py", line 40, in import linecache File "/usr/lib/python3.7/linecache.py", line 11, in import tokenize File "/usr/lib/python3.7/tokenize.py", line 33, in import re File "/usr/lib/python3.7/re.py", line 143, in class RegexFlag(enum.IntFlag): AttributeError: module 'enum' has no attribute 'IntFlag' >>> I've tried reinstalling python3.7 and its library packages without effect. Any hints on how I might find what's causing this pathological behavior? TIA! -Frank
Re: argparse missing from libpython3.7-stdlib ?
On Fri, 22 Feb 2019 22:00:01 +0100, Michael Lange wrote: > Hi, > > On Fri, 22 Feb 2019 20:15:12 - (UTC) > Frank Miles wrote: > >> It seems that the new testing/buster python3.7 lacks 'argparse'. >> Simply trying to import this causes an error, probably due to only a >> python2.7 version on my system. >> >> I didn't see any indication of a missing library, though there is >> apparently some kind of transition going on. Is my system missing some >> other python library? >> >> > according to > https://packages.debian.org/buster/amd64/libpython3.7-minimal/filelist > argparse should be in libpython3.7-minimal > > Regards > > Michael > > .-.. .. ...- . .-.. --- -. --. .- -. -.. .--. .-. --- ... .--. . > .-. > > Sometimes a feeling is all we humans have to go on. > -- Kirk, "A Taste of Armageddon", stardate 3193.9 Thanks, that's a good clue! There's something messed up with my installation. argparse is installed per libpython3.7-minimal. However if I simply start python, then do a help(), it gives me a traceback. Simpleminded forced reinstallation of python3.7, libpython3.7 idle-python3.7 idle3 libpython3-stdlib libpython3-minimal do not fix the problem :( Will explore further... Thanks again... -Frank
argparse missing from libpython3.7-stdlib ?
It seems that the new testing/buster python3.7 lacks 'argparse'. Simply trying to import this causes an error, probably due to only a python2.7 version on my system. I didn't see any indication of a missing library, though there is apparently some kind of transition going on. Is my system missing some other python library? Thanks for any insights! -F
firefox cannot download after jessie->stretch
firefox/iceweasel has worked just fine for years I just upgraded my home jessie machine to stretch. That went very smoothly, so far only one exception. I can no longer download anything with this browser. I've reinstalled, cleared out my profile, run in 'safe' mode (no add-ons such as no-script). I've tried other destinations, including "ask me" (it never asks). I only get the truncated error message: "could not be saved, because you cannot change the contents of that folder." Turned off my firewall - no change. Tried downloading from other sources, no change. Epiphany downloads the same files with no problem. Has anyone else had this problem recently ? Looking at the Debian bug reports I saw a year-old report that looked kind of similar. Thanks for any insights! -Frank
Re: SAMBA problems on Debian 8.8
On Sat, 15 Jul 2017 17:10:03 +0200, Jason Wittlin-Cohen wrote: > You are running an out-of-date version of samba (2:4.2.14+dfsg-0+deb8u6 > vs. 4.2.14+dfsg-0+deb8u7+b1). In addition, you seem to be missing > samba-common-bin, samba-dsdb-modules, and python-samba, all of which are > dependencies of samba. > > Try 'sudo apt-get install --reinstall samba' > Thanks, Jason - I had reinstalled samba-libs but not some other apparently critical part of samba. The --reinstall fixed it! -F (Not the OP, just another still-running-jessie guy)
Re: Efficiently finding information 'known' to exist "somewhere"
On Wed, 19 Apr 2017 16:10:01 +0200, Richard Owlett wrote: > I've had two instances recently. I've found the "immediately" needed > information, but they are samples of more generic problems. > > 1. Today's problem was easily solved. I had seen a post discussing an > application of the "tree" command. When I tried it, I got "command not > found". In _this_ case it was easily solved by using Synaptic's search > function -- there is a package named "tree". > > However that is not always the case. Some months ago I got a "command > not found" message for a command that had a man page (do not recall the > specific command). It turned out it was one utility command among many > provided by a package with an unrelated name. > > Is there a general way to find such a package? > > 2. There are many commands whose man pages point to using the "info" > command. I personally find that format more annoying than useful. I > would prefer to access the TeXInfo formatted document and convert it > locally to desired format - usually HTML. > > If the command is on my machine (i.e. GRUB), I can generally find the > associated TeXInfo formatted file (usually concealed in a tarred or > zipped file). How to search for all TeXInfo files on debian.org? One built-in way with your example is simply: man -k tree (so ok, not exactly the info format)
Re: AMD Ryzen support for Debian Stretch?
On Fri, 03 Mar 2017 00:50:01 +0100, Janis Hamme wrote: > Now that AMD's Ryzen CPUs have been released, I'm wondering if they'll > be supported by the upcoming Debian Stretch release. I'm a bit concerned > as Stretch comes with Kernel 3.9 but Ryzen support was added to 4.10. > > Is Debian known to backport hardware support from newer kernel versions? > Otherwise a custom or backports kernel would be required from the start. > Ryzen CPUs could become quite popular during the lifespan of Debian > Strech. It seems to perform fine for Linux applications so far: > > http://www.phoronix.com/scan.php?page=article=ryzen-1800x-linux=1 > > Regards > Janis It's a common occurrence (but hard to predict). Example: the current stable Debian can use a backported kernel (I'm doing this on my home desktop).
Re: qemu installed but not found
On Wed, 26 Oct 2016 18:20:07 +0200, Matyas A. Sustik wrote: > I installed qemu with apt-get. However the qemu command is not found. What > am I missing? Can anyone point me in the right direction? > > Thanks! > -Matyas I'm not sure what you've got. If it's only the qemu package, and not one of the many associated packages, you won't have an executable. If this is the case, you might get a better idea of what other packages you need using https://packages.debian.org/search?keywords=qemu=names=stable=all (there are other ways depending on how you manage your packages) HTH-
Re: Vim help, tags, sudo
On Mon, 24 Oct 2016 18:30:02 +0200, jeremy bentham wrote: > This could be a vim question, but since sudo's involved I'll > start here. > > I am thrashing about, trying to get wheezy going on a new machine > (well, new to me. I think the huckster term-of-art is > "pre-owned": I had the pleasure of wiping dollarbill inc's crap > off it). > > Anyway, I got vim installed, opened it up and wanted to look at > the docs for something. > > :h > > resulted in, "E433: no tags file" > > I've been using vim for a long time, upgrading as time went > along, and I'd never thought about the mechanics of help. So I > had to go looking. > > Ctags wasn't installed, so I got that and ran it on the vim doc > directory. It didn't make a tags file, but something that looks > like a config file for who-knows-what. > > So I looked at another machine, decided that the doc directories > were sufficiently similar (vim 7.1 versus 7.3) and copied the > tags file over. Kludgy, I know, and generally a Bad Idea (tm). > But I figured any damage would be limited to something already > broken > > It looks like a permissions problem, but I can't see any > difference in that area between the machines. (Admittedly, I'm > comparing Lenny to Wheezy, but it this instance should that make > any difference?) > > Now I have help with "sudo vi", but not as a normal user. > > I'm enquiring here because I'm wondering, is this symptomatic of > some other problem that's going to leap on me from a Very High > Place? > > Ok, I want my vim docs too! I think vim would sing, if you found > the right configuration, and I keep learning stuff, control-]'ing > about the help files. Having to keep a root session around and > switching to it just wouldn't be the same. And I wouldn't know > why it's not working the way it's supposed to. Which version of vim did you install? It wasn't the "tiny" version, right? Did you install vim-docs? (not sure if that's necessary)
Re: jessie-kvm-qemu: Win7 guest fails to update
On Wed, 13 Jul 2016 23:20:01 +0200, Linux-Fan wrote: > [Wed, 13 Jul 2016 16:06:43 - (UTC)] Frank Miles > <f...@u.washington.edu> wrote: >> I have two jessie systems with kvm-qemu virtualized Windows7 guest OSs. >> These are mostly working well (including guest inter-networking) >> to the extent that I use Windows, with one glaring exception: when I >> try to do a Windows Update - the process never finds anything to do, >> nor does it ever terminate. Ordinary Windows systems (belonging to >> other users) are finding lots of updates and fairly quickly. > > [...] > >> Has anyone else seen this? And better yet, found a solution? >> DuckDuckGo has not helped so far... > > How long ago was the last successful update? Was there ever one? I was trying to guess that. I'd estimate 1.5 - 2 months ago now. Maybe more. I used to be able to update without it taking all day or longer. > I have recently read [1] that upon reinstalling Windows 7 from scratch > Windows Update often ``takes hours'' (like say 12 hours or such) until > it has finally/successfully scanned for updates. The problem you > describe sounds very similar. IIRC it can be solved by installing a > certain Windows KB... manually and then scanning for updates another > time. IIRC the last time I updated it took quite a while. I was hoping that this would correct whatever was wrong and subsequent updates would go more smoothly. I also have a virtualized Linux installation hosted via kvm-qemu. This does its update without undue delays. > [1] > http://www.heise.de/ct/ausgabe/2016-5-Windows-7-Windows-Update-Beine- machen-3100633.html > > ...although the link does not contain a description of the solution, it > has further links (below ``WEITERFÜHRENDE LINKS'') where the correct > update to manually install is likely to be linked. > > HTH Linux-Fan Thanks, I'll take a look at it.
Re: jessie-kvm-qemu: Win7 guest fails to update
On Wed, 13 Jul 2016 21:40:02 +0200, Karl E. Jorgensen wrote: [snip] > Sounds like general network gremlins... A couple of things that spring > to mind: > > (1) Can the windows boxes ping e.g. 8.8.8.8 ? If not, then network > connectivity is likely broken... That works. > (2) Can the windows boxes resolve names? E.g. ping www.google.com ? > If not, then DNS resolution is borked... That works. > (3) IP address collisions: If two systems on the same network have the > same IP address, then you will get inconsistent results... To see > whether a box suffers from this, obtain it's IP address and > disconnect it from the network. If the IP address is still pingable > (or just arping'able), then this is a red flag... They're different - no address collision. > Hope this helps Nope, but thanks for the attempt.
jessie-kvm-qemu: Win7 guest fails to update
I have two jessie systems with kvm-qemu virtualized Windows7 guest OSs. These are mostly working well (including guest inter-networking) to the extent that I use Windows, with one glaring exception: when I try to do a Windows Update - the process never finds anything to do, nor does it ever terminate. Ordinary Windows systems (belonging to other users) are finding lots of updates and fairly quickly. One of these is my home computer (Win7Pro); the other is my work computer (Win7Enterprise). Both were afflicted approx the same time (not sure exactly when, I use the Win7 guests less than once/wk). I've talked to our network/user-systems techs/admins, and they haven't a clue. They recommend their usual solution - full reinstall of Windows. A hideous amount of time (ISTM) especially for my home system on a slow internet feed. Has anyone else seen this? And better yet, found a solution? DuckDuckGo has not helped so far... TIA for any clues!
Re: backported Kicad missing pcbnew?
On Fri, 01 Apr 2016 21:40:01 +0200, Sven Joachim wrote: > On 2016-04-01 19:14 +0000, Frank Miles wrote: > >> Had my hopes up when I saw the kicad package in backports. >> Unfortunately (at least on my jessie machine) it doesn't seem to >> include the pcbnew executable. From the bug reports, it would seem >> that it should be there & work. > > Looking at https://packages.debian.org/jessie-backports/kicad, > /usr/bin/pcbnew seems to be in the list of files for all architectures. > > Which architecture do you use? > > Cheers, >Sven Whoops. May have found my problem. I have a self-compiled version in /usr/local/bin. Don't know how that could have caused the kicad "graphical caller" to grey-out pcbnew. Will try that Monday... Thanks for your answer, it made me look again! -F
backported Kicad missing pcbnew?
Had my hopes up when I saw the kicad package in backports. Unfortunately (at least on my jessie machine) it doesn't seem to include the pcbnew executable. From the bug reports, it would seem that it should be there & work. Am I missing something somewhere? Hopefully yours, -F
Re: Where is digikam?
On Mon, 09 Nov 2015 18:40:03 +0100, Matthias Bodenbinder wrote: > Hi, > > I see digikam in unstable for a very long time now. But it is not coming to > testing. Why is that? Wasnt the push to testing supposed to be an automatic > process? Or does digikam in unstable still have critical bugs? What is the > reason for the delay? > > Matthias digikam exists in many debian versions, including testing. So maybe you want to revise your question? To find out why the version currently in sid hasn't made it to testing, go to the debian package page, and click on the 'developer information' link (right side of page). It will tell you why... HTH!
Re: Where is digikam?
On Mon, 09 Nov 2015 19:30:02 +0100, Lisi Reisz wrote: > On Monday 09 November 2015 17:53:05 Frank Miles wrote: >> On Mon, 09 Nov 2015 18:40:03 +0100, Matthias Bodenbinder wrote: >> > Hi, >> > >> > I see digikam in unstable for a very long time now. But it is not coming >> > to testing. Why is that? Wasnt the push to testing supposed to be an >> > automatic process? Or does digikam in unstable still have critical bugs? >> > What is the reason for the delay? >> > >> > Matthias >> >> digikam exists in many debian versions, including testing. >> So maybe you want to revise your question? > > It doesn't appear to be in Testing currently. > https://packages.debian.org/search?keywords=digikam=names=testing=all > https://packages.debian.org/search?keywords=digikam=names=all=all > > If you have it in a testing installation, it presumably was in testing and > has > been removed. > > Lisi Sorry - in my haste to describe how the OP could find the needed information I didn't notice that it wasn't in testing. Obscured by it being in all the other versions listed.
Re: Screen resolution in Jessie
On Wed, 29 Apr 2015 18:30:03 +0200, Gary Roach wrote: I just upgraded to Jessie with no problems. I use a 24 16x9 monitor. With wheezy, there was a 1920 x 1080? mode that gave the correct aspect ratio for the 16x9 screens. Jessie seems to only have the 4x3 1600x1200 mode. All of my circles are now elipses. Is there a solution to this problem. Is there a different driver out there. I'm using the on board video card on my Intel DP55KG mother board. I am certain that I didn't have this problem with Wheezy. Gary R. My 1920 x 1200 is working fine using the on-chip Intel video system and Jessie. Never tried the x1080. Can you alter the configuration using xrandr ? HTH.. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/mhr2dg$spg$1...@dont-email.me
Re: Foolproof disk device name in fstab
On Mon, 19 Jan 2015 03:50:01 +0100, andmalc wrote: I have a Jessie VPS with external disks attached. The disks are specified in /etc/fstab with traditional /dev/sdXX naming. I recently made changes to the disks that made a device name invalid but didn't notice. When I rebooted, the disk couldn't be found and boot halted in rescue mode. My question is: how can I specify devices in fstab so if they can't be found boot proceeds proceeds normally instead of halting? Would mounting with systemd with the 'device-timeout' option as described here be a good way? https://wiki.archlinux.org/index.php/fstab#Automount_with_systemd Regardless of whether you use systemd or some other init system, using UUIDs is supposed to be less susceptible. You can get the proper UUIDs using blkid() (see its man page). Use of UUIDs is at least partially explained in the fstab man page. HTH- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/m9i1bt$oub$1...@dont-email.me
Re: Moving LVM volume?
On Fri, 02 Jan 2015 21:10:02 +0100, Bob Proulx wrote:
Joel Rees wrote:
Bob Proulx wrote:
the disk as physical volumes for lvm. For you I might suggest:
/dev/sdb1 /boot {256M} /dev/sdb4 extended
{remainder}
Why extended? I generally put my LVM partition straight in a DOS
primary partition, unless I needed more than three non-LVM partitions
for some reason.
I sure appreciate everyone's recommendations, even if only for learning
things I was completely ignorant of (e.g. GPT).
What I've done is rebuilt from scratch without any LVM. IMHO LVM is
[a] overkill for my simple system; and [b] lacking in a few key functions.
I might have been able to move the LVM as originally wanted - by turning
off the LVM within gparted, it seemed like it might have worked. It
was estimated that it was going to take over 6 hours, and the system
had already shown some scuff marks, so I did the rebuild.
Initially I kept the root partition (including /boot) separate from /usr.
To my disappointment systemctl still reported that the system was
'degraded', indicating that it had 'failed to start Load Kernel Modules'.
So I merged /usr into the root partition - and now systemctl indicates
that the system is 'running' without any errors. I have the impression
that boot is faster but that will remain unproven since I'm not inclined
to restore the system to its split /usr-root partition state. It seems
most likely that splitting /boot and /usr into separate partitions would
not make systemctl (and whatever init process it is hosting) any happier.
I have a 'jessie' computer at work that produces the same error message
and it seems to work just fine (local server, running 24/7) so the
message seems to be fairly insignificant (so far).
Thanks again, all-
Frank
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/m87stb$9i2$1...@dont-email.me
Re: Moving LVM volume?
On Thu, 01 Jan 2015 11:30:02 +0100, Joe wrote:
On Thu, 1 Jan 2015 01:54:39 + (UTC)
Frank Miles f...@u.washington.edu wrote:
I recently added a new hard drive to my home system. I decided to use
it to create an all-new bootable 'jessie' system. I created a
partition table that I thought would be flexible:
/dev/sdb1 / (root) {7G} /dev/sdb2 /swap
{4GB} /dev/sdb3 /oldjunk{1G} /dev/sdb4 extended
{remainder} /dev/sdb5 LVM{one large volume}
Most of the partitions- /usr, /home, /var, ... were in LVM2.
What I've learned since then is that /usr seems to have special status,
and probably shouldn't be part of LVM as certain tasks early in the
boot process can't seem to access the interior of LVM.
I've moved 'oldjunk' into the LVM, and want to expand this partition to
become the new /usr. I've shrunk the LVM, but the freed space is all
at the far end of the LVM. I have been unable to move it towards the
end of the disk space,
so I can expand /dev/sdb3. gparted, resize2fs, pvmove,... (running
from a CDROM-based rescue disk) have all failed.
Is there some method that I've overlooked?
Is the system installed and running yet? If so, check the space used by
the main mountpoints. Almost certainly, /usr is the largest of the
system partitions. My workstation /usr is about 8GB, and I don't have
any modern games. Excluding /home, the total is just over 10GB.
Next, there's no problem having the entire system on LVM, including
/boot. I still have a /boot partition, for legacy reasons, but the rest
is in one LVM volume, indeed in a single partition apart from /home. On
a workstation, there's no great advantage to using separate partitions
for anything else.
Next, unless you want to mess with the building of the boot ramdisk, the
issue with /usr is that it must be mounted at the same time as the root
partition gets mounted during boot, so it needs to be physically stored
under /, and any separate /usr partition will still potentially have
problems. At the moment, I'm not aware of any show-stoppers caused by
having a separate /usr, but I've no doubt it will happen in time.
To be honest, unless you already have a significant investment in the
new system, I'd suggest starting again.
--
Joe
Thanks to everyone (Joe,Joel,Pascal,Tapani,Mark,...) for your interesting
replies. In response to your answers and questions:
I've been using the new-drive-system (jessie) for a bit over a week.
With the stock kernel it's mostly functional though 'systemctl status'
is whining about being 'degraded'. I probably wouldn't have discovered
that if my custom-kernel efforts* weren't failing in mysterious ways.
I was hoping that cleaning up the boot process would resolve the deeper
mysteries without having to think or work too hard.
/usr is currently taking ~11G. So there's not enough space to simply
merge the primary partitions and have / and /usr live on that. I do too
many different kinds of things to limit it - I expect it to grow, possibly
double in the next couple of years (and at that point this system will
probably get replaced anyway. the new disk is mainly to keep this old
system doddering along for a bit longer)
My proximal backup is the original disk, and I have tried to keep that
unaltered in developing the new system. I haven't (yet) backed-up to
my external backup drive - that's a bit of a pain to setup. I've had
drives fail on me, don't want to use that as /usr. I certainly wouldn't
trust a USB drive as /usr.
I guess I'm going to have to reinstall jessie from scratch. This will
take time as my network link is a slow DSL :( The alternatives don't
sound appealing. I have a lot of customization of my machines - hey,
if I just wanted to do what everyone else did I could run Windows.
So thanks again for your replies - I am at least satisfied that I've made
a reasonable effort to 'fix' the problems. I will have to think some
more about LVM - it seems like a great idea, but possibly not the best
option for my particular circumstances.
-Frank
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/m842nn$ml$1...@dont-email.me
Moving LVM volume?
I recently added a new hard drive to my home system. I decided to use it
to create an all-new bootable 'jessie' system. I created a partition
table that I thought would be flexible:
/dev/sdb1 / (root) {7G}
/dev/sdb2 /swap {4GB}
/dev/sdb3 /oldjunk{1G}
/dev/sdb4 extended {remainder}
/dev/sdb5 LVM{one large volume}
Most of the partitions- /usr, /home, /var, ... were in LVM2.
What I've learned since then is that /usr seems to have special
status, and probably shouldn't be part of LVM as certain tasks
early in the boot process can't seem to access the interior of
LVM.
I've moved 'oldjunk' into the LVM, and want to expand this
partition to become the new /usr. I've shrunk the LVM, but
the freed space is all at the far end of the LVM. I have
been unable to move it towards the end of the disk space,
so I can expand /dev/sdb3. gparted, resize2fs, pvmove,...
(running from a CDROM-based rescue disk) have all failed.
Is there some method that I've overlooked?
TIA!
-Frank
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/m829cv$kfq$1...@dont-email.me
Re: Is geda-gaf in debian?
On Sun, 21 Sep 2014 13:20:02 +0200, Tixy wrote:
On Sat, 2014-09-20 at 05:09 +, Frank Miles wrote:
[...] It's simply odd that the packages derived from geda-gaf do not
contain the gaf utility {none of the derived binary packages seem to
have it, not even geda-utils}.
Is 'gaf' meant to be a separate program? I'm running Debian Jessie and
seem to have all the utilities listed on
http://wiki.geda-project.org/geda:gaf
Yes - at least that's my understanding from the geda web-site. It says:
gaf(1) is a multipurpose command line utility implementing
setting up the above programs, exporting schematics and symbols
into various formats, and shell for command line processing of
their data.
See also: http://wiki.geda-project.org/geda:gaf_utility
What I want to be able to do is to print schematics without needing X[11]
running. I was able to make this work for a while using a utility
that simulated the X environment - but this stopped working nearly a
year ago, and I haven't had the time to see if I could restore it
somehow. 'gaf' provides a mechanism for printing that does
not require X.
Thanks for your interest!
-Frank
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/lvoarq$q4f$1...@dont-email.me
Is geda-gaf in debian?
Strange. Doing a web-search, I find a page for the geda-gaf package (old form at https://packages.qa.debian.org/g/geda-gaf.html which seems to say that geda-gaf is part of stable, testing, and unstable. There are update dates from earlier in 2014. But if I search on the Debian web-site, doing a package search for 'gaf' I only find 'gafitter'. Doing a direct package search either on wheezy or testing similarly finds no geda-gaf package. I'm confused - can anyone explain what is going on here? TIA! -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/lvibgf$3vf$1...@dont-email.me
Re: Is geda-gaf in debian?
On Sat, 20 Sep 2014 01:20:01 +0200, Cindy-Sue Causey wrote:
On 9/19/14, Cindy-Sue Causey butterflyby...@gmail.com wrote:
On 9/19/14, Frank Miles f...@u.washington.edu wrote:
Strange. Doing a web-search, I find a page for the geda-gaf package
(old form at
https://packages.qa.debian.org/g/geda-gaf.html
which seems to say that geda-gaf is part of stable, testing, and
unstable.
There are update dates from earlier in 2014.
But if I search on the Debian web-site, doing a package search for
'gaf' I
only find 'gafitter'. Doing a direct package search either on wheezy
or testing similarly finds no geda-gaf package.
I'm confused - can anyone explain what is going on here? TIA!
Will this help at all?
http://ftp.debian.org/debian/pool/main/g/geda-gaf/
On afterthought, tried apt-cache search on ANY form of it (geda,
gaf, OR geda-gaf).. That did bring up a couple geda packages, including
geda but not your sought after geda-gaf.
*hm* ? :)
Cindy
--
Cindy-Sue Causey Talking Rock, Pickens County, Georgia, USA
* I comment, therefore I am (procrastinating elsewhere) *
Ok, I'm blind. I had missed the 'source' aspect, which now makes
sense. It's simply odd that the packages derived from geda-gaf do
not contain the gaf utility {none of the derived binary packages
seem to have it, not even geda-utils}.
Thanks Cindy and Brian!
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: https://lists.debian.org/lvj26o$ofi$1...@dont-email.me
Re: The case of the read-only USB sticks.
On Sat, 22 Feb 2014 15:40:01 +0100, Hendrik Boom wrote: I have a problem with my USB sticks mysteriously becoming read-only. I decided to investigate. I bought three identical 8G USB sticks, identical except for colour). None of them appear have any switches on them. The first I used my Linux laptop to write a file into the top-level directory of the first stick: I mounted it, wrote it, and unmounted it. I handed it to my wife, who was to read it on her Mac. She told me it failed to even notice there was a USB stick plugged in. But returned to me, I could mount it and read it. I put the second into my Linux laptop, mounted it, listed the top-level directory (it was empty), unmounted it. I passed it to my wife, who plugged it into her Mac, and it immediately noticed the USB stick and allowed her to look at its contents. It was, of course, empty. I'm running Debian testing on an ASUS netbook. Speculation: Now this doesn't tell me anything about how my USB sticks turn read- only. But it does tell me that something weird is happening to them. Perhaps the two OS's have different ieas as to how USB sticks are to be written or read? Perhaps one of the other machined in the house it writing the in such a was that Linux can't read them? What do I need to know to investigate this. Has anyone else had problems like this? Online all I found was some people on Windows with read-only USB sticks. One of them said that some friend using Linux had fixed them. No one else had any luck. I have no idea if their experience has any relevance. -- hendrik You said you wrote to the top level directory. I'm guessing you were running as root and wrote to a section that you shouldn't have tampered with. For example, a drive might appears both as /dev/sdd and /dev/sdd1. You don't want to mess with /dev/sdd - loosely speaking, that's just for the partition table (i.e. use fdisk or one of its kin to alter if necessary). Read/write/mount only the /dev/sdd1. Of course the drive could have failed, but it seems unlikely. Have you tried to fsck the drive? HTH-- -F -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/leg008$p6k$1...@dont-email.me
Re: Vim config problem in jessie
On Sat, 15 Feb 2014 18:30:02 +0100, rpr nospam wrote: On Sat, 15 Feb 2014 04:05:48 + (UTC), Frank Mile wrote: I should have added one more interesting detail. If I comment out the (python) plugin call in /etc/vim/vimrc: if has(autocmd) filetype plugin indent on endif then the tabstops work without the manual override. Having, of course, lost all the nice stuff associated with the plugin. Just to make sure I hadn't messed up the plugin I commented-out the 'set ts=4' that I had appended to the plugin - no change in its behavior :( Frank, try this: $ locate python.vim /usr/share/vim/vim74/ftplugin/python.vim /usr/share/vim/vim74/indent/python.vim /usr/share/vim/vim74/syntax/python.vim I'd say your issue is caused by the fact that all that three scripts are loaded when you edit python files. According to vim help (see :help filetypes) the filetype plugin indent on command that you have enabled in /etc/vim/vimrc actually means the following: - vim tries to detect the type of the file that is edited (e.g. let's suppose the file type is my_file_type) - vim loads the appropriate plugin script: /usr/share/vim/vim74/ftplugin/my_file_type.vim - vim loads the appropriate indent script: /usr/share/vim/vim74/indent/my_file_type.vim Moreover, if the syntax highlighting is enabled in vim (:syntax on) then the appropriate syntax script is also loaded: /usr/share/vim/vim74/syntax/my_file_type.vim In /usr/share/vim/vim74/ftplugin/python.vim I see the following line: setlocal expandtab shiftwidth=4 softtabstop=4 tabstop=8 So, it seems that /usr/share/vim/vim74/ftplugin/python.vim is loaded last and overrides tabstop value set by other scripts. --rpr. Nice work, rpr! This explains it - down to the differences between jessie and wheezy or squeeze. I should be able to get it to work _my_ way... thanks! -F -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/ldpdpn$n1g$1...@dont-email.me
Vim config problem in jessie
I'm having problems setting the vim configuration -
and having it mean something - in a fairly new computer.
One simple example is the tab stops. Since I'm the
only direct user of this machine, I've simply edited
/etc/vim/vimrc, where I have the line:
set tabstop=4
In addition, I've appended the line
set ts=4
to some of the 'plugin' files, for example:
/usr/share/vim/vim74/syntax/python.vim
{the normal ending line with ts=8 has been commented-out;
otherwise the plugin file is unaltered from the Debian original}.
However when I open a python file, it's obvious that
8-space tabs are being used. Doing a set: shows that
softtabstop=4
But there is no
tabstop=4
whereas opening a 'C' file will achieve the latter (and
'proper' spacing.
The file is recognized as python - the keywords are highlighted
as expected. Manually entering the set ts=4 sets things right.
I've tried creating a ~/.vim/vimrc to no avail. I've not had
any problem with previous incarnations of Debian.
One has to suspect that some config file is overriding mine.
A quick examination of the installed vim packages using
dpkg -L ...
has turned up nothing other than links to /etc/vim/vimrc
Anyone have any hints regarding the source of this annoyance?
TIA!
-F
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/ldlnld$j3j$1...@dont-email.me
Re: Vim config problem in jessie
On Fri, 14 Feb 2014 20:50:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 06:35:25PM +, Frank Miles wrote:
I'm having problems setting the vim configuration -
and having it mean something - in a fairly new computer.
One simple example is the tab stops. Since I'm the
only direct user of this machine, I've simply edited
/etc/vim/vimrc, where I have the line:
set tabstop=4
In addition, I've appended the line
set ts=4
to some of the 'plugin' files, for example:
/usr/share/vim/vim74/syntax/python.vim
{the normal ending line with ts=8 has been commented-out;
otherwise the plugin file is unaltered from the Debian original}.
However when I open a python file, it's obvious that
8-space tabs are being used. Doing a set: shows that
softtabstop=4
But there is no
tabstop=4
whereas opening a 'C' file will achieve the latter (and
'proper' spacing.
The file is recognized as python - the keywords are highlighted
as expected. Manually entering the set ts=4 sets things right.
I've tried creating a ~/.vim/vimrc to no avail. I've not had
any problem with previous incarnations of Debian.
One has to suspect that some config file is overriding mine.
A quick examination of the installed vim packages using
dpkg -L ...
has turned up nothing other than links to /etc/vim/vimrc
Anyone have any hints regarding the source of this annoyance?
TIA!
-F
I use
~/.vimrc
and haven't experienced any of these issues.
--
Jeremiah Mahler
jmmah...@gmail.com
http://github.com/jmahler
Thanks .. but tried that too, no change.
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/ldltta$1lt$1...@dont-email.me
Re: Vim config problem in jessie
On Fri, 14 Feb 2014 23:10:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 08:22:02PM +, Frank Miles wrote:
On Fri, 14 Feb 2014 20:50:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 06:35:25PM +, Frank Miles wrote:
I'm having problems setting the vim configuration -
and having it mean something - in a fairly new computer.
One simple example is the tab stops. Since I'm the
only direct user of this machine, I've simply edited
/etc/vim/vimrc, where I have the line:
set tabstop=4
In addition, I've appended the line
set ts=4
to some of the 'plugin' files, for example:
/usr/share/vim/vim74/syntax/python.vim
The only entry in my ..syntax/pthon.vim with ts=8 is the modeline at
the bottom. This would only have an effect with python.vim is opened.
It would not effect the editing of Python files.
{the normal ending line with ts=8 has been commented-out;
otherwise the plugin file is unaltered from the Debian original}.
However when I open a python file, it's obvious that
8-space tabs are being used. Doing a set: shows that
softtabstop=4
But there is no
tabstop=4
whereas opening a 'C' file will achieve the latter (and
'proper' spacing.
The file is recognized as python - the keywords are highlighted
as expected. Manually entering the set ts=4 sets things right.
Are there any modelines in this file?
Does this issue occur with all Python files or just this one?
Can you post an example?
All python files with tabs; for example:
#!/usr/bin/python
def f() :
return 1
print f()
---
And yes, I know that some people consider tabs sinful.
Thanks so much for considering this!
I've tried creating a ~/.vim/vimrc to no avail. I've not had
any problem with previous incarnations of Debian.
One has to suspect that some config file is overriding mine.
A quick examination of the installed vim packages using
dpkg -L ...
has turned up nothing other than links to /etc/vim/vimrc
Anyone have any hints regarding the source of this annoyance?
TIA!
-F
I use
~/.vimrc
and haven't experienced any of these issues.
--
Jeremiah Mahler
jmmah...@gmail.com
http://github.com/jmahler
Thanks .. but tried that too, no change.
--
Jeremiah Mahler
jmmah...@gmail.com
http://github.com/jmahler
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/ldm65j$i5r$1...@dont-email.me
Re: Vim config problem in jessie
On Sat, 15 Feb 2014 01:00:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 10:42:59PM +, Frank Miles wrote:
On Fri, 14 Feb 2014 23:10:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 08:22:02PM +, Frank Miles wrote:
On Fri, 14 Feb 2014 20:50:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 06:35:25PM +, Frank Miles wrote:
I'm having problems setting the vim configuration -
and having it mean something - in a fairly new computer.
One simple example is the tab stops. Since I'm the only direct
user of this machine, I've simply edited /etc/vim/vimrc, where I
have the line:
set tabstop=4
In addition, I've appended the line
set ts=4
to some of the 'plugin' files, for example:
/usr/share/vim/vim74/syntax/python.vim
The only entry in my ..syntax/pthon.vim with ts=8 is the modeline at
the bottom. This would only have an effect with python.vim is
opened.
It would not effect the editing of Python files.
{the normal ending line with ts=8 has been commented-out;
otherwise the plugin file is unaltered from the Debian original}.
However when I open a python file, it's obvious that 8-space tabs
are being used. Doing a set: shows that
softtabstop=4
But there is no
tabstop=4
whereas opening a 'C' file will achieve the latter (and 'proper'
spacing.
The file is recognized as python - the keywords are highlighted
as expected. Manually entering the set ts=4 sets things right.
Are there any modelines in this file?
Does this issue occur with all Python files or just this one?
Can you post an example?
All python files with tabs; for example:
#!/usr/bin/python
def f() :
return 1
print f()
---
And yes, I know that some people consider tabs sinful. Thanks so much
for considering this!
I've tried creating a ~/.vim/vimrc to no avail. I've not had any
problem with previous incarnations of Debian.
One has to suspect that some config file is overriding mine.
A quick examination of the installed vim packages using
dpkg -L ...
has turned up nothing other than links to /etc/vim/vimrc
Anyone have any hints regarding the source of this annoyance?
TIA!
-F
I use
~/.vimrc
and haven't experienced any of these issues.
--
Jeremiah Mahler jmmah...@gmail.com http://github.com/jmahler
Thanks .. but tried that too, no change.
--
Jeremiah Mahler jmmah...@gmail.com http://github.com/jmahler
If I remove my ~/.vimrc and open a Python file I get ts=8 and sw=8. And
when I edit the file the tabs are 8 as expected.
I have not changed any of the system wide Vim configs in Debian.
If I create a ~/.vimrc with only one line
set ts=4
and then open the Python file again I will get ts=4, sw=8 and tabs will
be 4 as expected.
In your case the ts=4 setting is non-existent even if set in a ~/.vimrc.
This seem particularly odd.
What Vim versions are you running?
jeri@hudson:~$ dpkg -l vim* | grep ii ii vim2:7.4.161-1
amd64Vi IMproved - enhanced vi editor ii vim-common
2:7.4.161-1 amd64Vi IMproved - Common files ii vim-doc
2:7.4.161-1 all Vi IMproved - HTML documentation ii vim-gtk
2:7.4.161-1 amd64Vi IMproved - enhanced vi editor - with
GTK2 GUI ii vim-gui-common 2:7.4.161-1 all Vi IMproved -
Common GUI files ii vim-runtime2:7.4.161-1 all Vi
IMproved - Runtime files jeri@hudson:~$
--
Jeremiah Mahler jmmah...@gmail.com http://github.com/jmahler
My versions (look just like yours)
ii vim 2:7.4.161-1 amd64Vi IMproved - enhanced vi
editor
ii vim-common2:7.4.161-1 amd64Vi IMproved - Common files
ii vim-doc 2:7.4.161-1 all Vi IMproved - HTML
documentation
ii vim-runtime 2:7.4.161-1 all Vi IMproved - Runtime
files
ii vim-scripts 20130814 all plugins for vim, adding
bells and whistles
ii vim-tiny 2:7.4.161-1 amd64Vi IMproved - enhanced vi
editor - compact version
I'm mystified.
-F
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/ldmols$blj$1...@dont-email.me
Re: Vim config problem in jessie
On Sat, 15 Feb 2014 01:00:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 10:42:59PM +, Frank Miles wrote:
On Fri, 14 Feb 2014 23:10:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 08:22:02PM +, Frank Miles wrote:
On Fri, 14 Feb 2014 20:50:01 +0100, Jeremiah Mahler wrote:
On Fri, Feb 14, 2014 at 06:35:25PM +, Frank Miles wrote:
I'm having problems setting the vim configuration -
and having it mean something - in a fairly new computer.
One simple example is the tab stops. Since I'm the only direct
user of this machine, I've simply edited /etc/vim/vimrc, where I
have the line:
set tabstop=4
In addition, I've appended the line
set ts=4
to some of the 'plugin' files, for example:
/usr/share/vim/vim74/syntax/python.vim
The only entry in my ..syntax/pthon.vim with ts=8 is the modeline at
the bottom. This would only have an effect with python.vim is
opened.
It would not effect the editing of Python files.
{the normal ending line with ts=8 has been commented-out;
otherwise the plugin file is unaltered from the Debian original}.
However when I open a python file, it's obvious that 8-space tabs
are being used. Doing a set: shows that
softtabstop=4
But there is no
tabstop=4
whereas opening a 'C' file will achieve the latter (and 'proper'
spacing.
The file is recognized as python - the keywords are highlighted
as expected. Manually entering the set ts=4 sets things right.
Are there any modelines in this file?
Does this issue occur with all Python files or just this one?
Can you post an example?
All python files with tabs; for example:
#!/usr/bin/python
def f() :
return 1
print f()
---
And yes, I know that some people consider tabs sinful. Thanks so much
for considering this!
I've tried creating a ~/.vim/vimrc to no avail. I've not had any
problem with previous incarnations of Debian.
One has to suspect that some config file is overriding mine.
A quick examination of the installed vim packages using
dpkg -L ...
has turned up nothing other than links to /etc/vim/vimrc
Anyone have any hints regarding the source of this annoyance?
TIA!
-F
I use
~/.vimrc
and haven't experienced any of these issues.
--
Jeremiah Mahler jmmah...@gmail.com http://github.com/jmahler
Thanks .. but tried that too, no change.
--
Jeremiah Mahler jmmah...@gmail.com http://github.com/jmahler
If I remove my ~/.vimrc and open a Python file I get ts=8 and sw=8. And
when I edit the file the tabs are 8 as expected.
I have not changed any of the system wide Vim configs in Debian.
If I create a ~/.vimrc with only one line
set ts=4
and then open the Python file again I will get ts=4, sw=8 and tabs will
be 4 as expected.
In your case the ts=4 setting is non-existent even if set in a ~/.vimrc.
This seem particularly odd.
What Vim versions are you running?
jeri@hudson:~$ dpkg -l vim* | grep ii ii vim2:7.4.161-1
amd64Vi IMproved - enhanced vi editor ii vim-common
2:7.4.161-1 amd64Vi IMproved - Common files ii vim-doc
2:7.4.161-1 all Vi IMproved - HTML documentation ii vim-gtk
2:7.4.161-1 amd64Vi IMproved - enhanced vi editor - with
GTK2 GUI ii vim-gui-common 2:7.4.161-1 all Vi IMproved -
Common GUI files ii vim-runtime2:7.4.161-1 all Vi
IMproved - Runtime files jeri@hudson:~$
--
Jeremiah Mahler jmmah...@gmail.com http://github.com/jmahler
I should have added one more interesting detail. If I comment out the
(python) plugin call in /etc/vim/vimrc:
if has(autocmd)
filetype plugin indent on
endif
then the tabstops work without the manual override. Having, of course,
lost all the nice stuff associated with the plugin. Just to make sure
I hadn't messed up the plugin I commented-out the 'set ts=4' that I
had appended to the plugin - no change in its behavior :(
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/ldmp2s$dgr$1...@dont-email.me
Re: How well is wheezy supported by haswell platforms?
On Wed, 29 Jan 2014 10:50:01 +0100, M C wrote: Hi, I'm in the process of considering compute servers for a small HPC facility, and I would like to run Debian wheezy on them. Until recently I would have opted for Intel's Ivy Bridge based platforms, but with the advent of Haswell I'd like to consider that as an option. Searching the web reveals quite a few instances of people running into trouble with wheezy+Haswell, but many of those stories are now a few months old, so can I ask what the perceived situation currently is? Have most/all problems been ironed out, or would one need to backport a number of packages, and if so, which? Thanks for any help, m [snip html] I had problems with Wheezy, but Jessie is working pretty well. The answer is probably going to depend on your hardware, especially whether you're using the Haswell video stuff or a separate video card. In that former case I'm guessing you'll probably end up with Jessie - but I'd suggest you try it. It shouldn't take long to find out if there's something wrong on one of your units. -F -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/lcbdr0$2i5$1...@dont-email.me
Re: Printing in Jessie
On Tue, 17 Dec 2013 02:50:01 +0100, Zenaan Harkness wrote: On 12/17/13, Frank Miles f...@u.washington.edu wrote: I'm having serious difficulty trying to print PDFs in Jessie. Any insights as to where I should look? Is this something that jessie is doing to everyone, or is my configuration strangely mangled? Have you tried removing and re-adding the printer again (or simply add new printer with a new id/name for the printer)? Are you using gtklp,system-config-printer or some kde thing or something else? I've been using cups directly as previously described, not through any other interface. The good news is that in implementing your first suggestion I realized that I'd always been using the recommended Foomatic-Postscript driver in cups. I noticed that there was a gutenprint driver for this printer - and it _works_ with that driver! So it seems as though the Foomatic driver is broken, at least on this system. So thanks, Zenaan, for your suggestion - it led me to a reasonable solution! -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l8puq9$4ir$1...@dont-email.me
Re: Printing in Jessie
On Tue, 17 Dec 2013 18:00:01 +0100, Frank Miles wrote: On Tue, 17 Dec 2013 02:50:01 +0100, Zenaan Harkness wrote: On 12/17/13, Frank Miles f...@u.washington.edu wrote: I'm having serious difficulty trying to print PDFs in Jessie. Any insights as to where I should look? Is this something that jessie is doing to everyone, or is my configuration strangely mangled? Have you tried removing and re-adding the printer again (or simply add new printer with a new id/name for the printer)? Are you using gtklp,system-config-printer or some kde thing or something else? I've been using cups directly as previously described, not through any other interface. The good news is that in implementing your first suggestion I realized that I'd always been using the recommended Foomatic-Postscript driver in cups. I noticed that there was a gutenprint driver for this printer - and it _works_ with that driver! So it seems as though the Foomatic driver is broken, at least on this system. So thanks, Zenaan, for your suggestion - it led me to a reasonable solution! -Frank Well, spoke too soon. Evince is now causing the printer to emit many pages, each with a single line of gibberish - so its control is not yet right. Well at least I have reason to suspect the cups driver, not some more obscure part of the system. -F -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l8q2m9$7qf$1...@dont-email.me
Re: Printing in Jessie
On Tue, 17 Dec 2013 20:00:02 +0100, Brian wrote: On Tue 17 Dec 2013 at 17:46:49 +, Frank Miles wrote: Well, spoke too soon. Evince is now causing the printer to emit many pages, each with a single line of gibberish - so its control is not yet right. Well at least I have reason to suspect the cups driver, not some more obscure part of the system. You could equally as well suspect Evince's interaction with CUPS. I would try printing all your PDFs with lp after first altering cupsd.conf in /etc/cups to have LogLevel debug in it. Then compare successful and unsuccessful (if any) prints in /var/log/cups/error_log. Except that it was problematic with other printing sources, which (AFAICT) did not go through evince. Actually this second problem turned out to be a likely operator-error problem - somehow the driver got mangled from the gutenprint to a 'raw' driver. No idea how that happened, but it's restored now. The weird thing is that everything looked good in /etc/cups/printers.conf, but looking at the config via the web interface showed the 'raw' driver. Recreating the printer conf using the web interface yet again fixed this. Thanks for the tip for setting the cups loglevel though, that could be useful in the future! -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l8qdp0$hj7$1...@dont-email.me
Printing in Jessie
I'm having serious difficulty trying to print PDFs in Jessie. Printing PDFs works from gimp (gutenprint), but evince just causes blank pages. This is all with an old HPLJ5M (postscript) network printer. Similarly I can telnet into the printer and have random keystrokes print in Courier. Taking down my firewall changes nothing. Several other applications (not just evince) similarly generate blanks. I've also installed (on the same machine) a virtual Wheezy via KVM. Attempting to print from that also results in blank pages though this should be through cups on the virtualized system to the jet-direct card in the printer (bridged networking). The cups web interface and lpstat -tv say everything is good. My older system (this is a new machine) printed over the same network without difficulty (wheezy and older). I'm getting some hate mail from the system: dbus[4399]: [system] Rejected send message, 2 matched rules; type=method_return, +sender=:1.1 (uid=106 pid=4500 comm=avahi-daemon: starting up ) interface=(unset) member=(unset) error +name=(unset) requested_reply=0 destination=:1.35 (uid=1001 pid=3144 comm=/usr/bin/evince +testfile.pdf ) but so far my searching hasn't pulled up anything constructive relevant to dbus, avahi-daemon, evince,... Oh yes - this is with a 3.11 kernel, which I need to make the video system work (Intel haswell, video is on-chip). Any insights as to where I should look? Is this something that jessie is doing to everyone, or is my configuration strangely mangled? Thanks for any clues! -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l8o63u$396$1...@dont-email.me
Re: Growing number of packages not being upgraded
On Tue, 10 Dec 2013 16:20:01 +0100, Frank McCormick wrote: Each upgrade it seems there are a growing number of packages being held back by apt-get. It started off as a few packages a few weeks ago and now is 23: Calculating upgrade... Done The following packages have been kept back: cups-filters gir1.2-pango-1.0 libcdr-0.0-0 libdee-1.0-4 libgegl-0.2-0 libharfbuzz-dev libharfbuzz-icu0 libjavascriptcoregtk-1.0-0 libjavascriptcoregtk-3.0-0 libmateweather-common libmatewnck-common libmspub-0.0-0 libpango-1.0-0 libpango1.0-0 libpango1.0-dev libpangocairo-1.0-0 libpangoft2-1.0-0 libpangoxft-1.0-0 libtiff4 libtiff4-dev libtiffxx0c2 libwebkitgtk-1.0-0 libwebkitgtk-3.0-0 0 upgraded, 0 newly installed, 0 to remove and 23 not upgraded. root@frank-debian:/home/frank# How can I find out what is holding these packages ? Thanks Don't know if this will answer your question... if you search for a particular package on the Debian web site, then select Developer Information (PTS) on the right side, there's generally information on package status. HTH- Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l87f0e$3il$1...@dont-email.me
Re: New install (intel) can't video mode-switch - SOLVED
On Thu, 28 Nov 2013 05:30:02 +0100, Frank Miles wrote: On Thu, 28 Nov 2013 00:20:02 +0100, Georgi Naplatanov wrote: [snip] I have similar motherboard - ASUS H87-PRO, but a very kind man explained to me on this list that Haswell video won't work with Wheezy so I'm going to use my old Radeon video until Jessie become stable. In my opinion your options are: - use external/additional video card what is compatible with Wheezy - try to use VESA X.org driver with Haswell video - upgrade to testing (Jessie) - use another Linux distribution - check for other options HTH Best regards Georgi Thanks, Georgi! I'll probably try testing/Jessie, perhaps along with another system so that I can personally test updates before introducing them to this new system (which once up must stay running reliably...). -Frank Jessie (with its newer 3.11 kernel and newer libc6, perhaps other libs) works! A second attempt with Wheezy (was going to try again at a backported 3.11 kernel) failed. Whatever else is changed in the update, frame rate is less than 10% of what it was before (~60fps ~750fpm), but that's not critical for me. At least now I can login - use X apps - and logout while the server keeps chugging. Thanks Bob and Georgi! -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l7q8s3$iug$1...@dont-email.me
Re: New install (intel) can't video mode-switch
On Wed, 27 Nov 2013 07:30:02 +0100, Bob Proulx wrote: Frank Miles wrote: This is insane. I have a new system with an Asus H87M-Pro MB, an intel i4770t. Fresh install of wheezy. Things seemed mostly functional... but if I switched out of X to a terminal (Alt-Ctrl-Fn) the system would crash. In addition, shutdown would only start, then it would crash (even if X was never started). ... If anyone has some spare clues, I could use one about now! TIA! Even just where to start, what to look for would be appreciated. What graphics card do you have there? This could be any of several things. It sounds like radeon modsetting requiring the nonfree blobs. Do you have firmware-linux-nonfree installed? http://bugs.debian.org/696571 If that is it then try enabling nonfree in your sources.list file. deb http://http.debian.net/debian wheezy main contrib non-free apt-get update apt-get install firmware-linux-nonfree Not sure that lack of firmware-linux-nonfree is the problem but it seems like a good possibility. Bob There is no graphics card - the CPU does the graphics. A preliminary check: I got about 750fps on glxgears, which is plenty as my needs are computational, not gaming/graphics. Yes, I installed the nonfree firmware excepting a whole lot intended for wifi devices that aren't in my system. -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l755as$6fs$1...@dont-email.me
Re: New install (intel) can't video mode-switch
On Wed, 27 Nov 2013 19:00:03 +0100, Bob Proulx wrote: Frank Miles wrote: Bob Proulx wrote: Frank Miles wrote: but if I switched out of X to a terminal (Alt-Ctrl-Fn) the system would crash. In addition, shutdown would only start, then it would crash What graphics card do you have there? There is no graphics card - the CPU does the graphics. But if you didn't have a graphics adapter then you wouldn't have anything to plug the cable to the monitor into! :-) Almost certainly you are using the onboard one. In this case most likely the Intel integrated graphics controller. lspci lspci | grep VGA You should see some type of graphics controller listed. Probably an Integrated controller. Being integrated onboard it still resides on the bus. Yes, of course. I guess I took your question about a 'card' too literally. lspci - VGA compatible controller: Intel corporation Haswell integrated graphics controller (rev 06) Yes, I installed the nonfree firmware excepting a whole lot intended for wifi devices that aren't in my system. I would then check 'dmesg' for any errors. In particular I would look at the KMS (kernel mode setting) parts. In recent Linux kernels they have made massive changes to the console. It is possible, perhaps even likely, that the problem is KMS related. Other than an ACPI 'namespace lookup failure, AE_NOT_FOUND (which occurred in the past - if I try to boot without the rescue disk it doesn't leave anything new in the logs). This makes me suspect a disk-related problem (for the boot issue, not the video mode-switch), but how or why?? I haven't seen anything explicit regarding KMS, but I may have missed it - will look further. Other than this I have no ideas. Hopefully someone else will have a better suggestion. Bob Thanks for trying! I hope so too. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l75gpn$9gi$1...@dont-email.me
Re: New install (intel) can't video mode-switch
On Thu, 28 Nov 2013 00:20:02 +0100, Georgi Naplatanov wrote: [snip] I have similar motherboard - ASUS H87-PRO, but a very kind man explained to me on this list that Haswell video won't work with Wheezy so I'm going to use my old Radeon video until Jessie become stable. In my opinion your options are: - use external/additional video card what is compatible with Wheezy - try to use VESA X.org driver with Haswell video - upgrade to testing (Jessie) - use another Linux distribution - check for other options HTH Best regards Georgi Thanks, Georgi! I'll probably try testing/Jessie, perhaps along with another system so that I can personally test updates before introducing them to this new system (which once up must stay running reliably...). -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l76fsm$kv3$1...@dont-email.me
New install (intel) can't video mode-switch
This is insane. I have a new system with an Asus H87M-Pro MB, an intel i4770t. Fresh install of wheezy. Things seemed mostly functional... but if I switched out of X to a terminal (Alt-Ctrl-Fn) the system would crash. In addition, shutdown would only start, then it would crash (even if X was never started). I thought that perhaps it needed a newer kernel, so I loaded one from backports (3.10.0). Now the machine won't even boot. It gets to about 1.5(sec?) in rescue mode, with the last line ...switching to clocksource tsc. Then nothing except every 60(sec) it repeats some messages about the USB mouse. And something has been screwed up so that it will no longer boot with the old kernel (there were some other programs dragged in with the backported kernel; and there were some complaints about the Realtek firmware not being loaded, even when I loaded the nonfree realtek firmware package. If I use a 'rescue' CD to boot, switching to the machine's own filesystems (but the kernel is from the CD).In this mode, I can shutdown correctly. I still can't switch to a terminal without causing the system to crash (kernel 3.4.something). If anyone has some spare clues, I could use one about now! TIA! Even just where to start, what to look for would be appreciated. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/l73h5c$chv$1...@dont-email.me
Re: Gain owner of a file using vim :w!
On Thu, 23 May 2013 00:50:01 +0200, Beco wrote: Dear users, I'm astonished by this (maybe I'm naive and I'm missing something). Yesterday as root I saved a file skel.bashrc in my /home/beco user, owned by root, group root. Today I edited it, logged as beco, and vi told me warning, read only!. I edited anyway, just to test, and saved with :w! After that I checked the file and it has changed to owner beco, group beco. How is that possible? Thanks, Beco Did you want the modified file to remain owned by root? That would, of course, be _more_ dangerous! -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/knlgup$sno$1...@dont-email.me
Re: rpm
On Tue, 30 Apr 2013 18:40:01 +0200, ChadDavis wrote: I'm doing some software development that uses RPM packages. I would like to have RPM installed on my debian system for trivial and development only usage. In other words, I don't really plan to manage my system with it at all; i just want to use it for my dev purposes. My question is whether this is a problem. Will it somehow corrupt my deb system to simply install RPM and use it to play with the odd rpm package? div dir=ltrI#39;m doing some software development that uses RPM packages. I would like to have RPM installed on my debian system for trivial and development only usage. In other words, I don#39;t really plan to manage my system with it at all; i just want to use it for my dev purposes. div br/divdivMy question is whether this is a problem. Will it somehow corrupt my deb system to simply install RPM and use it to play with the odd rpm package?/divdivbr/divdivbr/div/div Have you tried the alien package? This converts rpms to debs, thereby not confusing the packaging system. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/klosak$vfv$1...@dont-email.me
Re: Console showing control characters for input on console keyboard
On Wed, 20 Mar 2013 18:50:02 +0100, francis picabia wrote:
After some random keys were hit on a KVM, the console is showing control
characters as the input. We switched KVM console and the PS/2 KVM
adapter, but the problem sticks with the system/OS. Killing all getty
to allow them to respawn has not helped.
ssh logins are fine. Only the console input is effected. It is showing
^R^O^O^T when 'root' is input (like a caps lock for control).
Many years ago I saw systems which would switch to all upper case if the
login input was given in upper case. This problem is similar.
Is there something which can be entered on the keyboard to reset this or
some process restarted? Obviously I'm trying to avoid the disruption of
a reboot on a server.
It seems like you may have sent some control sequence to the video
subsystem. Suggest that you look into resetting that. Try
man reset
which may get you started in the right direction.
{Years ago I'd occasionally do something similar via dosemu and some
marginal old DOS programs. Sometimes reboot was the only way out :( }
Good luck!
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/kicvt4$ocn$1...@dont-email.me
wpa_supplicant.conf - where?
The old Thinkpad finally having bit the dust, I got a cute new Asus ZenBook that seems (from other reports) to work with linux. Wiped windows, replacing it with Debian/testing (wheezy). Mostly it's working, with the most glaring problem being wireless. (iwlist wlan0 scanning) can find our router - showing all the address, channel, frequency, ssid/name,... but nothing seems to clue wpa_supplicant to it. So far my attempts at editing /etc/network/ interfaces has not helped. The wpa_supplicant.conf man page has many items of interest, but lacks _where_ this file is supposed to be. Creating this file in /etc/ wpa_supplicant/ seems to have no effect. Listing the package (dpkg -L ...) shows a wpa_supplicant.conf, but it's in /etc/dbus-1/system.d/, and looks to be in XML format. The man page says nothing about any XML file - where this is accurate or obsolete is uncertain. Issuing: wpa_cli status indicates that it's SCANNING, and list_networks shows the router. My son's Mac finds and logs into the wireless router with no problems. I'm probably missing something simple - if someone could point me to the docs or otherwise clue me into where wpa_...conf should be, I'd appreciate it! TIA- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1206200933220.31...@homer02.u.washington.edu
Re: kernel compile error in wheezy [solved!]
Thanks to everyone who helped - yes, it was indeed something about my environment. In /etc/profile, I've long exported an environment variable of the form: export LIBRARY_PATH=/home/myname/devel/lib:. I have a dim memory that when compiling cross-compilers that the '.' directory is a problem. Commenting this out allows the kernel compile to proceed. What's strange is that I've had this for many years without any kernel-compile problem. No longer! There may be some gcc docs that warn of this, not sure exactly where, but it might shed more light on this situation. Thanks again, everyone- Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1107182120300.6...@homer01.u.washington.edu
Re: kernel compile error in wheezy
Thanks to Stan, Stephen, and Maderios! - Stan wrote: Do you get the same error using the (new) Debian kernel method? $ make KDEB_PKGVERSION=custom.1.0 deb-pkg I'll have to learn more about the new method for the future. For right now, unfortunately the answer is yes, I get the same error. - Stephen wrote: ... I'm guessing that elks-libc is what is needed. Is that package installed on your system? ... No, it's not installed. Sure seems strange, requiring a 16-bit library for the build of a 64-bit system. I have stdarg.h from gcc versions 4.3, 4.4, and 4.6, and libstdc++ of the same versions. But it would be an easy solution, so... (installing,.. retrying...): - still fails :( - Maderios wrote: For my part: make-kpkg kernel_image --initrd and it works. Same error - it bombs on compiling arch/x86/kernel/asm-offsets.c, in each case looking for included file stdarg.h [again, I have at least 6 versions of this file depending on compiler]. Thanks for trying! I'm mystified, having compiled many kernels over the years on a various computers without difficulty. Is there any way that I could have mangled the include search path? I'm using bash as the shell interpreter, that's still ok isn't it? -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1107171013290.13...@homer01.u.washington.edu
kernel compile error in wheezy
I just tried compiling the kernel for My 'wheezy' system (2.6.39) [amd64]. As I've done many times - using make-kpkg --revision N kernel_image But with the recent linux-source update - shortly after starting I get: CC arch/x86/kernel/asm-offsets.s In file included from /usr/src/linux-source-2.6.39/arch/x86/include/asm/percpu.h:44:0, from /usr/src/linux-source-2.6.39/arch/x86/include/asm/current.h:5, from /usr/src/linux-source-2.6.39/arch/x86/include/asm/processor.h:15, from /usr/src/linux-source-2.6.39/arch/x86/include/asm/atomic.h:6, from ./include/linux/crypto.h:20, from arch/x86/kernel/asm-offsets.c:8: ./include/linux/kernel.h:12:20: fatal error: stdarg.h: No such file or directory compilation terminated. make[3]: *** [arch/x86/kernel/asm-offsets.s] Error 1 make[2]: *** [prepare0] Error 2 There don't seem to be any current bugs like this. A previous 2.6.39 compiler was uneventful. Any ideas would be appreciated! -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1107161104500.31...@homer02.u.washington.edu
Re: How to use serial ports?
Serial ports can be pretty opaque. Do you know if both serial ports were enabled in your BIOS? Can you find out if the relevant ports appear enabled in the system logs? Do you have any hardware diagnostic tools? i.e. one of those LED-based serial port monitors? Do you have flow control (RTS/CTS) disabled? HTH... -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1105130801530.3...@homer02.u.washington.edu
BUG DOSEMU I can't understand this but still exist
Dosemu should work fine with DPMI memory handling - if configured properly. Have you checked your dosemu.conf for its cpu_emu setting? HTH-- -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1104220802190.5...@homer02.u.washington.edu
Re: Squeeze update = disk unbootable
Firstly, thanks Tom for asking about the udev version. Since I
cannot boot the system I can't ask udev directly; however in
/var/cache/apt/archives, the only udev pkg is 164-3.
-
I think I'm closer to understanding the problem.
To clear up one thing: the system is partway to grub-2. The grub-2
chain completely bombs with a VFS... error. Altering the grub-1
menu.lst is what I've seemed to get. No matter what I do -
how I rephrase the disk reference, I get a
VFS: Cannot open root device...
error. From what I have gathered, this is probably due to my
using a custom-made kernel which is lacking in some key component
for booting with the new udev. Unfortunately I don't have any
of the standard precompile kernels on the system, so it may be
a bit awkward at best.
Perhaps I can manually mount the drives as needed into a pseudo-root
partition {having booted a rescue disk}, and chroot into the boot
partition, and load the kernel deb from a USB drive?
That's my next attempt... if there's a better way I'd like to
hear about it.
Thanks all! It's been a real learning experience...
-f
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive:
http://lists.debian.org/alpine.lrh.2.01.1102162133040.1...@homer02.u.washington.edu
Re: Squeeze update = disk unbootable
Thanks, Andrei. Unfortunately so far it hasn't worked. I've got the swap and all the ext3 partitions labeled (verified by cfdisk), and with the rescue disk boot the labels seem to work. I've modified /etc/fstab, replacing the original /dev/hdaX entries with the new LABEL=Y entries. But when I try to boot, the system fails to mount the drives. [By the way, grub has UUIDs. My initial attempts at changing this to labels hasn't been successful, so for now I've left the UUIDs in place. Does mixing UUIDs in grub with LABELs in fstab pose a problem?] The boot sequence first seems to fail at: findfs: Unable to resolve 'LABEL=ARoot' ...after which it fails on mounting the other partitions. One curious line is: EXT3 FS on hda2, internal journal This laptop has a single HD. I've commented non-linux partition information out of /etc/fstab. Any other thoughts would be appreciated! -f -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1102152214420.31...@homer01.u.washington.edu
Squeeze update = disk unbootable
I've just tried updating a seldom-used laptop from lenny (2.6.26 custom-compiled kernel) to squeeze. Everything seemed to be working perfectly until the reboot. Rebooting gives a repeated message fsck died with exit status 8. Using nano to rename the /etc/fstab entries /dev/hdaX - /dev/sdaX does no good - same message occurs. The kernel that it's trying to boot is still the 2.6.26. Booting from a rescue CD allows me to successfully e2fsck all the linux partitions - they all come out clean. The grub2 config references /dev/hdaX rather than /dev/sdaX. Is there some way to make this system self-consistently functional? Without scrubbing and starting all over? Thanks for any helpful hints! -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1102142021200.3...@homer01.u.washington.edu
Re: printing stopped after sid update
Can't be sure, but you might want to check the permissions of /usr/lib/cups/backend If the permissions are set to accessible only by root, try (as root): chmod go+rx /usr/lib/cups/backend HTH-- -f -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1007060814030.28...@homer01.u.washington.edu
fglrx blacked iceweasel?
Yesterday's squeeze update included the non-free fglrx driver. At least I think that is what is causing iceweasel to erratically blacken/erase/wipe many of the pages. Are others experiencing this? The browser has become almost completely unusable. Or was there some other update affecting the browser? -f -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1006301712290.20...@homer01.u.washington.edu
Re: Intel Core i5 integrated graphics
To Pasi: I'm using the xorg stuff - nothing special from Intel. Sorry for the ambiguity. To Stan: I didn't ask for, nor had I received any recommendation - you must be thinking of someone else. You had responded to a query I raised earlier when I had problems with the RealTek networking (which I thank you for), but that was some weeks after I bought the board. Before buying the board I had looked around for information on hardware compatibility, but didn't find much. This kind of information has an intrinsically short shelf life. For now the RealTek is working again, though needing a nonfree driver. -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1003240836490.27...@homer99.u.washington.edu
Re: Intel Core i5 integrated graphics
To Pasi: Sound has worked without any particular intervention - straight ALSA. I don't do anything complex with the sound system, though. As far as the video - while I'm presently using a custom-compiled kernel, I'm fairly sure I was getting the same rates with the stock 2.6.32 kernel available in 'testing'. I got a private e-mail telling me how I could do much better with a separate card - and I'm sure that's true. But [a] I'm not a gamer - so I don't really need super high rates - it is perfectly good enough for rotating 3-d math plots or views of devices; and [b] the low power consumption is important to me. Having a quiet machine is great! To Stan: I don't upgrade my computers very often. The last time I did, network cards seemed pretty much all operable with Linux drivers - including RealTek. This has been a real disappointment to find that something has gone backwards as far as Linux compatibility. -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1003230811250.20...@homer99.u.washington.edu
Re: Intel Core i5 integrated graphics
Hi Pasi: I have the same MB and processor. I have had a few issues, particularly with the RealTek chip, but as of now it's running fine - running the Debian/testing (squeeze) with the 2.26.32 kernel, and the RealTek driver out of nonfree. It's so annoying to have gotten a board thinking that I was avoiding the ATI/Nvidia proprietary mess, only to have to deal with this ugly RealTek problem. The straight Intel board may be a better choice for most. I needed the RS-232 port, which is available with the Gigabyte board (external connector added) but not (so far as I could determine) with the Intel board. I'm getting a bit less than 1000 fps with glxgears on a 1920x1280 monitor (DVI cable). With a moderate CPU load it's drawing around 50watts (not including display) according to my KillAWatt - pretty darn good for a fairly spiffy desktop! As Camaleon suggests, you may have to run a relatively new kernel to get this relatively new system running. Hope that helps... -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/alpine.lrh.2.01.1003220840040.17...@homer99.u.washington.edu
Re: eth0 - eth1 confusion vs. local network
Thanks so much to Stan, Tom H, and Cameleon! It seems that the consensus is that it's a NIC problem. In case it wasn't previously clear, the RealTek 8169 is part of the Gigabyte motherboard. I thought that I'd escaped non-free-firmware hell by getting a MB with the graphics based on an Intel chip. Never had a problem before, but then I usually stand far back from the bleeding edge. Stan, like you I usually use my own-build kernels. But I'd had problems getting my own kernels to run with full graphics capabilities, so had fallen back on the Debian 2.6.32-trunk. What I'm going to be doing in the short term is turning the RealTek off (BIOS setting), and installing another NIC. I should be able to get things running this way. I will post again once I've done this. Longer term, I'll try to get the RealTek running. All this flailing about has put me behind on other things, so that may not be right away. Thanks again to you all... it's been real educational. -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: eth0 - eth1 confusion vs. local network
I won't belabor this. Putting in a different NIC fixed things. No fuss, though interesting that it (presumably udev) wanted to call it eth2. I can live with that. Thanks again, everyone! -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
eth0 - eth1 confusion vs. local network
I am in the final throes of getting a new system running (Debian/squeeze). For the past 2 weeks it's had just eth0, and the network has worked fine. Now I want this system to have two network interfaces - the original eth0, and eth1 to a DSL modem, just like its precessor system. The strange thing is that wth the introduction of the second network card, the DSL connection works fine, but eth0 fails. A few relevant outputs: first, ifconfig shows (I masked the HWaddr's): eth0Link encap:Ethernet HWaddr xx:xx:xx:xx:xx:xx inet addr:192.168.0.10 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::6ef0:49ff:fe08:a40/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:198 errors:0 dropped:0 overruns:0 frame:0 TX packets:26 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:11880 (11.6 KiB) TX bytes:1916 (1.8 KiB) Interrupt:32 Base address:0x2000 eth1Link encap:Ethernet HWaddr xx:xx:xx:xx:xx:xx inet addr:192.168.1.33 Bcast:192.168.1.255 Mask:255.255.255.0 inet6 addr: fe80::260:8ff:fead:ec57/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:5811 errors:0 dropped:0 overruns:0 frame:0 TX packets:6876 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:4123103 (3.9 MiB) TX bytes:740692 (723.3 KiB) Interrupt:18 Base address:0xdf00 lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:212 errors:0 dropped:0 overruns:0 frame:0 TX packets:212 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:43609 (42.5 KiB) TX bytes:43609 (42.5 KiB) And route -n : Destination Gateway Genmask Flags Metric RefUse Iface 192.168.1.0 0.0.0.0 255.255.255.0 U 0 00 eth1 192.168.0.0 0.0.0.0 255.255.255.0 U 0 00 eth0 0.0.0.0 192.168.1.1 0.0.0.0 UG0 00 eth1 However, if I (even as root) try to ping another machine on the network: PING 192.168.0.4 (192.168.0.4) 56(84) bytes of data. ping: sendmsg: Operation not permitted ping: sendmsg: Operation not permitted ... ssh to the other machine similarly fails, though it worked when the new machine had a single interface. Similarly, I cannot ping or otherwise interact with the new machine from other machines on the local network. Strangely, arp -a shows: grebe (192.168.0.4) at xx:xx:xx:xx:xx:xx [ether] on eth0 rubberducky (192.168.1.1) at xx:xx:xx:xx:xx:xx [ether] on eth1 where 'grebe' is the machine that I tried to ping. It may have gotten the name from /etc/hosts, not necessarily via the network. No help if iptables rules are flushed. There is one troubling line in the logs from boot: udev: renamed network interface eth0 to eth1 Doing an ifdown eth1 does not fix the eth0 problem. So far, I haven't seen anything in the udev rules that might cause this. This is all with the 2.6.32-trunk kernel in squeeze. If anyone has any ideas on how to diagnose or fix this problem I'd appreciate it. TIA! -F -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: eth0 - eth1 confusion vs. local network
Thanks, Camaleon (sorry - don't know how to generate the proper characters).
That file includes:
# PCI device 0x10ec:0x8168 (r8169)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*, ATTR{address}==xx:xx:xx:xx:xx:xx, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth0
# PCI device 0x10b7:0x9050 (3c59x)
SUBSYSTEM==net, ACTION==add, DRIVERS==?*, ATTR{address}==xx:xx:xx:xx:xx:xx, ATTR{dev_id}==0x0,
ATTR{type}==1, KERNEL==eth*, NAME=eth1
On further checking, it may be that renaming is acceptable - in
/var/log/messages:
Feb 7 04:51:22 puffin kernel: [6.122403] 3c59x :03:02.0: PCI INT A - GSI
18 (level, low) - IRQ 18
Feb 7 04:51:22 puffin kernel: [6.122407] 3c59x: Donald Becker and others.
Feb 7 04:51:22 puffin kernel: [6.122411] :03:02.0: 3Com PCI 3c905
Boomerang 100baseTx at 0001df00.
Feb 7 04:51:22 puffin kernel: [6.148201] Linux agpgart interface v0.103
Feb 7 04:51:22 puffin kernel: [6.153035] udev: renamed network interface
eth0 to eth1
Feb 7 04:51:22 puffin kernel: [6.156559] r8169 Gigabit Ethernet driver
2.3LK-NAPI loaded
Feb 7 04:51:22 puffin kernel: [6.156573] r8169 :02:00.0: PCI INT A - GSI
17 (level, low) - IRQ 17
Feb 7 04:51:22 puffin kernel: [6.157040] eth0: RTL8168d/8111d at
0xc9c78000, x:x:x:x:x:x, XID 083000c0 IRQ 32
Feb 7 04:51:22 puffin kernel: [6.161239] r8169 :02:00.0: firmware:
requesting rtl8168d-2.fw
Feb 7 04:51:22 puffin kernel: [6.234448] eth0: unable to apply firmware
patch
Perhaps the kernel brings eth1 into existence by first establishing it as
eth0, then renaming it to eth1; then bringing the real eth0 into existence.
The unable to apply firmware patch seems potentially alarming, but it
used to work as a single-interface system. lspci -v indicates both
NICs have Kernel driver in use.
FWIW the kernel is the amd64 variant.
Any other thoughts? I should probably pull the 2nd NIC to verify that I'm not
delusional, or haven't bunged the configuration somehow, but I'm stumped in
solving this... especially since it is in most respects much like a previous
system.
Thanks, all!
-F
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: eth0 - eth1 confusion vs. local network
[snip] I fail to see what it's doing, but I cannot see any reference to eth1, it's like only one interace is being recognized :-? What is the output of dmesg | grep eth? [6.317161] eth1: RTL8168d/8111d at 0xc9c4e000, xx:xx:xx:xx:xx:xx, XID 083000c0 IRQ 32 [6.384830] eth1: unable to apply firmware patch [7.190453] udev: renamed network interface eth1 to eth0 [7.229390] udev: renamed network interface eth0_rename to eth1 [ 11.276999] r8169: eth0: link up [ 11.277005] r8169: eth0: link up [ 12.215716] eth1: setting full-duplex. [ 21.531029] eth0: no IPv6 routers present [ 22.599867] eth1: no IPv6 routers present Again, eth1 is working fine; eth0 seems completely blocked/nonfunctional, despite all the configuration files and netstats looking fine. I made a minor effort earlier to suppress the IPv6 modules, but [a] didn't succeed; and [b] hadn't suppressed them earlier with the one-interface system so wasn't convinced it was worth trying - why shouldn't this cause eth1 to quit as well as eth0? Also the previous system showed some indications of IPv6 in its reports, and it worked fine. Thanks, -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: eth0 - eth1 confusion vs. local network
... ok, started... [snip] I fail to see what it's doing, but I cannot see any reference to eth1, it's like only one interace is being recognized :-? What is the output of dmesg | grep eth? [6.317161] eth1: RTL8168d/8111d at 0xc9c4e000,xx:xx:xx:xx:xx:xx, XID 083000c0 IRQ 32 [6.384830] eth1: unable to apply firmware patch [7.190453] udev: renamed network interface eth1 to eth0 [7.229390] udev: renamed network interface eth0_rename to eth1 [ 11.276999] r8169: eth0: link up [ 11.277005] r8169: eth0: link up [ 12.215716] eth1: setting full-duplex. [ 21.531029] eth0: no IPv6 routers present [ 22.599867] eth1: no IPv6 routers present Again, eth1 is working fine; eth0 seems completely blocked/nonfunctional, despite all the configuration files and netstats looking fine. Errr, sir... something goes wrong here. As per your /etc/udev/rules.d/70-persistent-net.rules: eth0 - realtek eth1 - 3com But that is not what dmesg says above. That's the reason for my earlier fact-free speculation , based on a kern.log entry: Perhaps the kernel brings eth1 into existence by first establishing it as eth0, then renaming it to eth1; then bringing the real eth0 into existence. The kern.log entries don't appear in the dmesg output. Also, there is no link up or link down for eth1 but *both eth0 going up. Not sure how to interpret that. I don't know how to interpret that either, but the message is completely unchanged other than time - why assume it is referring to the 3com card in either case? And in any event, the 3com card is functioning - it's the realtek that isn't. I made a minor effort earlier to suppress the IPv6 modules, but [a] didn't succeed; and [b] hadn't suppressed them earlier with the one-interface system so wasn't convinced it was worth trying - why shouldn't this cause eth1 to quit as well as eth0? Also the previous system showed some indications of IPv6 in its reports, and it worked fine. I don't think this issue can have any relation with ipv6 :-?. How about your /etc/network/interfaces? Besides, you can make a quick probe by disabling eth1 and test if the network works as expected (ping et al) and then disable eth0 and perform the same test. I mean, test the network adapters separately. Greetings, -- Camale??n Tom H: Thanks for your queries also. I agree that the ipv6 warning is probably not an issue. What follows may help answer the questions that both of you have raised: - ...and diverted and continued... I decided to go back and re-establish the system when it only had a single NIC. Unfortunately this is the r8169, which has the possible firmware issue. I was _unable_ to get that working, though it had worked at one time. In the hopes of satisfying your curiousity, here are some reports from that experiment: puffin:~# netstat -nr Kernel IP routing table Destination Gateway Genmask Flags MSS Window irtt Iface 192.168.0.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0 puffin:~# arp -a grebe (192.168.0.4) at xx:xx:xx:xx:xx:xx [ether] on eth0 puffin:# route -n Kernel IP routing table Destination Gateway Genmask Flags Metric RefUse Iface 192.168.0.0 0.0.0.0 255.255.255.0 U 0 00 eth0 0.0.0.0 192.168.0.100.0.0.0 UG0 00 eth0 puffin:# ping 192.168.0.4 PING 192.168.0.4 (192.168.0.4) 56(84) bytes of data. ^C --- 192.168.0.4 ping statistics --- 3 packets transmitted, 0 received, 100% packet loss, time 2016ms puffin:~# iptables -L -n Chain INPUT (policy ACCEPT) target prot opt source destination Chain FORWARD (policy ACCEPT) target prot opt source destination Chain OUTPUT (policy ACCEPT) target prot opt source destination puffin:~# /sbin/ifconfig eth0 Link encap:Ethernet HWaddr xx:xx:xx:xx:xx:xx inet addr:192.168.0.10 Bcast:192.168.0.255 Mask:255.255.255.0 inet6 addr: fe80::6ef0:49ff:fe08:a40/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:18 errors:0 dropped:0 overruns:0 frame:0 TX packets:9 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1080 (1.0 KiB) TX bytes:594 (594.0 B) Interrupt:32 Base address:0x6000 loLink encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:69 errors:0 dropped:0 overruns:0 frame:0 TX packets:69 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:18577 (18.1 KiB) TX bytes:18577 (18.1 KiB) puffin:~# dmesg | fgrep eth [0.534289] Driver 'rtc_cmos' needs updating - please use bus_type methods [6.658237] eth0: RTL8168d/8111d at
Re: Daemon impotent after dropping priviledges with setuid()
I think I understand what's happening, though I'm not sure that it should...
After the daemon's double-fork (with setsid()), the utility of the group
identification seems lost.
With a bit more detail:
Say I have a directory owned by user user1, who is a member of
group1; and that user2 is also a member of group1. The directory
is set up to have group write permissions.
Before the double-fork, yada yada, all members of group1 can
write to this directory. After the double-fork, only user1 can write
to the directory.
It doesn't seem right, but that is what I observe.
The even stranger thing is that the program will append to an existing file...
just not create a new file. This seems wrong.
Am I ignorant? Does this happen on other people's machines? Here's a simple
program that should demonstrate the problem, if you substitute three names,
see below:
=== testuid.c
/* testuid.c Testing post-setuid() functionality.
*/
#includesys/types.h // pid_t
#includeunistd.h // getuid()
#includestdio.h
#includestring.h
// -- MAKE APPROPRIATE SUBSTITUTIONS FOR THESE:
#define DIR_FILE_NAME /full/path/filename
#define ALT_GRP_ID 1000// substitute group id
#define ALT_USER_ID 1000// substitute user id
// -- WRITE FILE TO PATH/NAME
#define MAX_FILE_LEN512
#define MAX_EXTN_LEN10
#define MAX_PARTIAL_FLEN(MAX_FILE_LEN-(MAX_EXTN_LEN+2))
static int writeDiagFile(const char *extn, const char *msg)
{
char fname[MAX_FILE_LEN];
FILE *f= NULL;
strncpy(fname, DIR_FILE_NAME, MAX_PARTIAL_FLEN);
fname[MAX_PARTIAL_FLEN - 1]= '\0';
strcat(fname, .);
strncat(fname, extn, MAX_EXTN_LEN);
fname[MAX_FILE_LEN-1]= '\0';
f= fopen(fname, a+);
if (!f)
{ fprintf(stderr, Unable to open file [%s] for writing by (%d,%d),
fname, getuid(),geteuid());
perror(fopen);
return -1;
}
fprintf(f, Msg {%s} = %s as %d\n, extn, msg, getuid());
fclose(f);
return 0;
}
int main(int argc, char * const argv[])
{
int retval;
pid_t xpid, ypid;
if (writeDiagFile(beginning user, 1))
return -10;
xpid= fork();
if (0 xpid)
{ perror(fork problem);
return xpid;
}
if (0 == xpid) // only child does more
{ if (writeDiagFile(child - after 1st fork, 2))
return -11;
setsid();
if (writeDiagFile(child - after fork/setsid(), 3))
return -12;
ypid= fork();
if (0 ypid)
{ perror(fork problem);
return ypid;
}
if (0 == ypid) // only child does more
{ if (writeDiagFile(child - after fork/setsid/fork, 4))
return -13;
retval= setgid(ALT_GRP_ID); // valid group on system
if (retval)
{ perror(setgid);
return retval;
}
retval= setuid(ALT_USER_ID); // valid user on system
if (retval)
{ perror(setuid);
return retval;
}
if (writeDiagFile(child - after fork/setsid/fork/setuid, 5))
return -14;
putc('\n', stderr);
}
}
return retval;
}
=== testuid.c
gcc -Wall testuid.c -o testuid
Run as root to be able to change user group.
Feel free to tell me about the stupid and obvious mistake[s].
I'm assuming that my use of stderr isn't the problem - the
earliest incarnation (in a daemon) didn't have the fprintf's,
but it had the same problem writing to a directory.
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Daemon impotent after dropping priviledges with setuid()
I have a peculiar problem with a daemon. Some of the tasks that the daemon needs to accomplish should be done with reduced priviledges, particularly if they are complex or depend on user input. So the daemon forks a child process, which setuid's to some lesser user. Unfortunately, as soon as the daemon setuid(some_other_userid), it can no longer write files. Not into the home directory of some_other_userid, not into /tmp, ... I haven't found anywhere that the daemon can write without receiving a permission denied error. The files didn't exist before the daemon action. Permissions to /tmp are 777+t. Doing a getuid(), geteuid() show that the permissions were set properly in the daemon's child process. The same user can write files into these same directories without error from the console or from a plain 'C' program. The daemon's child (before setuid) also succeeds at writing files. 'strace -f' doesn't show any problem until the open() call results in the permission denied error. This is on a 'squeeze' system, 64 bit i7 860 that otherwise seems to run properly. Anyone have any hints? Suggestions for diagnosing things? Would be appreciated!! -f -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
compiled C function for postgresql fails (64-bit only)
I have a function that I've used since early postgres-7.x days. Now migrating
to a 64-bit
(i7) machine - running 2.6.30/'squeeze' with postgresql-8.4.1-1. This function
is a simple
one that takes a string representation of a number as its sole argument
(engineering
notation), and returns the number. The floating point value that it returns is
the same
value {6.9484692498e-310} with every call, even when the string changes.
The same function works perfectly on a separate 2.6.30/'squeeze'
system, PG-8.4.1-1,
that is a 32-bit system - just like it has on previous postgres versions.
If I enable a diagnostic mid-functon dump-to-a-file, I can see that the
function
is receiving its input and computing the output properly - it's just a matter
of
getting the output translated into sql-land. This is with the Version 0
calling
conventions, with a return type of float8* . Attempting Version 1 calling
conventions
on the 64-bit system has not yet been successful - I haven't yet been able to
avoid a client
segfault, whether I use a stack variable for the float8 result, or use palloc().
In case it would help, the key parts of the function are:
float8 *sick_fct(const char *strng)
{
float8 *result;
char stack_string_buffer[BIG_STRING_SPACE];
// string copy operation : strng - stack_string_buffer
...
result= (float8*)palloc(sizeof(float8));
*result= inner_conversion_function(stack_string_buffer);
// printing this *result to a file shows that the conversion is correct
// the mysteriously wrong part seems to be:
return result;
}
I'm compiling it as:
g++-4.3 -c -I/usr/include -I/usr/lib/gcc/x86_64-linux-gnu/4.3/include
-Wall -Wshadow -Wpointer-arith -Wcast-align -I/usr/include/postgresql -I(yet
more includes) -fPIC -x c sick_fct.c -o sick_fct.o
and building the library as:
g++-4.3 -shared -Wl,-soname,lib_sickfct.so.1 -o lib_sickfct.so.1.0 *.o
and creating the function within PG as:
CREATE OR REPLACE FUNCTION sick_fct(text) RETURNS float
AS '/lib-path/lib_sickfct', 'sick_fct' LANGUAGE C STRICT;
Note that changing the float on the last line to double precision changes
nothing.
Odds are that I'm doing something stupid, but so far I haven't figured
out
what it is. Thanks for any insights you might have!
-f
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: debian-installer (squeeze-ia64) fails on new machine
On Fri, 23 Oct 2009, Dave Witbrodt wrote: Frank Miles wrote: Has anyone tried installing 'squeeze' recently on a i7/860 machine? [...] Any insights welcome! Isn't Core i7 an amd64 architecture, instead of ia64? I think you need to burn another installer CD, instead of that first coaster you burned. Ah...I will have to try that - re-reading the architechture description, I think you are quite right. Thanks!! -f -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
debian-installer (squeeze-ia64) fails on new machine
Has anyone tried installing 'squeeze' recently on a i7/860 machine? Trying the small-CD method - the line Loading Operating System ... comes up... and never goes away. No error messages, beeps, or any other complaints. The md5sum of the iso image is correct, and I've used 2 separately written CDs - same result. The system is ordinarily running 'lenny' without any problem (different partitions, eventually we hope). Oh...and 'reportbug' can't be used to report the problem, because the 'debian-installer' isn't installed ... :( Any insights welcome! -f -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: crond.daily: paralellel or sequential?
Hi Joost, You may be able to work around the sequential nature of cron jobs by the simple ruse of making a small script which backgrounds each job that you want to run contemporaneously. Something like: #!/bin/sh nohup jobToRun This script will return immediately while the job runs, allowing cron to start up the other jobs. Of course, cron has to call the script, not jobToRun. HTH-- -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: Second ethernet card seems to cause networking failure?
Thanks, Adrian. I was sure that it wasn't just the firewall at this point - that it was necessary to kill eth1 to get eth0 fully functional. I was wrong. This is now just a firewall problem. And this is a handcrafted beast, with lotsa rules. I wouldn't dream (well, not yet anyway) of asking anyone else to find the wart. Regrettably, I have a fire to put out - one of the ongoing experiments that we support is in dire need of some hardware we are developing - so I will have to defer that (and some clarification/simplification of the firewall) until after our hardware is out the door. Thanks again! -Frank -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: Second ethernet card seems to cause networking failure?
Following up, particularly on Adrian Levi's suggestion to eliminate the gateway spec in /etc/network/interfaces: Thanks, Adrian! Your idea makes sense. Trying it: it changes the routing table exactly as you described, causing my routing table to match yours (excepting, of course, the specific IP#s). Regrettably, the problem persists - though possibly with a different threshold of sorts, as pinging now seems to work. However- apt-get update still hangs. I have to kill BOTH the firewall and eth1 in order to make this work (not seeming to wait indefinitely for communication). I think the gateway correction is definitely a step in the right direction. In case someone want to know (I'm not trying to withhold information!) my /etc/apt/sources.list is very boring: deb http://http.us.debian.org/debian stable main contrib non-free deb http://security.debian.org/ stable/updates main contrib non-free Thanks to everyone for your ideas... hopefully the next one will resolve matters entirely. -f -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: Second ethernet card seems to cause networking failure?
Sure, can provide more info... /etc/network/interfaces : auto lo iface lo inet loopback # The primary network interface auto eth0 #iface eth0 inet dhcp iface eth0 inet static address xxx.yyy.zzz.32 network xxx.yyy.zzz.0 netmask 255.255.255.0 broadcast xxx.yyy.zzz.255 gateway xxx.yyy.zzz.100 pre-up /etc/iptables/iptables.sh start post-down /etc/iptables/iptables.sh stop # The secondary network interface auto eth1 #iface eth0 inet dhcp iface eth1 inet static address 192.168.42.100 network 192.168.42.0 netmask 255.255.255.0 #broadcast 192.168.42.255 gateway 192.168.42.100 === route result: Kernel IP routing table Destination Gateway Genmask Flags Metric RefUse Iface xxx.yyy.zzz.0 * 255.255.255.0 U 0 00 eth0 192.168.42.0* 255.255.255.0 U 0 00 eth1 === To reiterate: * The fundamental breakdown involves communication over the eth0 interface. Things just seem to hang when trying stuff like apt-get update. * ssh'ing into this machine from another host (directly to the IP of this machine) always works. * firewall is unchanged; well, ok, added: $IPT -A OUTPUT -o eth1 ! -s 192.168.42.0/25 -j DROP $IPT -A OUTPUT -o eth1 -s 192.168.42.0/24 -j ACCEPT $IPT -A INPUT -i eth1 ! -s 192.168.42.0/24 -j DROP $IPT -A INPUT -i eth1 -s 192.168.42.0/24 -j ACCEPT All communication with 192.168.42.x devices is functional. Listing iptables -L -n -v shows eth0 where it should. * simply turning firewall off (allowing everything) does not (at least by itself) fix eth0 communication. * as you can see, this is IPs are entirely static - no dhcp * network-manager not installed Since turning eth1 entirely OFF seems key to restoring eth0 full functionality, I agree that somehow the system seems confused about which interface to use. Any other thoughts/ideas welcome! -f * -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Second ethernet card seems to cause networking failure?
I recently added a second networking card to a hardware-test PC. This elderly machine had been working reasonably well. The second networking card is for eth1, etc., and /sbin/ifconfig shows things as properly connected, with eth0 being the outside interface and eth1 being an internal 192.168.x.x interface for some special internal systems that have absolutely no need to communicate with the outside world, just this one PC. The weird thing is that with normal booting configuration, pinging INet addresses fails. This seems to be related to the order in which the interfaces come up: doing ifdown eth1 ; ifdown eth1 ; ifup eth0; ifup eth1 causes pings to fail; but if eth0/eth1 are reversed (bringing up eth0 last), or eth1 is simply suppressed, pinging URLs works (i.e. ping www.debian.org). Regrettably this last does not entirely solve things - for example, I cannot do system updates: apt-get update fails to connect. Eventually, if I play around long enough (killing eth1, killing my firewall - which hasn't changed since before adding the second NIC,...) I can do a system update but it's not entirely clear what the critical steps were to get that working. I freely admit that I'm a hardware guy - I don't know much about networking. Does anyone have a suggestion on where I might look to get this working properly? Without sacrificing eth1? Or at least some better diagnostic[s] to track down where packets are getting lost? TIA! -f -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: python/smtplib fails on rebuilt system
[snip]
Florian has written:
Check if exim is really listening on port 25:
# netstat -plant | grep ':25 '
tcp0 0 127.0.0.1:250.0.0.0:* LISTEN
3271/exim4
Try
s=smtplib.SMTP('127.0.0.1')
to see if the problem is related to resolving localhost.
Very good and thanks, Florian! Though I have tried various ways to incorporate
localhost into my exim4 config, it is _not_ listening on 127.0.0.1.
netstat only shows the host IP address for port 25. Furthermore
the SMTP call works for the host IP address.
I'll clearly have to get the exim4 config sorted out.
Thanks again, Florian!
-f
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
python/smtplib fails on rebuilt system
A hard drive failure forced me to rebuild my main system. Just a few
things haven't been restored; one of them is a python script which is
used to email users of important events.
In attempting to diagnose the cause, I tried directly executing the
lines:
import smtplib
s= smtplib.SMTP('localhost')
but this causes an error with a traceback:
File stdin, line 1, in module
File /usr/lib/python2.5/smtplib.py, line 244, in __init__
(code, msg) = self.connect(host, port)
File /usr/lib/python2.5/smtplib.py, line 310, in connect
raise socket.error, msg
socket.error: (97, 'Address family not supported by protocol')
This is with exim4 and python2.5 on a newly installed lenny system.
No error messages appear in /var/log or /var/log/exim4 directories.
Helpful clues or pointers to relevant documentation would be
appreciated!
-f
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Re: Unable to mount cdrom in Lenny
[snip] Does /dev/scd* exist? -- Ron Johnson, Jr. Jefferson LA USA Ron! Thanks! It does indeed, and is apparently the proper mount point for the cdrom, regardless of the lines in /var/log/dmesg. Not to seem unappreciative, but: how should I have known about this? What did I miss? Again, Ron, thanks your question was perfect. -f -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Unable to mount cdrom in Lenny
Why can't I mount cdrom?
Running Lenny with custom 2.6.26 kernel. I don't often read CDROMs, so I don't
know when this failed (yes, it used to work). Putting a CD into the drive no
longer
causes udev to provide a /dev/hda device. Kernel modules loaded (lsmod)
include:
ide_cd_mod
cdrom
sr_mod
Looks good to me. Also /etc/udev/rules.d/70-persistent-cd.rules has:
# BENQ_DVD_DD_DW1620 (pci-:00:06.0-ide-0:0)
ENV{ID_CDROM}==?*, ENV{ID_PATH}==pci-:00:06.0-ide-0:0, SYMLINK+=cdrom,
ENV{GENERATED}=1
ENV{ID_CDROM}==?*, ENV{ID_PATH}==pci-:00:06.0-ide-0:0, SYMLINK+=cdrw,
ENV{GENERATED}=1
ENV{ID_CDROM}==?*, ENV{ID_PATH}==pci-:00:06.0-ide-0:0, SYMLINK+=dvd,
ENV{GENERATED}=1
ENV{ID_CDROM}==?*, ENV{ID_PATH}==pci-:00:06.0-ide-0:0, SYMLINK+=dvdrw,
ENV{GENERATED}=1
which seems reasonable, given the info. In /var/log/dmesg:
NFORCE-CK804: :00:06.0 (rev f2) UDMA133 controller
NFORCE-CK804: IDE controller (0x10de:0x0053 rev 0xf2) at PCI slot :00:06.0
NFORCE-CK804: not 100% native mode: will probe irqs later
NFORCE-CK804: IDE port disabled
ide0: BM-DMA at 0xe000-0xe007
Probing IDE interface ide0...
hda: BENQ DVD DD DW1620, ATAPI CD/DVD-ROM drive
hda: host max PIO5 wanted PIO255(auto-tune) selected PIO4
hda: UDMA/33 mode selected
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide-scsi is deprecated for cd burning! Use ide-cd and give dev=/dev/hdX as
device
Is the IDE port disabled significant? Even though a few lines later hda and
ide0 seem
to be set up properly? I haven't found anything missing from the kernel
configuration.
Unfortunately, this is on a server system that I'd rather not put through a lot
of reboot cycles.
Any ideas would be appreciated!
-f
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Keychain use no longer avoids password requirement
I have a simple backup system : a remote system periodically tars some important data, and notifies a server that it should read those files. The server then tries to read the files: su - --command=scp URL:/path/filename.tjz /backup-path/filename.tjz backupUser Until mid-May, this was working great. Keychain had been invoked so that this could operate without user intervention. However now the scp operation is asking for a password, which the script does not have. If done manually, the operation works. In trying to resolve this, I've not found how to interrogate the ssh-agent to determine what keys it has access to. So first question - is there some way to determine that? One probably-more-than-a-coincidence is that the failure began on rebooting the remote system, probably due to a kernel/security update. It's running Etch/2.6.18. As it would seem to be the server rather than the remote system that was rebooted, I'm mystified as to why any kernel change (and these are self-compiled kernels) on the remote system would have any effect on this operation. Any clues - including references to man pages possibly overlooked - would be appreciated. TIA! -f -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Cannot write file 1GB to DVD-RAM : any way to fix?
I changed how I was backing up my system, and now need to write larger files to my Panasonic LF-D521 DVD-RAM (backup drive). Unfortunately it chokes a bit over 1GB, reporting that the file size is too large. 'umount'ing takes a really long time as well. The disk[s] have been formatted with the udf file system. This problem occurs with each of the several disks that I tried. Any suggestions on how I can store larger files? TIA for any hints or pointers to Fine Manuals! -frank -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Scanner won't work under gimp
Hello... I had to rebuild my home machine (hardware failure). Since the rebuild (with Etch, same as before), neither scanimage nor gimp seem to find my Epson/USB scanner. Running sane-find-scanner finds the scanner without any problem. Running scanimage -L hangs. Turning on the debug environment variables SANE_DEBUG_DLL and SANE_DEBUG_EPSON show that the scanner is found, but scanimage never exits (until I hit Ctrl-C). If I explicitly specify the backend, scanimage works. Otherwise it hangs. I've provided the appropriate link to the scanner backend through the link as recommended on the man page. I'm guessing that what is lacking is a link or parameter setting somewhere that will route gimp to the correct scanner backend. Can anyone suggest where that might be? (Many of the man pages seem out of date with the changeover to the temporary /dev entries). TIA!! -frank -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
