Using (old) Ubuntu libjpeg packages to allow some older programs still running on recent debian
The subject is considered as dealing with Utnubu - but according to the appropriate wiki page, this team has stopped to exist "many years ago". Therefore I write in this list. Next I describe a case which make me think that Debian has less a focus on keeping less recent lib packages part of more recent debian versions. More important: two other packages part of both Ubuntu and Debian prove to result in removal of old programs on Debian, but not on Ubuntu! Regarding better backward compatibility: libjpeg8 and libpeg62 exists both in Ubuntu (till today); but in Debian only the latter. In my case I had to install libjpeg8 from ubuntu on my debian-based system (as well as gstreamer0.10-plugins-good) Yesterday, I experienced that installation of jhead on a Debian based-machine results in installing just a slightly different version of libjpeg-turbo-progs and libturbojpeg0 (compared to Ubuntu). In my case this would also incur removals of libjpeg8 and libjpeg-turbo8 & gstreamer0.10-plugins-good, which I "imported" some time ago from Ubuntu packages in order to continue the use of old application stuff. Doing the same (installing jhead on an ubuntu-based machine) does not affect anything. But: when apt install jhead is executed on Debian 11 *after* the next two Ubuntu packages are installed via dpkg, no complaints are made. To be precise: Debian| Ubuntu libjpeg-turbo-progs 1:2.0.6-4 | libjpeg-turbo-progs 2.0.6-0ubuntu2 libturbojpeg0 1:2.0.6-4 | libturbojpeg 2.0.6-0ubuntu2 Please take notice of the extra 0-suffix in the libturbojpeg name from the Debian 11 repository. My actual question: what does the Debian-version of these two libs make incompatible with libjpeg8 and three other libs - all "imported" from Ubuntu? A practical problem arises with these two libs. When apt upgrade (or e.g. use KDE discovery) is being executed, the system will try to "upgrade" these two packages to the original Debian version. And that's annoying because I must decline those upgrades in order to prevent the automatic removal of four lib packages and one (old) application... Regards, Rob
Re: Fwd: Re: ASTM Lab equipment protocol
Hi there On 12/07/2021 15:40, Markos wrote: OK. Now I understand a little bit more the situation. You are referring to this pattern: https://www.astm.org/Standards/E1381.htm I didn't know this standard. What I have seen in my experience with laboratory automation with some instruments/equipment is that each manufacturer has its own ASCII communication protocol. Therefore, it is more viable to use some standard to specify a communication interface but without specifying the specific protocol of each equipment/instrument. The only ASTM standard for laboratory automation with this approach (define interface) was the LECIS standard, which is now discontinued: https://www.astm.org/DATABASE.CART/WITHDRAWN/E1989.htm I discovered, a few days ago, the SiLA project to standardize automation in the laboratory: https://sila-standard.com/ But I still don't know the details. I have already had to implement programs for serial communication with instruments and equipment in the laboratory. But I don't remember seeing any Debian packages related to this ASTM standard. The thing to do is probably to use github search and then download, install and test each relevant program. I found two which are programmed in C, so I can modify them myself if need be. The vet however is perfectly happy with using the native skyla protocol, so I used this instead. They are already using my software and perfectly happy with it. The software simply dumps the data in a text file which they then can access. Nothing fancy here. There is an other blood analyses machine heading this way, with yet an other protocol for which I will also write software. In the long run ASTM is probably the way to go. With lots of googling you can actually find bits and pieces of the standard without having to pay for it. I've found some programs for acquisition and graphic display of data received through the serial port. But as they didn't have all the resources I decided to develop my own programs using the Tcl/Tk language, always using Debian. I tried to document some of these projects on my website, which might be useful. The material is in Portuguese but it is possible to translate it with the help of Google Translator: https://tinyurl.com/46vnrp68 Email-me if you have any question. Regards, Rob
Re: ASTM Lab equipment protocol
Hi On 09/07/2021 20:43, Markos wrote: Em 09-07-2021 10:21, Rob van der Putten escreveu: Which Debian packages support the ASTM lab equipment (over TCP) protocol? An overview would be nice. Please, explain with more detail, and some example, what exactly are you looking for? The sister of a friend is a vet. She has blood analyses equipment which sends analysis results to a serial port. Each brand and model has it's own protocol. Recently she acquired Skyla VB1. This device has both a serial port and Ethernet. It sends a ^A to the serial port, expects a ^F and then sends the data followed by a ^D, after which it expects an other ^F. I did manage to get data from the device this way. The device can also use the ASTM protocol. With ASTM it can also use Ethernet. Which is more practical. I could not find a free version of the ASTM standard. Apparently, there is some Linux software which supports ASTM, but what I could not find is a nice overview. Regards, Rob
Re: ASTM Lab equipment protocol
Hi there On 09/07/2021 19:11, Brian wrote: On Fri 09 Jul 2021 at 19:04:06 +0200, Rob van der Putten wrote: On 09/07/2021 18:23, Polyna-Maude Racicot-Summerside wrote: On 2021-07-09 9:21 a.m., Rob van der Putten wrote: Which Debian packages support the ASTM lab equipment (over TCP) protocol? An overview would be nice. Have you tried usign the packages.debian.org search engine for going thru the packages descriptions ? Possibly using astm keyword ? ASTM is not necessarily in the description. That should give you some idea If only things could be that easy. The debian-med and/or the debian-science mailing list couls help out. https://lists.debian.org/debian-med/ https://lists.debian.org/debian-science/ Now that is easy. I did not know about the archives. Or the search. No results though. So this might be a bit of a dead end. It doesn't mean that software doesn't exist. Just that ASTM was never discussed on a Debian list. Regards, Rob
Re: ASTM Lab equipment protocol
Hi there On 09/07/2021 18:23, Polyna-Maude Racicot-Summerside wrote: On 2021-07-09 9:21 a.m., Rob van der Putten wrote: Which Debian packages support the ASTM lab equipment (over TCP) protocol? An overview would be nice. Have you tried usign the packages.debian.org search engine for going thru the packages descriptions ? Possibly using astm keyword ? ASTM is not necessarily in the description. That should give you some idea If only things could be that easy. Regards, Rob
ASTM Lab equipment protocol
Hi there Which Debian packages support the ASTM lab equipment (over TCP) protocol? An overview would be nice. Regards, Rob
Does anyone else use Bibus
*Bibus* I'm using Debian Stretch and have been for some time. For bibliographic work, I want to use Bibus. If anyone else out there is using this useful bit of software I'd like to know, or if anyone knows of a mailing list to discuss issues, I'd also like to know. My problem with Bibus is to do with editing styles. In the "Fields Ordering" tab I am unable to delete any of the items (such as "Author") that refer to database entries. I can insert new entries, but can't delete ones I don't want. Is this a known bug? *Python* Thinking that Bibus using Python 2.7 might be the problem, I installed Python 3.5: debianrob [14:42] /usr/bin#rm python debianrob [14:42] /usr/bin#ln -s python3 python but Bibus did not like it: debianrob [14:40] ~>bibus & [1] 7175 debianrob [14:50] ~>Traceback (most recent call last): File "/usr/bin/bibus", line 21, in import os, sys, gettext, locale, ConfigParser, csv ImportError: No module named 'ConfigParser' [1]Exit 1bibus (python 3.5 has a "configparser" module, but linking "ConfigParser" to it raises other issues and I'm not competent with python). Any ideas, or direction to other sources of help would be most welcome. Thanks heaps, Rob Hurle - Rob Hurle e-mail:rob1...@gmail.com Mobile: +61 417 293 603 (Australia) Telephone: (02) 6236 3895 28 Mirrormere Rd, Burra, NSW 2620, Australia
Re: How to write in Chinese
I use Debian 9 and have installed iBus. It is possible to type Vietnamese, Chinese (in pinyin and/or characters) and English. Many other languages are available as input methods. The method of swapping between the language is configurable (I use alt/space). Been very reliable. 你 好 Rob Hurle - Rob Hurle e-mail:rob1...@gmail.com Mobile: +61 417 293 603 (Australia) Telephone: (02) 6236 3895 28 Mirrormere Rd, Burra, NSW 2620, Australia On Fri, 20 Nov 2020 at 15:00, Markos wrote: > Please, > > I am studying Chinese and I need to install fonts to type ideograms in > Debian 9. > > I use Brazilian Portuguese, but I don't want to reconfigure my computer > to type only in Chinese. > > Using LibreOffice or with a specific program to generate the ideograms > only when needed. > > Any tip? > > Thank you, > > Markos > >
Re: technical terms overhaul
Hi there On 19/06/2020 21:14, Eike Lantzsch wrote: On the danger of starting a flame war ... thinking about the article by Gunnar Wolf on Planet Debian instead of "whitelist" and "blacklist" I would like to propose the terms: "allowlist" and "rejectlist" instead of (for example on disk drives) "master" and "slave" I like to propose the terms: "lead" and "lag" which are e.g. used when electrically managing two pumps working on the same hydraulic circuit. What do you think? How about parent, child, kill, zombie, cooked, raw, sane. Male, female, bisex connectors. Red and black. In a classic concurrent server, a client connects. The parent creates a child. The child does all the work (child labour). After the work is done the child commits suicide or gets killed by the parent. These metaphors are not necessarily very nice. The main thing is that they are clear. Regards, Rob
Re: IPv4 v IPv6
Hi there On 17/06/2019 12:11, Aidan Gauland wrote: On 17/06/19 9:09 PM, to...@tuxteam.de wrote: On Mon, Jun 17, 2019 at 10:05:11AM +0100, mick crane wrote: hello, I know nothing about IPv6. Can somebody point to a good explanation ? I'd recommend skimming the relevant Wikipedia [1] page. Cheers [1] https://en.wikipedia.org/wiki/IPv6 I don't entirely agree that's a good introduction for someone without any background knowledge in TCP/IP, but certainly a good resource when one has some of the basics already. To answer OP's questions, Without knowing anything about it I'm wondering if I should request an IPv6 range from my ISP to use locally. Given that you do not already have a reason to do so, I would say not. A network card have IPv4 and IPv6 addresses that are different, not the same address in different notation? A network card (also called an interface) can have both, and they are not the same address. IPv4 and IPv6 are two different network stacks, with different addressing, routing, etc. Then with firewalling do you need to specify both IPv4 and IPv6 ranges? Yes, for the same reason as above. Sadly, most of the world is still on IPv4, so until IPv6 is deployed across the majority of the Internet, then if you're not a large entity, there's not much reason to use IPv6 other than playing around. Here's a gentler introduction to IPv6 that might also help: http://www.steves-internet-guide.com/ipv6-guide/ I really like this one; http://tldp.org/HOWTO/Linux+IPv6-HOWTO/ Regards, Rob
Re: please do *never* use GitHub for free software, was Re: Salsa vs Github
Hi there On 17/05/2019 09:28, Dominik George wrote: please do*never* use GitHub for free software Please explain, in detail, why. If discrimination against parts of the community is not enough for you, here's why: https://mako.cc/writing/hill-free_tools.html An overview; https://en.wikipedia.org/wiki/Comparison_of_source_code_hosting_facilities https://www.adamhyde.net/another-good-reason-not-to-use-github/ https://www.mirbsd.org/permalinks/wlog-10_e20170301-tg.htm Regards, Rob
iotop not working
Hi List,
I run Stretch, I've just installed iotop and when I try to run it I get
this:
# iotop
Traceback (most recent call last):
File "/usr/sbin/iotop", line 17, in
main()
File "/usr/lib/python3/dist-packages/iotop/ui.py", line 620, in main
main_loop()
File "/usr/lib/python3/dist-packages/iotop/ui.py", line 610, in
main_loop = lambda: run_iotop(options)
File "/usr/lib/python3/dist-packages/iotop/ui.py", line 508, in run_iotop
return curses.wrapper(run_iotop_window, options)
File "/usr/lib/python3.5/curses/__init__.py", line 94, in wrapper
return func(stdscr, *args, **kwds)
File "/usr/lib/python3/dist-packages/iotop/ui.py", line 501, in
run_iotop_window
ui.run()
File "/usr/lib/python3/dist-packages/iotop/ui.py", line 155, in run
self.process_list.duration)
File "/usr/lib/python3/dist-packages/iotop/ui.py", line 434, in
refresh_display
lines = self.get_data()
File "/usr/lib/python3/dist-packages/iotop/ui.py", line 415, in get_data
return list(map(format, processes))
File "/usr/lib/python3/dist-packages/iotop/ui.py", line 388, in format
cmdline = p.get_cmdline()
File "/usr/lib/python3/dist-packages/iotop/data.py", line 292, in
get_cmdline
proc_status = parse_proc_pid_status(self.pid)
File "/usr/lib/python3/dist-packages/iotop/data.py", line 196, in
parse_proc_pid_status
key, value = line.split(':\t', 1)
ValueError: not enough values to unpack (expected 2, got 1)
Any help appreciated.
--
Regards,
Rob McCathie
Re: Still unable to restart networking on Debian 9 text mode only
Hi there On 24/11/2018 18:25, Gary Dale wrote: Reco has already explained why this approach is incorrect. My own two cents on the problem is that Interfaces is meant to define how the network is brought up, not to change a running network. If you want to change a running network, use ifconfig or ip to change the address. e.g. ifconfig enp-s3 10.5.0.3 should work since all you are changing is the ip address. I found this out the 'hard' way; Edit '/etc/network/interfaces', '( ifdown eth0 ; sleep 1 ; ifup eth0 ) &' used to work, even over SSH. But with Stretch this no longer works. With Stretch, I even had to write a script to bring up the WAN (VLAN + PPPoE); http://www.sput.nl/internet/xs4all/config.html This doesn't bother me, but it does show that high level out of the box doesn't necessarily behave the way you might expect. In which case low level plus scripts works just fine. Regards, Rob
Re: DNS Key rollover
Hi there On 04/10/2018 20:32, Reco wrote: Please do not top post. On Thu, Oct 04, 2018 at 02:15:52PM -0400, Default User wrote: Hi, Henning. I am running Unstable, with 4.18.0-2 amd-64 kernel, all updated. I don't know anything about bind. How do I know what bind version I am running, and if I need to do anything regarding the change you mentioned? Stretch's bind has this public part of root's KSK: # grep -A2 20326 /etc/bind/bind.keys # This key (20326) is to be published in the root zone in 2017. # Servers which were already using the old key (19036) should # roll seamlessly to this new one via RFC 5011 rollover. Servers I have an old config with just contains 19036. However, the mkeys file in /var/cache/bind/ contains both. I think this is due to 'dnssec-validation auto' in named.conf. If you have the same - there's nothing to do. If you don't - DNSSEC will stop working for you in seven days. If you do not use BIND - there's nothing to do. Regards, Rob
Re: DNS Key rollover for dnsmasq [SOLVED}
Hi there On 07/10/2018 12:36, Eduardo M KALINOWSKI wrote: On 07-10-2018 07:11, Rick Thomas wrote: On further study, it seems that (in Debian Stretch, at least) the root KSK’s used by dnsmasq are taken from the file /usr/share/dns/root.ds, which is provided by the package dns-root-data; and that package seems to be part of the standard Stretch installation. That file lists both keys (the new “20326” and the old “19036”). So it’s all set to go. No need to panic… (-: Where did you get that information from? I found nothing about dns-root-data in dnsmasq package. It depends on dnsmasq-base, which recommends dns-root-data. Stretch bind9 does not depend on dns-root-data. Backports does. I'd just add a new trust-anchor to the configuration. Just copy and paste from https://github.com/imp/dnsmasq/blob/master/trust-anchors.conf Regards, Rob
Re: cannot open display: localhost:0.0 [solved]
Hi there On 23/09/2018 20:07, Grzesiek Sójka wrote: On 9/23/18 2:36 PM, Rob van der Putten wrote: On 23/09/2018 13:48, Grzegorz Sójka wrote: I need to allow remote applications to connect to xorg. Since I log in using lxdm i have tcp_listen=1 in /etc/lxdm/lxdm.conf. Thus Xorg is running without -nolisten tcp flag. Unfortunately: $ xhost +localhost; DISPLAY=localhost:0.0 xterm localhost being added to access control list xterm: Xt error: Can't open display: localhost:0.0 Any suggestions?? Did you check port 6000? root@pc6:~# netstat -anp | grep 6000 tcp 0 0 0.0.0.0:6000 0.0.0.0:* LISTEN 2684/Xorg tcp6 0 0 :::6000 :::* LISTEN 2684/Xorg Bingo! The command gave me empty output. Always start low level. Now -nolisten tcp is the default Xorg behaviour. Thus, you need to add -listen tcp flag explicitly. That changed some time ago. I added: arg=/usr/bin/X -background vt07 -listen tcp to /etc/lxdm/lxdm.conf and now it works perfectly. Sometimes it's in /etc/X11/xdm/Xservers or /etc/X11/xinit/xserverrc With lightdm it's in /etc/lightdm/lightdm.conf; xserver-allow-tcp=true Regards, Rob
Re: cannot open display: localhost:0.0
Hi there On 23/09/2018 13:48, Grzegorz Sójka wrote: I need to allow remote applications to connect to xorg. Since I log in using lxdm i have tcp_listen=1 in /etc/lxdm/lxdm.conf. Thus Xorg is running without -nolisten tcp flag. Unfortunately: $ xhost +localhost; DISPLAY=localhost:0.0 xterm localhost being added to access control list xterm: Xt error: Can't open display: localhost:0.0 Any suggestions?? Did you check port 6000? root@pc6:~# netstat -anp | grep 6000 tcp 0 0 0.0.0.0:6000 0.0.0.0:* LISTEN 2684/Xorg tcp6 0 0 :::6000:::*LISTEN 2684/Xorg Regards, Rob
Bug#908349: firefox-esr: no sound after upgrading from 52.9 to 60.2
Hi there Note: A reply in bugs failed. On 09/09/18 10:08, Marco Lucidi wrote: > On Sat, 8 Sep 2018 21:41:11 +0200 Samuel Thibault > wrote: > > Unfortunately AIUI upstream has stopped supporting ALSA, so we are stuck > > with pulseaudio for firefox. > > Is there any particular reason for pulseaudio not being listed in the > dependencies? > I mean, even if it's not a "core" dependency, I think it should be > listed in rec or at least in sug. Apparently there alternatives to pulseaudio. See; http://forums.debian.net/viewtopic.php?p=639534#p639534 Does anyone know why ALSA is no longer supported? Does Firefox run in a chroot? How much work would it be to add ALSA support? Regards, Rob
Bind bug
Hi there See; https://kb.isc.org/article/AA-01639/0 I don't think not using deny-answer-aliases is really an option. Regards, Rob
Re: mailing list vs "the futur"
Hi there On 10/08/18 01:03, Rich Kulawiec wrote: No. This is an absolutely terrible idea. Here's why mailing lists are (along with Usenet newsgroups) vastly superior to web-based anything: I prefer Usenet to mailing lists. I read dozens of mailing lists and I don't want all that data on my computer. This is why I use a mail to news gateway. What I really like about Usenet, is the way everything is neatly organized by subject. The same applies to mailing lists when using a gateway. Further more I don't have to create an account for each group. I just subscribe. Regards, Rob
Linphone dependencies
Hi there Package: linphone depends on linphone-nogtk. But I don't see why. It contains no libs, just the binaries linphonec and linphonecsh. Which, as far as I can tell, are never called from linphone. In fact, I can move them to an other dir and still run linphone. What did I miss? Regards, Rob
Re: stretch and DNS name resolution service for other devices on a LAN
Hi On 19/01/18 15:08, Andy Hawkins wrote: I guess I don't *need* IPv6, but as a technology geek it's just something I wanted to play around with. I've set up a tunnel with Hurricane Electric that gets me a fully routable IPv6 prefix that I can assign an address from to each of the devices on my LAN. If you have got your own IPv6 /48 combined with RFC1918 IPv4 then split horizon DNS is probably a good idea; https://en.wikipedia.org/wiki/Split-horizon_DNS I do use a 192.168/16 for my local IPv4 addressing, but given that moving forward IPv6 is likely to become more and more available, I thought it worth having a play with. I can now (if I want) allow IPv6 access to every device on my internal network without having to worry about NAT and port forwarding to get connections through my firewall. Thanks for the pointer to tinydns, I'll take a look. AFAIK Bind is the only named with a good split horizon implementation. Regards, Rob
Re: Kernel problem?
Thanks to everyone who provided help with this failure of the 4.9.0-5-686-pae kernel. I have tried both "nopti" and "pti=off" kernel parameters to see if it is the fix for "Meltdown" which is causing the problem, but neither parameter changes anything. Since the system doesn't even get to the kernel fsck stage, I don't have any logs to be able to either report a bug or analyse the problem. I also realise now that auto-update is turned on by default in Stretch, so that is how the new kernel became installed. I have temporarily turned auto-update off because I don't want any automatic update to clobber my only working kernel - the 4.9.0-4-686-pae. The updating system only seems to keep 2 generations of kernel? Any further ideas would be very welcome. Cheers, Rob Hurle - Rob Hurle e-mail:rob1...@gmail.com Mobile: +61 417 293 603 (Australia) Telephone: (02) 6236 3895 28 Mirrormere Rd, Burra, NSW 2620, Australia On 8 January 2018 at 13:11, tv.deb...@googlemail.com < tv.deb...@googlemail.com> wrote: > On 07/01/2018 21:27, Marc Auslander wrote: > >> The new kernel implements the "fix" for meltdown. You could try booting >> with the fix turned off - I believe the kernel parameter is pti=off >> Rob Hurle writes: >> >> Hi All, >>> >>> I'm running Stretch and yesterday I did my normal: >>> >>> sudo apt-get update >>> sudo apt-get upgrade >>> >>> It seemed to install vmlinuz-4.9.0-5-686-pae (and associated config >>> and image files, etc) in place of 4.9.0-4-686-pae versions. Now the >>> system won't boot at all. I have reverted to 4.9.0-4-686-pae and all >>> is well. My questions are: >>> >>> 1. Does anyone else see this? >>> >>> 2. How can I revert without losing my working 4.9.0-4-686-pae system? >>> Can I just change the soft links for initrd.img and vmlinuz at / to >>> point to the 4.9.0-4-686-pae versions instead of the 4.9.0-5-686-pae >>> ones? Will this break something else for a future upgrade? >>> >>> Any help much appreciated. Thank you. >>> >>> Cheers, Rob Hurle >>> >>> - >>> Rob Hurle >>> e-mail: rob1...@gmail.com >>> Mobile: +61 417 293 603 (Australia) >>> Telephone: (02) 6236 3895 >>> 28 Mirrormere Rd, Burra, NSW 2620, Australia >>> >> >> > Hi, if the hang is due to memory isolation option, then "nopti" added as a > kernel boot parameter will cancel it. If you are on an Intel machine this > will leave you exposed to the new class of "Meltdown" attacks. > This kind of boot problem seems to happen to a very small number of > systems, patches are already queued in the kernel to (hopefully) correct > this, but you will have to wait for them to be merged. > > Hope it helps. > >
Kernel problem?
Hi All, I'm running Stretch and yesterday I did my normal: sudo apt-get update sudo apt-get upgrade It seemed to install vmlinuz-4.9.0-5-686-pae (and associated config and image files, etc) in place of 4.9.0-4-686-pae versions. Now the system won't boot at all. I have reverted to 4.9.0-4-686-pae and all is well. My questions are: 1. Does anyone else see this? 2. How can I revert without losing my working 4.9.0-4-686-pae system? Can I just change the soft links for initrd.img and vmlinuz at / to point to the 4.9.0-4-686-pae versions instead of the 4.9.0-5-686-pae ones? Will this break something else for a future upgrade? Any help much appreciated. Thank you. Cheers, Rob Hurle ----- Rob Hurle e-mail:rob1...@gmail.com Mobile: +61 417 293 603 (Australia) Telephone: (02) 6236 3895 28 Mirrormere Rd, Burra, NSW 2620, Australia
Re: Lightweight Audio-Only DLNA player
On 17/10/17 23:37, Robert Latest wrote: Hello, there's plenty of DLNA capable media players around, but all of them (or at least, all that I could find) are gargantuan packages with tons of functionality and dependencies that I don't need or want. This is supposed to run audio-only on a small headless SBC with a USB connected DAC, nothing else. Thanks, robert minidlna is not suitable? -- Regards, Rob McCathie
Re: Asterisk security update
Hi there On 04/10/17 20:10, Don Armstrong wrote: On Wed, 04 Oct 2017, Rob van der Putten wrote: The Asterisk security update seems to exclude amd64; I can't find a 13.14.1~dfsg-2+deb9u2_amd64.deb in pool. It's in stable-new: % rmadison asterisk|grep 9u2 asterisk | 1:13.14.1~dfsg-2+deb9u2 | stable-new | source, arm64, armel, armhf, i386, mips, mips64el, mipsel, ppc64el, s390x and it will (likely) get accepted into proposed-updates for stable as soon as the amd64 build is present: https://release.debian.org/proposed-updates/stable.html https://release.debian.org/proposed-updates/stable_missing.html I'm not sure why the amd64 buildd hasn't yet built that package, though. Apparently it's a dak bug. Regards, Rob
Re: Asterisk security update
Hi there On 04/10/17 17:33, Roberto C. Sánchez wrote: On Wed, Oct 04, 2017 at 05:24:21PM +0200, Rob van der Putten wrote: The Asterisk security update seems to exclude amd64; I can't find a 13.14.1~dfsg-2+deb9u2_amd64.deb in pool. That is quite strange. There have been reports of uploads disappearing from the ftp-master host some days ago as a result of an outage: https://lists.debian.org/debian-devel/2017/09/msg00385.html https://lists.debian.org/debian-devel/2017/10/msg6.html Those threads may shed some light on the matter. According to https://packages.debian.org/stretch/asterisk there is no 13.14.1~dfsg-2+deb9u2_amd64.deb. Which is a bit confusing when one expects an update. Regards, Rob
Asterisk security update
Hi there The Asterisk security update seems to exclude amd64; I can't find a 13.14.1~dfsg-2+deb9u2_amd64.deb in pool. Regards, Rob
Re: DVD won't eject after playing DVD
Hi there On 20/08/17 17:59, Rob van der Putten wrote: An other bit stretch 'weirdness': After playing a DVD with VLC, the eject button is dysfunctional and I have to type 'eject' on the command line to get it to eject. Found an other one; File type icons are missing in a file listing. ^L fixes this. How do I fix this? Regards, Rob
Re: Debian v9 it's a stretch
Hi there On 24/08/17 15:39, Dan Ritter wrote: As of Stretch, the standard OpenSSH sshd does not support Protocol 1, so there's no particular reason to enforce it by stating Protocol 2. I assumed as much. It's just a simple way to keep rkhunter happy. PermitRootLogin now defaults to "prohibit-password", which means that you can log in as root with a proper SSH key or via other methods you have set up, but not with a password. Another useful argument is forced-commands-only, which requires both public-key authentication and a command="blah blah" option in the authorized_keys file, and only allows those commands to be run. If you've got a pull backup system, that can help. The alternative would be to reconfigure rkhunter. Regards, Rob
Re: Debian v9 it's a stretch
Hi there On 22/08/17 21:16, Rob van der Putten wrote: Upgrade from amd64 Jessie (insserv, bare ALSA). I kind of miss xfce-mixer Alsamixergui works, but xfce-mixer looked better. I use qasmixer now; http://www.sput.nl/software/qasmixer.png I removed the xfce4 meta package, since it insists on pulse related stuff. There is no xfce4-artwork in Stretch, but the Jessie version works. I'm very happy that Firefox works on bare ALSA. I posted a few things, probably kernel bugs. Apart from that, seamless transition. Even my serial mouse still works. Rkhunter nags a bit about SSH. Even though things seem to be OK. I still have to look into that. I added; Protocol 2 PermitRootLogin no to /etc/ssh/sshd_config Regards, Rob
Re: NFS creates hidden port
Hi there On 22/08/17 18:01, Thomas Schmitt wrote: Question is whether it can be unambiguously recognized in netstat output as long as it is visible. Further: Is it always only one hidden port ? It's always a callback from a Stretch NFS server to a Jessie NFS client. It occurs when the client mounts. The port moves from established, to wait to hidden. When the Jessie box mounts a dir exported by Stretch, the Stetch box does a callback. When the Stretch box mounts a dir exported by the Jessie box, the Jessie box doesn't callback. As a temporary fix a disabled the callback by mounting with NFS version three (vers=3) on the Jessie box. Regards, Rob
Re: Debian v9 it's a stretch
Hi there On 21/08/17 18:29, deloptes wrote: No issues, even with new installation on new disk :) Upgrade from amd64 Jessie (insserv, bare ALSA). I kind of miss xfce-mixer Alsamixergui works, but xfce-mixer looked better. I'm very happy that Firefox works on bare ALSA. As Greg wrote, please share the issues. I posted a few things, probably kernel bugs. Apart from that, seamless transition. Even my serial mouse still works. Rkhunter nags a bit about SSH. Even though things seem to be OK. I still have to look into that. Going back - might be an option perhaps for the next 2-4y. changing dists - you may face same on other dist as well Regards, Rob
Re: NFS creates hidden port
Hi there On 22/08/17 15:23, Thomas Schmitt wrote: It seems that it was fixed or suppressed intermediately. The newer post says "It's back!". I already stated my enthusiasm on occasion of your post about DVD ejecting. It is discouraging to get ignored after having invested substantial effort in diagnosing or at least reliably reproducing a kernel problem. Well, complaining is futile. Try to work around in user space. E.g. try to patch unhide-tcp so that it reads the NFS port number from a file which you create before the Rkhunter run. You could let function checkoneport() return "ok" if "port" is the registered NFS zombie. This would be done before the function runs netstat by if (NULL != (fich_tmp=popen (command, "r"))) in https://sources.debian.net/src/unhide/20130526-1/unhide-tcp.c/#L190 I would have to find out when NFS does a callback an then dump the local port into a file. Regards, Rob
Re: NFS creates hidden port
Hi there On 22/08/17 12:38, Thomas Schmitt wrote: Rob van der Putten wrote: I think this may be a kernel bug. A valid theory for now. I googled on: https://askubuntu.com/questions/851986/rkhunter-reports-hidden-tcp-port-probably-nfs-server brings me to http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg910866.html Some suspicious kernel commit ids are mentioned in http://www.mail-archive.com/linux-kernel@vger.kernel.org/msg911346.html It looks like the original poster ended up speaking to himself. Insightful but lonely. One year late the problem appeared again https://patchwork.kernel.org/patch/9207481/ And this post is over a year old. One would expect this to be fixed by now. Regards, Rob
Re: NFS creates hidden port
Hi there On 22/08/17 11:44, to...@tuxteam.de wrote: This raises the question why netstat does not show Rob's NFS ports. Does NFS change the port fast enough so that netstat and port scan differ ? A good question. I guess we need more details from the OP. The hidden port lingers on for days. Until one restarts NFS. NFS then uses an other port which clearly shows in netstat, until it becomes hidden again. And the daily rkhunter [1] starts complaining about it. [1] https://en.wikipedia.org/wiki/Rkhunter I think this may be a kernel bug. Regards, Rob
NFS creates hidden port
Hi there More stretch weirdness: Rkhunter alerts me to a hidden port. Restarting NFS changes the port number. Today I did a netstat after restarting NFS and then run unhide-tcp a few times: It's the client side of RPC NFS callback. What can I do about this? Regards, Rob
DVD won't eject after playing DVD
Hi there An other bit stretch 'weirdness': After playing a DVD with VLC, the eject button is dysfunctional and I have to type 'eject' on the command line to get it to eject. How do I fix this? Regards, Rob
Re: Stretch vim doesnt cut and paste
Hi there On 19/08/17 14:10, Brian wrote: "set mouse=" in ~/.vimrc. In /etc/vim/vimrc it doesn't work. In ~/.vimrc it does. Thanks! Regards, Rob
Stretch vim doesnt cut and paste
Hi there After a upgrade from jessie to stretch I can't use the mouse to cut and paste in vim. nvi works, nano works. As quick fix I removed xxd and installed the jessie version of vim, vim-common and vim-runtime. This does work. Regards, Rob
Re: Replace systemd
Hi there On 05/07/17 17:27, Don Armstrong wrote: It already exists: https://lists.debian.org/debian-user/2017/04/msg00097.html « You can just append: preseed/late_command="in-target apt-get install -y sysvinit-core" to the installer command line. Or you can roll your own install media with its own syslinux.cfg which adds that or something more complicated in a preseed file. You don't need to fork the installer, or submit any patches upstream. If you want something more complicated, like not installing systemd at all, you'll have to pass --include and --exclude options to debootstrap using the base-installer/includes and base-installer/excludes preseed options; something like: base-installer/includes=sysvinit-core base-installer/excludes=systemd-sysv but that's totally untested. » You need to modify your net install DVD too; http://without-systemd.org/wiki/index.php/How_to_remove_systemd_from_the_Netinst_CD Regards, Rob
Re: Apology to siduction 17.01 (was Re: why??why?why??)
Hi there On 13/03/17 15:54, GiaThnYgeia wrote: Out of my frustration and lack of understanding, or the belief that all systems should run as trouble-free as clean-debian, and possibly due to just getting tired of fighting something too long, I bad-mouthed siduction in public. My source of frustration came from fighting an installation in a tired old system (64bit) and a mediocre not so smart monitor. Most installers and systems I've seen use 800x600 as their boot/grub screen. Siduction probably doesn't have access to an old monitor. What this meant was the screen would blank out and have an internal error showing not-compatible meanwhile the grub time would expire and boot up eventually to a graphical login screen. So I thought some weird boot system prevents me from seeing what's going on or giving me options, so in effect I lost access to all other installed systems. With a clean head and a different pc I went in as root and looked on the usb what the grub config looked like and simply replaced the high resolution to an 800x600 and delayed the default boot sequence just in case. Siduction ... the best grub-configuration I have seen yet, once you can get to it and edit it :) Sorry Siduction! kAt PS But you (siduction) should lower your graphics expectations down a notch or two. There are some 1024 monitors still around. My server has a 12 Inch 640x480 monitor (IBM 8513). Installers shouldn't make assumptions about resolutions. PS2 Maybe a way to scale back and confirm visibility of the grub menu before it boots the default may help, if you MUST high a high definition grub screen. PS3 Being more confident I can back up and restore my experimentation with siduction really did help as well ;) dd/xorriso and all! PS4 More than a week ago when I subscribed to their forum "I asked" on why would their installer not run on very similar systems. Never thought the monitor was IT. I got no response from them ... but I assume small teams of developers can't address everyone's complaints. Regards, Rob
Re: `Filter failed' no print error message
Hi there On 13/03/17 17:27, Rodolfo Medina wrote: Brian writes: On Sat 11 Mar 2017 at 21:06:23 +, Rodolfo Medina wrote: Brian writes: On Tue 07 Mar 2017 at 15:41:54 +, Rodolfo Medina wrote: Thanks, `cups' alone was enough. Now the printer seems to be configured. But it does not print, and, when I get into `jobs', I see they are stopped with "Filter failed" claim. What can I do? Please help. Thanks. Read the wiki and find out how to activate debug logging. Produce an error_log and examine it for the filters which are run and which one failed. Tell us about all the filters. Also, post your printer make and model and the output of 'lpstat -t'. I noticed that CUPS does not really provide the PPD file I'm adding. In fact, there in the managing section, `Modify Printer', where it says `Or Provide a PPD File', I select the file: ML-191xspl2.ppd, which is in my home directory, then click on `Modify Printer' and it seems all right; but then, when I get into it again, the file is no more there, and there is `No file selected' instead. Maybe this is related to my `Filter failed' problem...? There are seven instructions in the advice given. You haven't provided a response to any of them. If there is a problem with following what is required, please ask. As the wiki suggests, I did: # cupsctl --debug-logging . Then I did: $ lpstat -t scheduler is running system default destination: Samsung_ML-191x_252x_Series device for Samsung_ML-191x_252x_Series: usb://Samsung/ML-191x%20252x%20Series?serial=Z2L9BABZ700571F. Samsung_ML-191x_252x_Series accepting requests since Sun 12 Mar 2017 07:44:58 GMT printer Samsung_ML-191x_252x_Series is idle. enabled since Sun 12 Mar 2017 07:44:58 GMT Sending data to printer. Samsung_ML-191x_252x_Series-16 rodolfo 457728 Sat 11 Mar 2017 19:56:36 GMT Samsung_ML-191x_252x_Series-17 unknown 1024 Sat 11 Mar 2017 20:09:12 GMT Samsung_ML-191x_252x_Series-18 unknown 457728 Sat 11 Mar 2017 20:20:18 GMT Samsung_ML-191x_252x_Series-19 rodolfo 457728 Sun 12 Mar 2017 07:44:49 GMT The file /var/log/cups/error_log is very long. I noticed in it the following lines: D [07/Mar/2017:14:35:23 +] [Job 1] Started filter /usr/lib/cups/filter/pdftopdf (PID 4741) D [07/Mar/2017:14:35:23 +] [Job 1] Started filter /usr/lib/cups/filter/pdftops (PID 4742) D [07/Mar/2017:14:35:23 +] [Job 1] Started filter /usr/lib/cups/filter/rastertosamsungspl (PID 4743) D [07/Mar/2017:14:35:23 +] [Job 1] Started backend /usr/lib/cups/backend/usb (PID 4744) D [07/Mar/2017:14:35:23 +] [Job 1] pdftops - copying to temp print file \"/var/spool/cups/tmp/0128658cb23ec\" D [07/Mar/2017:14:35:23 +] [Job 1] execv failed: No such file or directory Someone is calling a program which isn't there. Or the path is wrong. Does pdftops exist? D [07/Mar/2017:14:35:23 +] [Job 1] PID 4743 (/usr/lib/cups/filter/rastertosamsungspl) stopped with status 102 (No such file or directory) D [07/Mar/2017:14:35:23 +] [Job 1] Hint: Try setting the LogLevel to "debug" to find out more. Don't know what to do now, please help... While installing Jessie I noticed that I had to manually edit some file to get things to work. I think it was printers.conf, but I'm not sure. Regards, Rob
Re: Do have programs have poor documentation? (was ... Re: Why? -- "A Modest Proposal")
Hi there On 30/12/16 18:25, deloptes wrote: Lisi Reisz wrote: Try reading what I have actually said, whilst making some attempt to understand it, instead of just contradicting it. You do enjoy contradicting people, don't you? Hah, Lisi I just got the same impression from Xen. I am glad to hear that the education system in Holland is better than here. I am not surprised and do not doubt it. If nothing changes the education systems in all FED subordered countries will be equally bad in not more than 20-30y. We'll have 2 class society as in the middle ages. Xen, I knew older dutch people who were speaking 4-5 languages ... nowdays I heard languages are not more subject in school. Why learn German anyway, when everybody speaks English? Isn't it that way now in the schools in NL? And this is just in 10-15y period. Not to discuss about the TV channel quality in NL. Compared to UK it was indeed better (irony). I don't know about US, but here in EU all is getting averaged to the lowest mark possible. When I was studying dutch some 20y ago I watched dutch TV and it was still good, but now it is rubbish, and the same all over Europe. A lot of people stopped watching TV - I don't for 10y already. ARTE [1] and TV5 [2] are still interesting sometimes. The BBC used to be brilliant, but dumbed down considerably over last few decades. TV5 comes with subtitles in various languages. ARTE is dual language French - German. [1]: https://en.wikipedia.org/wiki/Arte [2]: https://en.wikipedia.org/wiki/TV5Monde Regards, Rob
Re: Do have programs have poor documentation? (was ... Re: Why? -- "A Modest Proposal")
Hi there On 29/12/16 14:44, cbannis...@slingshot.co.nz wrote: [sorry for the late response.] On Fri, Nov 18, 2016 at 11:05:48AM +0100, Thomas Schmitt wrote: It is not easy to describe a program with many use cases and even more particular settings and actions. What lacks to my experience as confused reader and as best effort writer is the user's view on the programs. man pages should document the details and often do sufficiently. But the user looks for solutions, not opportunities. An interesting thing to ponder is whether a tractor manual should explain how to prepare a field to plant carrots. No, but you might want to put a carrot man page in section 7 (I like section 7). Also, think of man pages as a reference (what does that switch do again?) not as a "first introduction" or tutorial. e.g. I'd be annoyed if a chess database program's documentation consisted of how to play the game, rules of the game, etc. Regards, Rob
Re: Whats the alternative to .xinitrc
Greg Wooledge wrote: > Debian's Xsession is not the same as Red Hat's. > > If you need a home directory that will work consistently across different > operating systems, then I *strongly* suggest you stick with startx and > make .xinitrc a symlink to .xsession for backwards compatibility. Any > attempt to reverse engineer and consolidate different display manager + > Xsession combos across multiple operating systems is likely to lead to > mental breakdown. > I'm a committed Debian user these days so no need for any backwards compatibility but Red Hat was the first distro I ran (must be 20 years ago now) but I can remember mapping out the startx processes but not the details. Thank you for Xsession WIKI, I wish I had found that a few days ago, it would explained all I needed to know. Rob
Re: Whats the alternative to .xinitrc
to...@tuxteam.de wrote: > > On Mon, Feb 20, 2017 at 01:32:37PM +0100, Vincent Lefevre wrote: >> On 2017-02-20 12:49:34 +0100, to...@tuxteam.de wrote: > > [...blah...] > >> > See man (5) Xsession and poke a bit through the scripts in >> > /etc/X11/Xsession.d/, then have a look at /etc/X11/Xsession.options, >> > then at /etc/X11/Xsession. >> >> I don't see any contradiction to what I've said. Note that I haven't >> mentioned .xsessionrc at all, only .xinitrc, which is a different >> thing. > Thank you for your explanations, I must admit that I didn't expect it to be so involved. I seem to remember mapping out the X startup processes a number of years ago but I was running RedHat then, so I suppose it has changed a lot in the meantime. Regards Rob
Re: Re: Whats the alternative to .xinitrc
>> I am running up to date KDE Sid with SDDM but I need to start the gnome >> keyring at x startup to satisfy Midori dependencies. >> >> The Midori FAQ suggests adding “export `gnome-keyring-daemon –start`” to >> the historical .xinitrc but this seems to be ignored as does .xsession as >> suggested by the startx man page. >> >> As .xinitrc has worked OK in the past, how do I now set the user >environment >> when X starts? > >My .xsessionrc is functional. The magic happens in > /etc/X11/Xsession.d/40x11-common_xsessionrc > > +--- > | # This file is sourced by Xsession(5), not executed. > | > | #Source user defined xsessionrc (locales and other environment variables) > | if [ -r "$USERXSESSIONRC" ]; then > | . "$USERXSESSIONRC" > | fi > +--- > >Attention: I'm running SysV init -- perhaps things are different in other >settings. Thank you for your help, I was wrong footed by the man page which suggested using .xsession not .xsessionrc so I guess the man page needs an update. Looks like it is now OK after changing to .xsessionrc. Also sorry for the double posting, I has DNS problems which caused me to be unsubscribed and I didn't see the replies. Rob
Whats the alternative to .xinitrc
I am running up to date KDE Sid with SDDM but I need to start the gnome keyring at x startup to satisfy Midori dependencies. The Midori FAQ suggests adding “export `gnome-keyring-daemon –start`” to the historical .xinitrc but this seems to be ignored as does .xsession as suggested by the startx man page. As .xinitrc has worked OK in the past, how do I now set the user environment when X starts? Rob
Re: potential damage to Debian "stable" when installing packages from "testing"
Hi there On 08/12/16 16:27, Alex Mestiashvili wrote: 0) backport it yourself. It is not that hard to dget a dsc file from testing and try to build it for the current release. Often works without additional efforts. That's what I do. I'm rather blunt about it; 1. Does it compile? 2. Does it install? 3. Does it work? 4. Am I still happy about it after a while? Recently I ran into some squid and libcap3 problems; http://www.sput.nl/software/squid/squid-backport.html According to a friend it's OK to replace the maintainer supplied symbols file with the generated one. And the libcap3 does indeed compile and build a package. I'm considering building Squid with it and want to name the new package 3.5.22-1~bpo8+2 instead of 3.5.22-1~bpo8+1. Is this possible and how do I do this? Regards, Rob
Re: Debian *not very good
Hi there On 26/11/16 18:38, Rob van der Putten wrote: I run XFCE on my desktop. I had to add myself to sudo to make things work properly. And admin. admin is needed to get xconsole syslog to work. sudo to keep xdm logout from complaining. I use lightdm now though. I edited the cups config manually. The web interface breaks stuff. Regards, Rob
Re: Debian *not very good
Hi there On 25/11/16 22:26, Latincom wrote: Is there a step by step guide or How to on line? I have 1 Wheezy without Systemd, and i would like to upgrade it. Thanks. You can do both an upgrade and an install from scratch without systemd; http://without-systemd.org/ http://without-systemd.org/wiki/index.php/How_to_remove_systemd_from_the_Netinst_CD http://without-systemd.org/wiki/index.php/How_to_remove_systemd_from_a_Debian_jessie/sid_installation http://people.skolelinux.org/pere/blog/How_to_stay_with_sysvinit_in_Debian_Jessie.html I had to make some other changes to get things to work though. Reading the release notes and keeping a close eye on the install process helps. I run XFCE on my desktop. I had to add myself to sudo to make things work properly. Regards, Rob
Re: IPv6 Problems
Hi there On 06/10/16 07:29, Ethan Rosenberg wrote: iface etho0 inet static See post by Georgi Naplatanov. address 2620:7:a000::1 netmask 64 gateway :::c0a8:101 :::c0a8:101 is 192.168.1.1; Kernels use ::/96 addresses internally and not on networks. So I don't think this is going to work. 2000::/3 addresses work. And probably fe80::/11 too. Regards, Rob
Re: Is nagle disabled?
Hi there On 10/09/16 13:37, Nicolas George wrote: Le quintidi 25 fructidor, an CCXXIV, Rob van der Putten a écrit : So the question should have been 'Is delayed ack disabled'. I have a hard finding decent information on the subject, so I did a bit of experimentation; I send tiny bits of data (two bytes at a time) to a little echo server. As it turns out, the use of delayed acks and therefore nagle, depends on the time between sending the bits of data and the RTT. I do not think you can really test delayed ACK with an echo server: the principle of delayed ACK is to leave a little time to the application to send a reply, because then the ACK can be bundled with the reply, saving a packet. With an echo server, the reply comes immediately, never letting the delay for the ACK expire. I use netcat6. From man nc; nc6 --continuous --exec cat -l -p So the data goes from nc to cat and then back to nc. Timestamps (epoch.microseconds) are dumped to a file on transmission and to an other file on reception of the 'echo'. Further check with tcpdump. Experiment on LAN and experiment over DSL; When the time between the bits of data is larger then the RTT, data is send at two bytes per TCP packet. Regards, Rob
Re: Is nagle disabled?
Hi there On 09/09/16 19:57, Rob van der Putten wrote: I thought I overlooked something. And this is it. Thanks! So the question should have been 'Is delayed ack disabled'. I have a hard finding decent information on the subject, so I did a bit of experimentation; I send tiny bits of data (two bytes at a time) to a little echo server. As it turns out, the use of delayed acks and therefore nagle, depends on the time between sending the bits of data and the RTT. Regards, Rob
Re: Is nagle disabled?
Hi there On 09/09/16 18:19, Stefan Monnier wrote: I suggest you re-read https://en.wikipedia.org/wiki/Nagle's_algorithm Nagle only kicks in when there are un-acknowledged packets. So on the first packet, there is no delay. There will be a delay on the *second* packet if it's small and we haven't yet received confirmation that the first packet was received. I thought I overlooked something. And this is it. Thanks! Regards, Rob
Is nagle disabled?
Hi there I'm experimenting with TCP to see how long it takes to send a small amount of data from A to B. One would expect a latency of a few hundred milliseconds, but it's a few hundred microseconds instead. It is as if Nagle's algorithm has been disabled. Regards, Rob
Problem with linphone and twinkle
Hi there Linphone doesn't sort contacts in alphabetical order. Twinkle crashes when I switch virtual desktop during a call. Any ideas? Suggestions for other softphones? Regards, Rob
Re: Squid security
Hi there Rob van der Putten wrote: The libs are different. So I build a backport. And a libecap3 backport. It wants libecap3. Regards, Rob
Re: Squid security
Hi there Rob van der Putten wrote: heqami...@runbox.com wrote: No need to build a backport. Just use the sid/unstable testing version on apt-get I didn't check the library compatibility. Thanks. The libs are different. It's a bit odd though. Every version patched except stable. Regards, Rob
Re: Squid security
Hi there heqami...@runbox.com wrote: No need to build a backport. Just use the sid/unstable testing version on apt-get I didn't check the library compatibility. Thanks. It's a bit odd though. Every version patched except stable. Regards, Rob
Squid security
Hi there This puzzles me a bit: Information on source package squid3; https://security-tracker.debian.org/tracker/source-package/squid3 Do I need a to build a backport to be safe? Regards, Rob
Re: [partially solved]Re: Iceweasel uses wrong charset
Hi there Michael Lange wrote: On Wed, 11 May 2016 18:43:50 +0200 Michael Lange wrote: According to this (yet probably dated ) page: http://www-archive.mozilla.org/projects/intl/chardet.html universal_charset_detector might help if it were enabled by default. Just looked at the above page again, and noticed that the "japanese" charset detector also supports UTF-8 and cp1252. So I tried to set intl.charset.detector to "ja_parallel_state_machine" and actually at least the minimal utf-8 text file I just tested (which contains no more than a bunch of german umlaut characters) is now displayed correctly. Yesterday I found https://bugzilla.mozilla.org/show_bug.cgi?id=910192 But gmane dropped my post. Regards, Rob
Re: Iceweasel uses wrong charset
Hi there to...@tuxteam.de wrote: On Wed, May 11, 2016 at 12:34:29PM +0200, Rob van der Putten wrote: I actually thought about this. But it just seems crude; vim some.txt ^vufeff :wq It does work though. But I want UTF-8 without a BOM. It is, after all, the default locale. Everyone wants that... except Microsoft, it seems. Is Firefox really this braindead? Sometimes I fear that too. Note: If you edit this again, vim doesn't show the BOM any more. It's there though. That's because it has been enshrined by the Unicode Consortium. I think I know who pushed that (Apple might have helped, though). So Vim's correct in tolerating that on input. I don't think they like BOM. After all they named it 'ZERO WIDTH NO-BREAK SPACE'. Perhaps you can help yourself by even uglier tricks (like with an http-equiv header in your HTML). But best would be if you could exorcise this abject behaviour from Firefox. For HTML this works. I can even put hieroglyphs in a plain text file and print it using a plain text print queue. UTF-8 just works. Firefox however, insists on putting a square peg into a round hole. Let us know... The only way to fix this is probably to edit the source. I have a very old iceape that actually gets it right. An older version had a 'remove all bookmarks' button, which no longer exists. With iceweazel one actually has to create a bookmark to switch javascript on and of (the plug-in for this does not work). And with jesie's apache there is no longer a default mime-type or cern-meta module because _NO_ _ONE_ uses this any more. Apparently I'm a figment of my own imagination. Regards, rob
Re: Iceweasel uses wrong charset
Hi there to...@tuxteam.de wrote: This is happening only on local files? Perhaps FF is expecting that brain damaged byte order mark to recognize the file as UTF-8? When the data come over HTTP there are other out-of-band hints to communicate the encoding. Try editing the file with some Microsoft cra^H^H^H text editor and saving as "Unicode": that might insert the BOM at the beginning. Just a shot in the dark. I actually thought about this. But it just seems crude; vim some.txt ^vufeff :wq It does work though. But I want UTF-8 without a BOM. It is, after all, the default locale. Is Firefox really this braindead? Note: If you edit this again, vim doesn't show the BOM any more. It's there though. Regards, Rob
Re: Iceweasel uses wrong charset
Hi there Michael Lange wrote: you can try the following: type "about:config" into iceweasel's address bar (without the quotes of course), then type "charset" into the config's search bar. Several options should appear. Look for the key "intl.charset.default". Is the value already set to "UTF-8"? If not, double-click the current value and change it to UTF-8 , this should hopefully fix it. This seems the obvious thing to do. It doens't work though. Regards, Rob
Iceweasel uses wrong charset
Hi there UTF-8 is my default locale. So obviously text files are UTF-8 as well. So why does iceweasel insist that these are windows-1252? How do I fix this? Regards, Rob
Analog, domain not given and IPv6
Hi there I was under the impression that Jessie's Analog would support IPv6. It turns out it doesn't. Older thread on the subject; https://lists.debian.org/debian-user/2012/12/msg00784.html So I applied the same hack as before; https://lists.debian.org/debian-user/2012/12/msg00932.html Regards, Rob
Re: Problems after upgrade to Jessie
Hi there Rob van der Putten wrote: How do I get cern_meta.load? Is compiling Apache myself the only option? And default mime type is no longer supported. Which is really annoying if you want to read compressed text files with a web browser. Regards, Rob
Problems after upgrade to Jessie
Hi there I run into the following problems after upgrading to Jessie; - logresolve segfaults. - Apache misses cern_meta.load - console-kit-daemon complains; console-kit-daemon[4676]: GLib-CRITICAL: Source ID 154 was not found when attempting to remove it How do I get cern_meta.load? Is compiling Apache myself the only option? Regards, Rob
Re: Sound card question
Hi there Danny wrote: No jumpers on the motherboard ... :( ... Sometimes it's the way the connector is plugged into the motherboard or the front panel. Sometimes it's a BIOS setting. Look for AC 97. See; https://en.wikipedia.org/wiki/Intel_High_Definition_Audio If all of this fails you need to hack the hardware a bit. See; Regards, Rob -- ISDS is evil. Abolish ISDS.
Re: Sound card question
Hi there Seeker wrote: If you do actually have an audio out and a line out and the line out doesn't produce audio when something is plugged into the audio out, it may be an indication that it's a hardwired mechanical function built into the audio out jack to break the circuit to the line out jack when anything is plugged into audio out. Systems with headphone jack at the front and a line out at the rear, will often switch of the the line out off when a headphone is plugged in. This feature can usually be switched of by means of a jumper on the motherboard. Consult your motherboard documentation. Regards, Rob
mate and ibus
Anyone using ibus on mate desktop? I've just moved to jessie and prefer the mate desktop to gnome. However ibus won't change to another input method (unikey for Vietnamese in my case). mate opens up ibus-unikey ready for Vietnamese input and will not recognise the key sequence to change to English ( blank in the default case, but nothing else works when any other combination is chosen). ibus-unikey works correctly for gnome. Anyone else had this problem? Cheers, Rob Hurle ----- Rob Hurle e-mail:rob1...@gmail.com Mobile: 0417 293 603 (Australia) 0948 243 538 (Vietnam) Telephone: (02) 6236 3895 28 Mirrormere Rd, Burra, NSW 2620, Australia
Re: Unwanted application autostarting at login with XFCE
Hi there Stephen Powell wrote: Reposting, since the original post seems to have gotten lost in the mail. I've got an annoying problem. Every time I login to the XFCE desktop, an application starts that I don't want. (It happens to be abiword.) I close the application window and go on, but it starts up again the next time I login. I've checked Check if it's running in de background. Applications -> Settings -> Session and Startup in the "Application Autostart" and "Session" tabs, but abiword is not listed there. I've searched the Internet, but did not find anything. Any ideas? Regards, Rob -- ISDS is evil. Abolish ISDS. -- 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/mp24kd$fd6$1...@ger.gmane.org
Re: Flash update
Hi there Curt wrote: https://addons.mozilla.org/en-US/firefox/blocked/p948 A rather fuzzy statement. Regards, Rob -- ISDS is evil. Abolish ISDS. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/mo50lp$pk9$1...@ger.gmane.org
Flash update
Hi there Apparently .481 is buggy as well. Just to see what would happen, I renamed libflashplayer.so; Youtube complains about a missing plugin, but works anyway. It seems that FF 31.8.0 has enough HTML 5 support to make this work. Regards, Rob -- ISDS is evil. Abolish ISDS. -- To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/mo3fag$fu0$1...@ger.gmane.org
Gnome 3 VNC server issue
When try to enable my vnc server on Debian Jessie: I have to unmanage my network interfaces from /etc/network/interfaces The only way to start remote desktop on my internal interface is to let the gnome network manager control my interface. This is unwanted behavior as i use my system als server and want to manage my interfaces by config file rather then using the graphical network manager. - start vnc server without network interfaces - show the interfaces managed by the /etc/network/interfaces -- 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/c29ba8fb-20d0-4720-b5e0-3b45404a0...@gmail.com
Re: Flash update
Hi there Sven Arvidsson wrote: Or consider making do without it. :) Any alternatives? I hate flash! Flash seems to be in a category of its own when it comes to security problems. Indeed. Regards, Rob -- 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/mme4gj$2rl$1...@ger.gmane.org
Flash update
Hi there For i386 the new version is; http://fpdownload.macromedia.com/get/flashplayer/pdc/11.2.202.468/install_flash_player_11_linux.i386.tar.gz Regards, Rob -- 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/mmdup9$58a$1...@ger.gmane.org
Re: Nova Desktop
On 18/06/15 10:43, Lisi Reisz wrote: On Thursday 18 June 2015 00:04:12 Jose Martinez wrote: Anyone know anything about the Nova Desktop application. I have it installed and set it up, but it doesn't seem to affect my desktop background. I have several .jpg images that I had wanted to cycle through the desktop background, and it seemed that Nova was just the ticket I have found references to Android and references to Ubuntu. Are you sure that it works on Debian? Which DE are you trying to use it on and why is the DE's own background "manager" not adequate? Lisi Debian package desktopnova rob -- 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/55829f5e.4090...@rektau.ukfsn.org
Re: CUPS default printer setting in Jessie
You should be able to use the CUPS interface: http://localhost:631 I use this for my Canon LBP7200Cdn printer, but I haven't gone to Jessie yet. It handled the setting up in the change from squeeze to wheezy. Rob Hurle ----- Rob Hurle e-mail:rob1...@gmail.com Mobile: 0417 293 603 (Australia) 0948 243 538 (Vietnam) Telephone: (02) 6236 3895 28 Mirrormere Rd, Burra, NSW 2620, Australia On 8 June 2015 at 12:43, Paul E Condon wrote: > My printer was working a few weeks ago at printing files in Emacs, > but now I get an error message that the default printer has not > been set. I don't recall having done any setting of the default > when it was working. I had thought it happened automagically > because of some special code in some package. But it seems not. > I think I should use lpoptions to set the default, but I can't > find any information about what a legitimate text string should > look like. Where can I find some examples of what one actually > types at the command prompt? And some discussion of how it works. > My printer is an HP Laserwriter 5MP, which is a very old model > that has centronix parallel and long distinct AppleTalk connector. > > The can't be much wrong except for configuration because I can > print documents from a modern MacBook Pro laptop running Mac OS X. > But that is more of a hope than a rational expection. > > Suggestions? > Please help. > > -- > Paul E Condon > pecon...@mesanetworks.net > > > -- > 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/20150608022609.ga6...@big.lan.gnu > >
Flash update
Hi there The most recent version is 11,2,202,457. For i386 that's: http://fpdownload.macromedia.com/get/flashplayer/pdc/11.2.202.457/install_flash_player_11_linux.i386.tar.gz Regards, Rob -- Comet: A very large bouncy castle. -- 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/mgl7d8$k7e$1...@ger.gmane.org
Re: apt stuck at "Reading database"
On 10/04/15 21:56, Luis Finotti wrote:
Dear all,
I've been trying to fix this problem, but have not found a solution
yet. (I've also asked at the aptosid list without success.)
Before I resigned myself to a reinstall, I thought I'd post the output of
strace apt-get dist-upgrade
(I killed the process a few seconds after it got stuck.) The first
4000 lines are here:
http://www.math.utk.edu/~finotti/misc/DU.log
At line 37886, we start to see
pselect6(43, [0 40 42], NULL, NULL, {0, 50}, {[], 8}) = 0 (Timeout)
wait4(8948, 0x7ffd2430a5c0, WNOHANG, NULL) = 0
repeat over and over. At line 38006, we see:
select6(43, [0 40 42], NULL, NULL, {0, 50}, {[], 8}) = 1 (in
[40], left {0, 264028})
read(40, "(Reading database ... \r", 1024) = 23
write(1, "(Reading database ... \r", 23) = 23
write(4, "(Reading database ... \r", 23) = 23
wait4(8948, 0x7ffd2430a5c0, WNOHANG, NULL) = 0
before the pattern above starts to repeat again.
Of the last 117592 lines, most are the (repeating ones) above, only
1012 are different. I've post these different ones, i.e., the result
of
tail -117592 DU.txt | grep -F -v 'wait4(8948, 0x7ffd2430a5c0,
WNOHANG, NULL) = 0' | grep -F -v 'pselect6(43, [0 40 42], NULL, NULL,
{0, 50}, {[], 8}) = 0 (Timeout)'
here:
http://www.math.utk.edu/~finotti/misc/DU2.log
Again, just a last attempt.
Best to all,
Luis
If you haven't already try:
rm /var/lib/apt/lists/* -vf
apt-get update
and if needed
dpkg --configure -a
--
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/552936bd.9090...@rektau.ukfsn.org
Flash update
Hi there Most recent version is 11.2.202.451. For i386 that's; http://fpdownload.macromedia.com/get/flashplayer/pdc/11.2.202.451/install_flash_player_11_linux.i386.tar.gz Regards, Rob -- Comet: A very large bouncy castle. -- 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/me14s1$gi1$1...@ger.gmane.org
Re: rsync question
On Fri, Feb 06, 2015 at 07:58:22PM +, Curtis Vaughan wrote: > I need to rsync several, but not all, the Maildir directories under /home. > I have a file with all the user accounts, so what would the script be to > say take each user account name and then rsync up each Maildir directory. > > The rsync command would be: > > rsync -arv ./[username]/Maildir/ root@[computer]:/home/[username]/Maildir/ > > I have already gone through each user's directory and manually done each > one over the course of the week. Now I want to be able to sink up using a > single command. > I think you want to use rsync's --include-from parameter. That lets you specify a file which contains all the includes. I don't think you need a catch-all --exclude=* after that, but I'm not 100% sure. Take a look at the FILTER RULES section of 'man rsync'. It would be something like: rsync -arv --include-from=my_include_file ./*/Maildir/ root@[computer]:/home/*/Maildir/ and my_include_file would be something like: user1/Maildir/ user2/Maildir/ user5/Maildir/ I think you want to specify 'Maildir' after each user. Otherwise you might end up including a directory named 'user1' which resides in user3's directory. In other words, be as specific as possible with your includes. I haven't tested any of this. Just going from memory. -Rob signature.asc Description: Digital signature
Re: An experiment in backup
On Fri, Jan 16, 2015 at 07:50:33PM -0800, Kevin O'Gorman wrote: > I've not tried other solutions. I worry that the ones folks seem to like > most do more > than I need or want in terms of management. I want my stuff where I can > see it, so > to speak, and where I can use the ancient tools (tar, dd, gzip and so on) > to work with it. > You might take a look at rsnapshot, then. It uses hard links to avoid backing up the same file more than once, but everything ends up in its own directory that you can see and manipulate with standard tools. Backuppc does put everything together where you can see it, but it does so by using its web interface or command line tools that are specific to backuppc. I understand you have a lot of storage space. But the less space your backups take up, the more backups you can have. I keep daily backups for x number of days, weekly backups for y days, yearly backups for z days. All of this for multiple machines. Another benefit of not backing up the same files over and over again is that you can perform backups over the internet to an offsite location, and it doesn't require as much bandwidth. But it is hard to argue against sticking with the tried and true "ancient tools". > I'm a bit ashamed I've never tried (or needed) a system restore before. So > whatever > this problem is, it's probably lurking in all of my backups. I may just > make image > backups of my bootable partitions real quick now, pending resolution of > this difficulty. > Yeah, you gotta test... But for my Linux systems I usually don't worry about backing up the OS. I keep install media handy and if I have a disaster, I reinstall the OS and then restore my data and certain things in /etc. But that is a more manual process than just restoring the whole thing at once. -Rob signature.asc Description: Digital signature
Re: An experiment in backup
On Thu, Jan 15, 2015 at 08:47:01PM -0800, Kevin O'Gorman wrote: > On Thu, Jan 15, 2015 at 8:41 PM, David Christensen < > > There are two basic kinds of "backups": > > > > 1. File system -- e.g. a copy of the files and directories on an mounted > > and operating drive. > > > > 2. Raw binary image -- e.g. a copy of the bytes on a drive taken when the > > drive is powered, but the partitions, volumes, file systems, etc., are not > > mounted. > > > > > > For system drives, the former won't work; you need the later. I connect a > > large hard drive (to hold the images), boot Debian installation media into > > rescue mode, and use 'dd' to backup/ restore system drive raw binary images. > > > > I was hoping for some details on why this won't work on system drives, or > conditions under which it just might. Another user has suggested I read > https://help.ubuntu.com/community/BackupYourSystem/TAR which suggests that > it actually should work. Image backups are definitely easier for doing disaster recovery of an entire machine. And when you have that kind of problem, you may really appreciate having to do less work / make fewer decisions. But filesystems backups can be used for disaster recovery. I've done it. One potential problem is that on a running system, things change. So at the start of your backup, you backup file A. At the end of the backup, you backup file Z. But in the middle of the backup, both file A and file Z have changed. And some software requires that file A and file Z be in sync. When you restore, those files are not in sync and you could have a problem. In practice, I haven't seen this be a problem much on home desktop machines. But that's not to say it couldn't be a problem. Another thing to consider is hardware changes. This can make certain devices be named differently when you restore. eth0 becomes eth1, and /etc/network/interfaces doesn't have a stanza for eth1. /dev/dvd becomes /dev/dvd1, and your cd burner was set to look for /dev/dvd which no longer exists. These things can be fixed in the /etc/udev/rules.d directory. UIDs of disk partitions will change. If /etc/fstab references UIDs, you need to update it. Same for /boot/grub/grub.cfg, although for that you run update-grub2 from the restored system (you'll need to boot with a live cd and chroot, or you'll need to boot with Super Grub Disk or similar). You will also need to install the bootloader on the new hard disk. 'grub-install /dev/sda' The UUID and Grub issues don't show up when restoring from an image backup, but the network card and cd burner issues can. There are a lot of free software backup solutions available. I would recommend using one of those, unless this endevour is more for learning experience than anything else. Backuppc may be overkill for your case, but it's pretty good. It will do file pooling and compression, so keeping multiple backups of one or more machines doesn't take up much disk space. -Rob signature.asc Description: Digital signature
Re: An experiment in backup
On Thu, Jan 15, 2015 at 07:19:52PM -0800, Kevin O'Gorman wrote: > I'm trying to develop a reliable backup method that does not use > proprietary tools or formats, and is free as in beer. I thought I had it, > but i just tried a restore, and it's a miserable failure. I wonder if > anyone here can point out the error of my ways. > > I have a tar backup of the entire system, excluding /sys, /proc and /dev. > I have a tar backup of a bind-mount of /dev. > These were taken while the system was running, but quiet. I did it this > way because I cannot get the system to boot into single user mode. Putting > "single" on the end of the "linux" like results in a black screen. > > I restored these, created /sys and /proc, and tried to boot the resulting > partition. It boots, but X does not come up, or even seem to try. I can > do a console login to my usual account, and stuff is there. > Check the permissions on /tmp. It should be drwxrwxrwt. Without the 't' at the end, stuff like X logins won't work (in my experience). signature.asc Description: Digital signature
Re: Preventing the computer from shutting down.
On Fri, Oct 31, 2014 at 04:10:11PM -0600, Mario Castelán Castro wrote: > Thanks everybody for their help. I will use molly-guard to guard > from accidental shut down from the CLI. Is there something like > molly-guard for the LXDE power off/close session button?. > > I must clarify that what I mean by “preventing the computer from > shutting down” is preventing the operator from inadvertently > shutting down or rebooting the system while it is undergoing a > backup. I know this is an old thread, but here's something that might work. I have a Wheezy system running LXDE and if you try to shut down using the gui while another user is logged in, it refuses. Actually, I think it requests the root password. So if your backup process involves an ssh login of a regular user (uid of 1000 or greater), other users won't be able to shut down the computer by normal gui methods. -Rob signature.asc Description: Digital signature
Re: problem with corrupted root password
On Thu, Jan 15, 2015 at 12:47:08PM +0100, mrr wrote: > On 15/01/2015 01:00, Gary Dale wrote: > >On 14/01/15 04:26 PM, Rob Owens wrote: > >>On Wed, Jan 14, 2015 at 03:07:09PM -0500, Comer Duncan wrote: > >>>I recently got wheezy up and running. I installed xfce4 and like it. > >>> > >>>However, today in the process of trying to spawn a root terminal (in > >>>Accessories) and going through a cycle of trying to get authorized but > >>>being prevented by repeated complaints that the system password I > >>>used was > >>>not correct, I now find that I can not get logged in in single-user > >>>mode! > >>>I have thus royally screwed up. So, how can I get the system password > >>>changed to something new? > >>> > >>>Thanks for help and apologies for making such an error. > >>Boot using a Live CD, then as root: > >> > >>mount /dev/sda1 /mnt/sda1 (or whatever device is your root partition) > >>chroot /mnt/sda1 > >>passwd > >I'd change the chroot command to > > chroot /mnt/sda1 bash > > > >to ensure you get the correct shell. System Rescue CD, for example, uses > >zsh by default so chrooting with specifying the shell will get you a > >not-found error. > > > > > > Wouldn't it work too if you delete the root line in /etc/shadow ? > When I say delete, I mean just the hash, you would leave a line: > > root: > I tried this with a test user and it worked. But there are fields after the password hash that remain. My user's line in /etc/shadow looked like this: junk::16450:0:9:7::: I was able to login as the 'junk' user without a password. I wasn't even asked for a password. However, 'su junk' from my regular user account did not work. > And just for fun (well maybe I'll try this later) if you take your > user hash in the same file and copy it in the place of the root > hash, wouldn't root get the same password as the user one? > I tested this too, and it worked for both a login and for su. Thanks for the ideas! -Rob signature.asc Description: Digital signature
Re: problem with corrupted root password
On Wed, Jan 14, 2015 at 03:07:09PM -0500, Comer Duncan wrote: > I recently got wheezy up and running. I installed xfce4 and like it. > > However, today in the process of trying to spawn a root terminal (in > Accessories) and going through a cycle of trying to get authorized but > being prevented by repeated complaints that the system password I used was > not correct, I now find that I can not get logged in in single-user mode! > I have thus royally screwed up. So, how can I get the system password > changed to something new? > > Thanks for help and apologies for making such an error. Boot using a Live CD, then as root: mount /dev/sda1 /mnt/sda1 (or whatever device is your root partition) chroot /mnt/sda1 passwd signature.asc Description: Digital signature
Re: Have I been hacked?
On Thu, Jan 08, 2015 at 02:20:27PM -0500, Jerry Stuckle wrote: > Danny, > > If you want to inspect further, I would suggest you look at each of the > jobs being run. See if they are what you expect them to be. Also check > your /etc/crontab and /etc/anacrontab to see what is in them. > And if you can't tell which job is the culprit, try running them one by one to see which one causes the problems you're seeing. Oh, and make sure you check users' cron jobs. You can find them in /var/spool/cron/*. All of this is in the name of curiosity only, however. Like others have said alread, a reinstall is in your future. -Rob signature.asc Description: Digital signature
Re: Please stop systemd-fsck on _every_ boot!
On Tue, Jan 06, 2015 at 06:23:18PM -0600, ~Stack~ wrote: > Greetings, > > This problem has been a minor annoyance for a while but only recently > have I started to use Jessie more and it is has finally peeved me off. I > have been trying everything I can find for the last two hours and I > still can't get systemd to STOP doing a fsck on _every_ boot! I suppose it's possible that the bios battery is no good, and the clock is getting reset after each shutdown. Then the "last time fsck'd" would be in the future, and I expect the system might fsck your disk as a precaution. You might never notice the clock issue if ntp or ntpdate is running and corrects the clock after the network is up. Good luck! -Rob signature.asc Description: Digital signature
Re: SFTP question
On Tue, Dec 23, 2014 at 02:52:28PM +0200, Danny wrote: > Hi guys, > > I am trying to setup SFTP (ssh) with ProFTP. > > My /etc/proftpd/conf.d/sftpd.conf looks like this: > > ### > > SFTPEngine on > Port 7003 > SFTPLog /var/log/proftpd/sftp.log > # Configure both the RSA and DSA host keys, using the same host key > # files that OpenSSH uses. > SFTPHostKey /etc/ssh/ssh_host_rsa_key > SFTPHostKey /etc/ssh/ssh_host_dsa_key > SFTPAuthMethods publickey > SFTPAuthorizedUserKeys file:/etc/proftpd/authorized_keys/%u > SFTPCompression delayed > > ## > > I added the following line in /etc/ssh/sshd_config: > > Subsystem sftp /usr/lib/openssh/sftp-server > > I generated a key for each user that will use SFTP located in their > /home/USER/.ssh/ directory > > As you can see, I have setup SFTP to listen on port 7003. > > My question is the following: The users that will connect to the ssh server > uses > FileZilla and mostly from Windows based machines. I copied the "id_rsa" key > files (which was generated on Debian) to the Windows user's "My Documents" > folder on Windows. I also added the (copied) "id_rsa" files to FileZilla. > > However, I get an "Authentication Failed" followed by a "Critical Error:Could > not connect to server" from FileZilla. Did you put the contents of id_rsa.pub into a file called /home/$USER/.ssh/authorized_keys on the sftp server? -Rob signature.asc Description: Digital signature
Re: init script can't find /bin/dirname
On Mon, Dec 22, 2014 at 10:55:15AM -0500, Rob Owens wrote: > On Mon, Dec 22, 2014 at 04:41:05PM +0100, Sven Joachim wrote: > > On 2014-12-22 04:10 +0100, Rob Owens wrote: > > > So I'm still not sure why /etc/init.d/mysql cannot find "dirname" and > > > "basename" when running at system boot, but it can find them when run > > > from a terminal after boot. > > > > Maybe you have a /bin/dirname executable whose interpreter is missing, > > e.g. it could be a 32-bit binary and you don't have libc6:i386 > > installed anymore. > > That's it! This system has a /bin/dirname as well as /usr/bin/dirname. > When I call /bin/dirname from a shell, I get "-bash: /bin/dirname: No > such file or directory". $ ls -l /bin/dirname -rwxr-xr-x 1 root root 13880 Jan 30 2007 /bin/dirname $ file /bin/dirname dirname: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.4.1, stripped Similar thing for /bin/basename and a bunch of other files that are present in both /bin and /usr/bin. For all the duplicates I've checked, the files in /bin are 32-bit and are old, while the files in /usr/bin are 64-bit and are recent. So I guess for now I'm going to move all these obsolete files into another directory and make sure the system works ok before I delete them for good. I'm not sure how this system ended up like this, but it's and old system that's been dist-upgraded many times as well as having the hard disk moved to different hardware. I checked Wheezy, Squeeze, and Lenny and the coreutils package doesn't put dirname or basename in /bin on any of those releases -- they go in /usr/bin. I wasn't able to go farther back than that. But 2007 is when Etch was released, and that was also the first Debian release I ever used. Sven, thanks for your help. Your tips got me heading down the right track. -Rob signature.asc Description: Digital signature
Re: init script can't find /bin/dirname
On Mon, Dec 22, 2014 at 04:41:05PM +0100, Sven Joachim wrote: > On 2014-12-22 04:10 +0100, Rob Owens wrote: > > So I'm still not sure why /etc/init.d/mysql cannot find "dirname" and > > "basename" when running at system boot, but it can find them when run > > from a terminal after boot. > > Maybe you have a /bin/dirname executable whose interpreter is missing, > e.g. it could be a 32-bit binary and you don't have libc6:i386 > installed anymore. That's it! This system has a /bin/dirname as well as /usr/bin/dirname. When I call /bin/dirname from a shell, I get "-bash: /bin/dirname: No such file or directory". My other Wheezy system (which was installed as a 64-bit system) only has /usr/bin/dirname. OK, I guess I've got a little research to do. If you have any clue how to straighten this out, I'm all ears. Thanks for all your help! -Rob signature.asc Description: Digital signature
Re: init script can't find /bin/dirname
On Sun, Dec 21, 2014 at 06:07:20PM -0500, Rob Owens wrote: > On Sun, Dec 21, 2014 at 10:32:07PM +0100, Sven Joachim wrote: > > On 2014-12-21 21:58 +0100, Rob Owens wrote: > > > I know I could add a PATH statement to the init script, but this problem > > > is my own doing and I'd like to fix it right. I "cross-graded" this > > > system from 32-bit to 64-bit using this guide: > > > > > > www.ewan.cc/?q=node/90 > > > > > > It worked pretty well, but not perfectly. Some things got missed, like > > > screen, ntp, and a couple other packages. I'm thinking that maybe > > > another missing package is preventing the mysql init script from > > > checking for /usr/bin/dirname. > > > > That seems to be unlikely, but you could add a line with "echo $PATH" to > > the script to find out what PATH actually is. > > > I will do this, but I can't reboot just yet to check it out. I will > post my findings later. I rebooted and found that the path is fine: /sbin:/usr/sbin:/bin:/usr/bin So I'm still not sure why /etc/init.d/mysql cannot find "dirname" and "basename" when running at system boot, but it can find them when run from a terminal after boot. Any advice would be appreciated! -Rob signature.asc Description: Digital signature
Re: init script can't find /bin/dirname
On Sun, Dec 21, 2014 at 10:32:07PM +0100, Sven Joachim wrote: > On 2014-12-21 21:58 +0100, Rob Owens wrote: > > > The /etc/init.d/mysql script on one of my systems is complaning that it > > can't find /bin/dirname and /bin/basename. Line 24 of the script is > > this: > > > > SELF=$(cd $(dirname $0); pwd -P)/$(basename $0) > > > > Both dirname and basename live in /usr/bin, not /bin. > > Which should be fine from the script's point of view as long as > $remote_fs is in the "Required-Start" header (i.e. /usr is guaranteed to > be mounted). > It is. In any case, /usr is part of the / partition. In case it matters, /var is a separate partition. And I'm using software raid 1 for both / and /var. > > I know I could add a PATH statement to the init script, but this problem > > is my own doing and I'd like to fix it right. I "cross-graded" this > > system from 32-bit to 64-bit using this guide: > > > > www.ewan.cc/?q=node/90 > > > > It worked pretty well, but not perfectly. Some things got missed, like > > screen, ntp, and a couple other packages. I'm thinking that maybe > > another missing package is preventing the mysql init script from > > checking for /usr/bin/dirname. > > That seems to be unlikely, but you could add a line with "echo $PATH" to > the script to find out what PATH actually is. > I will do this, but I can't reboot just yet to check it out. I will post my findings later. > > I can run '/etc/init.d/mysql start' from a terminal and it works fine, > > because it picks up the PATH associated with the terminal. So this is > > only a boot-time issue. > > > > So is there a package or a global setting somewhere that sets the PATH > > for init scripts? > > PATH is set directly by init. If I read the source code correctly, > sysvinit sets it to "/sbin:/usr/sbin:/bin:/usr/bin" while systemd uses > "/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin". > This is a Wheezy system, running with sysvinit. 'man init' says: ENVIRONMENT Init sets the following environment variables for all its children: PATH /bin:/usr/bin:/sbin:/usr/sbin So maybe it's not a path issue after all. What else could it be then? The boot messages clearly indicate that it's failing to find /bin/dirname and /bin/basename. Thanks for your help. -Rob signature.asc Description: Digital signature
init script can't find /bin/dirname
The /etc/init.d/mysql script on one of my systems is complaning that it can't find /bin/dirname and /bin/basename. Line 24 of the script is this: SELF=$(cd $(dirname $0); pwd -P)/$(basename $0) Both dirname and basename live in /usr/bin, not /bin. I know I could add a PATH statement to the init script, but this problem is my own doing and I'd like to fix it right. I "cross-graded" this system from 32-bit to 64-bit using this guide: www.ewan.cc/?q=node/90 It worked pretty well, but not perfectly. Some things got missed, like screen, ntp, and a couple other packages. I'm thinking that maybe another missing package is preventing the mysql init script from checking for /usr/bin/dirname. I can run '/etc/init.d/mysql start' from a terminal and it works fine, because it picks up the PATH associated with the terminal. So this is only a boot-time issue. So is there a package or a global setting somewhere that sets the PATH for init scripts? -Rob signature.asc Description: Digital signature
mounting removable media without systemd on Jessie
Just posting something I discovered recently in case anybody's interested... On a Jessie system without systemd you can mount USB sticks and other removable media using the pmount command, as others have pointed out on this list. I was looking for a gui method, and I've found it. Spacefm is a file manager that will use any of the following to mount removable media: udevil pmount udisks2 (depends on libpam-systemd) udisks I've tested its use with pmount and udevil and they both work. I find udevil a bit nicer, because I was able to get the mount point to be /media/$USER/label by setting the following in /etc/udevil/udevil.conf: allowed_media_dirs = /media/$USER I'm not claiming that this is a perfect solution. For instance, spacefm and udevil are currently unmaintained. But I thought some folks on this list would be interested to know about this option anyway. -Rob signature.asc Description: Digital signature
