Re: how do I restart lagg0 properly?
Erich Dollansky writes: I have used this configuration with success: ifconfig_em0=up ifconfig_iwn0=ether MAC address of em0 wlans_iwn0=wlan0 ifconfig_wlan0=WPA cloned_interfaces=lagg0 ifconfig_lagg0=laggproto failover laggport em0 laggport wlan0 Your IP address This configuration switched automatically between the available connections prefering em0. There are no routing problems, nothing of this sort at all. Thanks for confirming the setup. I basically use the same but with DHCP enabled. I've done quite a bit of testing lately. Turns out that *sometimes* lagg0 works as advertized. When the box was booted with WLAN available only, I can plug in a patch cable and connect to my DSL router via LAN. Most of the times though, this does not work. After plugging in the cable, I cannot make any outbound connections. The box can be pinged from other boxes in my LAN though. *Sometimes* the following commands will enable networking after plugging in the patch cable in these cases: service netif start service routing restart But then, most of the times this does not help either. I'm at a loss now. I'm actually wondering if the problem is located on the other end of the line (I'm using a Fritz!Box 7113 DSL router for both LAN and WLAN). regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
how do I restart lagg0 properly?
Hi, I've tried to setup link aggregation to allow networking through WLAN and ethernet on my laptop, see: http://lists.freebsd.org/pipermail/freebsd-questions/2013-January/248605.html (this thread contains my network setup information) The lack of replies may indicate that it is not possible to set up lagg in a way which allows to setup networking either via WLAN or via ethernet, whichever is accessible during boot. In any case I figured that I should be able to bring down networking completely and restart it completely after I plugged in a LAN cable to work around this limitation. I tried things like: service netif restart lagg0 I understand that both routing and DHCP will be affected by temporarily bringing down the network, so I added commands like these in various orders: service routing restart service dhclient restart All in all, to no avail. I can't even ping my DSL router after fiddling with lagg0 after boot. What is the correct incantation of commands to cleanly shut down and restart my networking? thanks, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: lagg problems (or lack of understanding?)
Jerry writes: On Sun, 27 Jan 2013 01:20:13 +0100 markus.hoeni...@mhoenicka.de articulated: hald_enable=NO Its defaults to NO. No reason to specifically set it. That's correct. This is a leftover of an attempt to track down some problem which turned out to be influenced by hald. IIRC I needed hald only for xfburn as I use an external CD burner. This screwed up other stuff, so today I start hald manually in the rare cases where I need it. regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: lagg problems (or lack of understanding?)
Erich Dollansky writes: Hi, On Sat, 26 Jan 2013 00:55:59 +0100 markus.hoeni...@mhoenicka.de wrote: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-aggregation.html Specifically, I use the rc.conf entries mentioned in the box at the very bottom on the page, except that my ethernet interface is re0 instead of bge0. The lagg module is loaded during boot using an appropriate entry in /boot/loader.conf. if you could post your settings here? Oh yes, certainly, although I think that I use pretty much exactly what the instructions said. /boot/loader.conf: verbose_loading=YES # print verbose loader output during boot coretemp_load=YES # intel temp sensor support hw.psm.synaptics_support=1 # improved touchpad support atapicam_load=YES # access CD drive through SCSI subsystem cuse4bsd_load=YES # required for webcamd sem_load=YES # required by firefox for displaying HTML5 if_lagg_load=YES #link aggregation /etc/rc.conf: hostname=wombat.mininet ifconfig_re0=up ifconfig_iwn0=ether 00:1e:33:26:4c:3e wlans_iwn0=wlan0 ifconfig_wlan0=WPA cloned_interfaces=lagg0 ifconfig_lagg0=laggproto failover laggport re0 laggport wlan0 DHCP sshd_enable=YES moused_enable=NO powerd_enable=YES powerd_flags=-a max -b adaptive -i 90 cupsd_enable=YES mysql_enable=YES linux_enable=YES dbus_enable=YES hald_enable=NO anacron_enable=YES # Set dumpdev to AUTO to enable crash dumps, NO to disable dumpdev=NO # -- sysinstall generated deltas -- # Thu Feb 16 22:33:22 2012 rpcbind_enable=YES rpc_statd_enable=YES rpc_lockd_enable=YES nfs_server_enable=YES nfs_client_enable=YES devfs_system_ruleset=localrules fusefs_enable=YES Is there anything suspcicious? regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
lagg problems (or lack of understanding?)
Hi, I'd like to use both ethernet and wlan, whichever is available, with my laptop. In many cases I prefer wlan at home as I'm not tethered and performance is good enough. However, in some cases I prefer to connect to a switch via ethernet to have faster connection to a desktop PC. I followed these instructions: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-aggregation.html Specifically, I use the rc.conf entries mentioned in the box at the very bottom on the page, except that my ethernet interface is re0 instead of bge0. The lagg module is loaded during boot using an appropriate entry in /boot/loader.conf. Things almost work: boot with ethernet plugged in, ethernet still in: connected via ethernet boot with ethernet plugged in, ethernet out: connected via wlan boot with ethernet unplugged, ethernet still out: connected via wlan boot with ethernet unplugged, ethernet in: no connection What bothers me is the last scenario: this would force me to boot with ethernet plugged in whenever I intend to use ethernet in the same session. Is this expected behaviour? Is there a way to fix this? regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: installworld strangeness
David Noel david.i.n...@gmail.com was heard to say: I tried upgrading to 9.1 also but ran into a wall there as well: === usr.sbin/lpr/filters.ru (install) install -o root -g wheel -m 444 /usr/src9.1/usr.sbin/lpr/filters.ru/bjc-240.sh.sample /usr/libexec/lpr/ru === usr.sbin/lpr/filters.ru/koi2alt (install) install -s -o root -g wheel -m 555 koi2alt /usr/libexec/lpr/ru strip: could not create temporary file to hold stripped copy of /usr/libexec/lpr/ru/koi2alt *** [_proginstall] Error code 70 Stop in /usr/src9.1/usr.sbin/lpr/filters.ru/koi2alt. *** [realinstall] Error code 1 Stop in /usr/src9.1/usr.sbin/lpr/filters.ru. *** [realinstall] Error code 1 Stop in /usr/src9.1/usr.sbin/lpr. *** [realinstall] Error code 1 Stop in /usr/src9.1/usr.sbin. *** [realinstall] Error code 1 Stop in /usr/src9.1. *** [reinstall] Error code 1 Stop in /usr/src9.1. *** [installworld] Error code 1 Stop in /usr/src9.1. *** [installworld] Error code 1 Stop in /usr/src9.1. Google points me to information that suggests the failure is caused by a lack of disk space, but that doesn't seem applicable to my case as I have plenty. Could anyone offer insight into either of these errors? Total disk space may not be an issue here, but I faintly recall that the default disk layout before 8.2 created too small a root partition. Rebuilding a kernel on 8.1 required more disk space on / than the default 500 MB. If that hits you here as well, you may need to resize partitions. regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: printing jpeg slides to a Postscript printer
Matthias Apitz g...@unixarea.de was heard to say: I was thinking in some UNIX way to do so, like: 1. converting the 10 jpeg to 10 .eps files 2. running psmerge to bring the 10 EPS files into one PS file 3. running psnup to get the 10 pages re-arranged, 2 on one page; If you're looking at 10 images, just fire up OpenOffice and make a presentation or a drawing with 2 images per page. Figure out the Unix way if you're looking at 10 images every day :-) just my 2 cc Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: cups slow when printing from firefox
Leslie Jensen writes: Hi Markus. If you do a search in the archives you'll see that some of us have printing problems with Firefox. Some even suspect that printing in Firefox is broken. I myself has not been able to solve my specific printing problems which include both Firefox but also Libre office. I'm running Win7 in Virtualbox in order to fix my printing :-( Hi Leslie, good to know, I thought it's just me being too stupid to press the right button. For the time being I'll resort to printing to a file and sending that to lpr, instead of running Win7. regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
cups slow when printing from firefox
Hi all, I'm struggling with a speed issue when trying to print stuff from firefox (10.0.2,1). I've set up cups (meta package 1.5.2) on my box which uses a Lexmark E360dn printer via ethernet. The printer has a postscript emulation which is set as the default (as opposed to PCL). I've used a PPD file from openprinting.org which claims that the printer is 100% supported. Printing works fine in general with all sorts of text and images, including PDF and LibreOffice and whatnot. However, trying to print maps from maps.google.de sends one CPU to 100% for a couple of minutes. Printing the same map from a Debian box takes a couple of seconds. I can print the map to a file, and print that file with lpr which also takes just a few seconds. While printing from firefox directly, I noticed a process gsc owned by cups which causes most of the CPU load. I take this as an indication that the postscript output from firefox is incorrectly rasterized on my box, instead of sending the postscript data directly to the printer. I did not make any changes to the default config files except for adding the printer through the localhost:631 interface. Is there anything else that I need to configure, either on the firefox or the cups end, to make printing maps faster? regards Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Emacs splash screen went to textmode only
Quoting Ashish SHUKLA ash...@freebsd.org: And please also evaluate following functions in Emacs (in X11) and paste their output: This is the output from 23.2.1, i.e. the one which doesn't show the fancy splash screen: (use-fancy-splash-screens-p) nil (display-graphic-p) t (display-color-p) t (image-type-available-p 'xpm) t (image-type-available-p 'xbm) t For comparison, this is what 23.1.1. reports: (use-fancy-splash-screens-p) t (display-graphic-p) t (display-color-p) t (image-type-available-p 'xpm) t (image-type-available-p 'xbm) t The obvious difference is the value of (use-fancy-splash-screens-p). Both Emacsen were started with -q -no-site-file to be on a level ground. regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Emacs splash screen went to textmode only
Quoting Ashish SHUKLA ash...@freebsd.org: Well I use a tiling WM (Xmonad), and to reproduce your issue, I started 12 instances of 'xterm' and then started an Emacs instance with '-q -no-site-file' which caused Emacs to start with a smaller window where I got a non-fancy splash screen. Try starting your Emacs in full-screen mode .e.g. 'emacs -fs'. That works indeed. That is, I just have to make sure Emacs starts up with a sufficient frame height. emacs -g 80x40 is all it takes. Thanks a lot for your help! Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Emacs splash screen went to textmode only
Quoting Olivier Nicole olivier.nic...@cs.ait.ac.th: Hi, 4) Another instance of Emacs on my desktop, this time running in a terminal. The startup screen is yet another one. In this one, I clearly recognize emacs being run inside an ascii terminal. Ascii terminal do not have graphic possibility (nor mousse) so it is only normal that there is no graphic splash screen. Yes, I am fully aware of that. In my original post I wondered if Emacs for some reason *thinks* it is running in a terminal and thus displays the terminal mode splash screen. But #2 and #4 are different, thus I proved myself wrong. If it is the same version of emacs that you run on the instances 2 and 3, I would look for a X issue. Do you have an environment variable set in the xterm? In that xterm, can you launch another xterm? Well, set displays a few screens full of environment variables, so, yes, of course there are some and they have to be there. But none of these affected Emacs before upgrading to 23.2.1. BTW launching another xterm from an xterm works without problems. regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Emacs splash screen went to textmode only
Quoting Ashish SHUKLA ash...@freebsd.org: Can you please provide output of following: 1. xrdb -q|grep '^Emacs' from an terminal running in X11. [mar...@wutz /usr/home/markus]$ xrdb -q|grep '^Emacs' Emacs*font: Bitstream Vera Sans Mono-8 2. (display-grayscale-p) in *scratch* buffer in GNU Emacs running in X11. (display-grayscale-p) t 3. system-configuration-options in *scratch* buffer in GNU Emacs. system-configuration-options '--with-x-toolkit=gtk' '--x-libraries=/usr/local/lib' '--x-includes=/usr/local/include' '--prefix=/usr/local' '--mandir=/usr/local/man' '--infodir=/usr/local/info/' '--build=i386-portbld-freebsd8.0' 'build_alias=i386-portbld-freebsd8.0' 'CC=cc' 'CFLAGS=-O2 -pipe -fno-strict-aliasing' 'LDFLAGS=-L/usr/local/lib -lintl' 'CPPFLAGS=-I/usr/local/include' BTW I've first built the Emacs port, and after I realized there was that splash screen issue I installed the package instead, just in case I managed to misconfigure the build somehow. Didn't make a difference. Thanks Thank you! Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: Emacs splash screen went to textmode only
Ashish SHUKLA writes: Can you upload the screenshot of what you're getting ? Hi, I've uploaded a screenshot right here: http://www.abload.de/img/emacs-splashaws1.png The image shows four incarnations of Emacs. These are, clockwise starting from upper left: 1) Emacs started remotely on my laptop through ssh. This Emacs is slightly older (23.1.1) than the one on my desktop (23.2.1). This is what an unfiddled-with splash screen is supposed to look like in recent versions. 2) Emacs started on my desktop, using -q -no-site-file as per your suggestion. There are no graphics and no colors. 3) Another instance of Emacs started on my desktop, also with -q -no-site-file. I switched to *scratch* and evaluated (fancy-startup-screen). The fancy startup screen pops up without a hitch. 4) Another instance of Emacs on my desktop, this time running in a terminal. The startup screen is yet another one. Any clues? regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Emacs splash screen went to textmode only
Hi, I've recently updated Emacs to the latest version (23.2_2,2 from 23.1.1) along with all dependencies. Since then, the startup screen when running in an X window shows b/w text only like in the days of yore. Not that the startup screen graphics are that important, but things that don't work as expected tend to make me nervous. According to startup.el, the following variables affect what is shown at startup. Evaluating them in the scratch buffer looks like this: inhibit-startup-screen nil initial-buffer-choice nil That is, the startup screen *should* be displayed, or am I missing something here? The interesting thing is that nothing else about the startup screen appears broken. If I evaluate (fancy-startup-screen) in the scratch buffer, i.e. the function which actually displays the startup screen, Emacs faithfully does so, including the graphics and all bells and whistles. Has anything fundamentally changed between these versions? Does Emacs think it runs in a terminal? Any clues would be greatly appreciated. regards, Markus -- Markus Hoenicka http://www.mhoenicka.de AQ score 38 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: xfburn fails with 'Undefined symbol __malloc_lock'
Mel Flynn writes: -PP will fail if for some reason the package is not available on the servers. It is better to use -P when crossing major releases, so that any restricted packages that are unavailable on the buildservers are built from source. I suspect this is the root of the problem, though they look to be available: I resorted to reinstalling everything, using -Pk instead of -PP this time. This apparently helped fix this problem. thanks Markus -- Markus Hoenicka markus.hoeni...@cats.de (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: xfburn fails with 'Undefined symbol __malloc_lock'
Manolis Kiagias writes: Upgrading between major versions requires all installed ports to be rebuilt, so they get linked to the new versions of the libraries. I suppose you missed this step, older apps may still work but there is a problem installing new ones. [...] (AFAIR, if you upgraded via source, you will also need to run make delete-old-libs in /usr/src after successfully recompiling ports) This was kind of an experience, as make delete-old-libs almost bricked my laptop. Apparently portupgrade failed to upgrade quite a few apps, for whatever reasons. I deleted all cached packages to make sure portupgrade -P would fetch the newest. I also had to coerce /usr/bin/objformat to return elf as the installed objformat prevented building many ports that could not be installed as packages. Anyway, everything seems to work ok now. thanks a lot Markus -- Markus Hoenicka markus.hoeni...@cats.de (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: xfburn fails with 'Undefined symbol __malloc_lock'
Quoting Manolis Kiagias son...@otenet.gr: Upgrading between major versions requires all installed ports to be rebuilt, so they get linked to the new versions of the libraries. I suppose you missed this step, older apps may still work but there is a problem installing new ones. Please see the instructions at the end of section 24.2.3 (portupgrade etc): This is why I'm at a loss here. I upgraded the base system from the sources and then used portupgrade -af to rebuild all installed packages. Right after upgrading the base system, essentially everything from X to Emacs was broken (I tried, just for the heck of it). After running portupgrade, these problems were all fixed - except for xfburn. (AFAIR, if you upgraded via source, you will also need to run make delete-old-libs in /usr/src after successfully recompiling ports) I'll try that, as I've never seen this before. I definitely didn't do that. regards, Markus -- Markus Hoenicka markus.hoeni...@cats.de (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
xfburn fails with 'Undefined symbol __malloc_lock'
Hi, I've upgraded my laptop from 6.4 to 7.2-RELEASE. Essentially everything went fine, except that for some reason xfburn no longer works. If I install a package using portupgrade -f -PP, I see the following at runtime: mar...@yeti:/usr/home/markus# xfburn [1] 47214 mar...@yeti:/usr/home/markus# /libexec/ld-elf.so.1: /lib/libpthread.so.2: Undefined symbol __malloc_lock If I build xfburn in the ports tree, I get the following error at compile time: /usr/bin/ld: warning: libcam.so.3, needed by /usr/local/lib/libburn.so, may conflict with libcam.so.4 /lib/libpthread.so.2: undefined reference to `__malloc_lock' I assume that I somehow managed to botch the 6-7 upgrade, but would anyone know how to fix this particular problem? regards, Markus -- Markus Hoenicka markus.hoeni...@cats.de (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: dlsym can't use handle returned by dlopen?
Markus Hoenicka writes: Don't mean to nag, but is there any news on this? just for the record: turns out this was a bug that got fixed in 6.3. See http://www.freebsd.org/cgi/query-pr.cgi?pr=kern/129031 regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: presentation application (other than OpenOffice)?
Quoting Andrew Gould [EMAIL PROTECTED]: I'm browsing around for smaller office apps. Abiword and Gnumeric are great. Does anyone have any recommendations for a presentation software? I'd rather not compile OpenOffice; and I'm just not getting KDE4. Hi Andrew, if you're into LaTeX, then prosper (http://sourceforge.net/projects/prosper/) might be an option. If you're into XML and don't need no fancy effects, then DocBook Slides (http://docbook.sourceforge.net/release/slides/3.4.0/RELEASE-NOTES.html) may come in handy. Also, if you don't need any effects or transitions you may consider any tool that creates multi-page PDF files. Adobe Reader has a full-screen option which turns it into a presentation viewer. hth Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: presentation application (other than OpenOffice)?
Quoting Paul B. Mahol [EMAIL PROTECTED]: xpdf also support full-screen option and is much lighter than Adobe Reader, and doesnt depends on linux stuff. evince can be of use if you already have it installed, because it supports fullscreen and presentation mode. Yes, this is correct. I was thinking about the systems which are available when you present your talk (you're not always allowed to plug in your own box). Whenever I give a talk I have to deal with Windows systems. That's why I suggested Adobe Reader. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: dlsym can't use handle returned by dlopen?
Don't mean to nag, but is there any news on this? regards, Markus Markus Hoenicka writes: Jeremy Chadwick writes: As promised: http://www.malkavian.com/~jdc/myprog.tar.gz This test program indeed works as expected. However, this doesn't quite reflect the situation in libdbi. I took your files and modified them accordingly, see: http://libdbi.sourceforge.net/downloads/dlsymtest.tar.gz To run the test use: LD_LIBRARY_PATH=. ./myprog We need to set the environment variable to let the linker pick up a shared object that gmake builds. myprog.c now just calls a function which is provided in libmylib (built from mylib.c). The latter file does most of what your test case did in myprog.c. The second major change is that myshared.so is linked against libmysqlclient (just like a libdbi database driver is linked against the client library). myfunc now calls a MySQL function to show that it is accessible (if you don't have libmysqlclient handy, you can replace it with whatever function from some .so is convenient) Finally, libmylib tries to obtain a pointer to that MySQL function by means of a dlsym call. This new dlsym call, in contrast to the existing one that acesses myfunc in myshared.so, indeed fails: myint = 0xdeadbeef (3735928559) == entered myfunc() == double = 3.141590 ==mysql client version is 50051 == exiting myfunc() dlsym() in shared lib failed: Undefined symbol mysql_get_client_version So, to make the problem clear again: while dlsym works when accessing symbols in dlopen()ed objects, it fails to access symbols which are linked into such an object if you use the handle returned by dlopen(). This is different from other OSes. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: dlsym can't use handle returned by dlopen?
Jeremy Chadwick writes: As promised: http://www.malkavian.com/~jdc/myprog.tar.gz This test program indeed works as expected. However, this doesn't quite reflect the situation in libdbi. I took your files and modified them accordingly, see: http://libdbi.sourceforge.net/downloads/dlsymtest.tar.gz To run the test use: LD_LIBRARY_PATH=. ./myprog We need to set the environment variable to let the linker pick up a shared object that gmake builds. myprog.c now just calls a function which is provided in libmylib (built from mylib.c). The latter file does most of what your test case did in myprog.c. The second major change is that myshared.so is linked against libmysqlclient (just like a libdbi database driver is linked against the client library). myfunc now calls a MySQL function to show that it is accessible (if you don't have libmysqlclient handy, you can replace it with whatever function from some .so is convenient) Finally, libmylib tries to obtain a pointer to that MySQL function by means of a dlsym call. This new dlsym call, in contrast to the existing one that acesses myfunc in myshared.so, indeed fails: myint = 0xdeadbeef (3735928559) == entered myfunc() == double = 3.141590 ==mysql client version is 50051 == exiting myfunc() dlsym() in shared lib failed: Undefined symbol mysql_get_client_version So, to make the problem clear again: while dlsym works when accessing symbols in dlopen()ed objects, it fails to access symbols which are linked into such an object if you use the handle returned by dlopen(). This is different from other OSes. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: dlsym can't use handle returned by dlopen?
Quoting Jeremy Chadwick [EMAIL PROTECTED]: When I looked at the resulting symbol names using nm or objdump, certain characters were prepended to them. There's a gcc or ld flag which disables this behaviour. I'll have to dig around to remind myself what it is. Once I read about it, it made perfect sense. Again, if you want me to write some code and provide some output of what I'm talking about, I can do so. I'd greatly appreciate any help here. I feel what I'm doing now is writing ugly hacks to make things work somehow. I'd prefer to do it properly. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: dlsym can't use handle returned by dlopen?
Quoting Jeremy Chadwick [EMAIL PROTECTED]: I know that the .so's you're loading with dlopen() need to be built a specific way/with certain arguments, otherwise they won't work (I believe what I saw was dlsym() returning NULL). My symbol names were getting stomped on, and there was a compiler flag that addressed that. Is that a BSD-specific problem? As mentioned previously, I don't run into trouble on other platforms. Is there any documentation available which tells me how to build a dlopen()'able object in a portable way? function_pointer = dlsym(RTLD_DEFAULT, function_name); Why is that? Or rather: what am I doing wrong? This code right here is *completely* wrong. RTLD_DEFAULT is a mode bit for dlopen(). I'm willing to bet a strict set of warnings would Citing the FreeBSD dlsym(3) man page: If dlsym is called with the special handle RTLD_DEFAULT, the search for the symbol follows the algorithm used for resolving undefined symbols when objects are loaded. You probably had RTLD_LAZY and RTLD_NOW in mind which are dlopen() flags. BTW RTLD_NEXT works just as well instead of RTLD_DEFAULT. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: dlsym can't use handle returned by dlopen?
Quoting Jeremy Chadwick [EMAIL PROTECTED]: As promised: http://www.malkavian.com/~jdc/myprog.tar.gz I couldn't figure out what the gcc flag was that I needed to keep certain characters from getting prepended to the symbol names. I believe the char added was an underscore, but I could be wrong. Either way, the example should help you, I think. (You can change RTLD_LAZY to RTLD_NOW and it still functions as expected) Thanks a lot, I'll have a look at it when I get home tonight. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
dlsym can't use handle returned by dlopen?
Hi, FreeBSD yeti.mininet 6.1-RELEASE FreeBSD 6.1-RELEASE #1: Mon Aug 28 22:24:48 CEST 2006 [EMAIL PROTECTED]:/usr/src/sys/i386/compile/YETI i386 I'm trying to do the following: libdbi (http://libdbi.sourceforge.net) is a database abstraction layer which is linked as a shared object into applications. The libdbi library uses dlopen() to load database drivers which in turn are linked against database client libraries. Now I want to access functions defined in the database client library from within the libdbi library. In brief, I thought this is going to work like this: dlhandle = dlopen(path, RTLD_NOW); ... function_pointer = dlsym(dlhandle, function_name); dlhandle is not NULL and does not crash the app when passed to dlclose(), so I assume the handle is valid. Accessing the functions does work on most systems (Linux, OSX, Cygwin, to name a few), but I get Undefined symbol errors on FreeBSD. Interestingly, the following does work: function_pointer = dlsym(RTLD_DEFAULT, function_name); Why is that? Or rather: what am I doing wrong? Any help is appreciated. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
linux-flashplugin-7.0 freezes firefox - weird fix included
Robert Huff writes: linuxpluginwrapper-20051113_8 There's a reason the latest version of this available in ports is dated over two years ago: it's no longer the correct tool. Instead, use www/nspluginwrapper. (Pay attention to the post-install messages.) I see, I must have missed that message. After removing linuxpluginwrapper and running nspluginwrapper everything seems to run just fine. Thanks a lot Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
linux-flashplugin-7.0 freezes firefox - weird fix included
Hi, I hardly dare to report this problem as it, let alone the fix, seems quite ridiculous to me. Before filing a bug report I'd like to ask if anyone can reproduce the problem (I see it on two computers running FreeBSD) or tell me what I'm doing wrong. Nevertheless the problem seems reproducible, so here goes. System information: FreeBSD yeti.mininet 6.1-RELEASE FreeBSD 6.1-RELEASE #1: Mon Aug 28 22:24:48 CEST 2006 [EMAIL PROTECTED]:/usr/src/sys/i386/compile/YETI i386 All packages were rebuilt from the ports about a week ago due to an upgrade of xorg from 6.9 to 7.3 (however, the problem was present before the upgrade as well). The relevant port versions are: firefox-2.0.0.12_1.1 linux-flashplugin-7.0r73 linuxpluginwrapper-20051113_8 The browser plugins directory looks like this after upgrading: [EMAIL PROTECTED]:~# ls -al /usr/local/lib/browser_plugins total 2202 drwxr-xr-x 2 root wheel 512 Apr 7 23:09 . drwxr-xr-x 86 root wheel40960 Apr 7 22:53 .. -rw-r--r-- 1 root wheel0 Mar 15 12:20 .firefox.keep -rw-r--r-- 1 root wheel0 Mar 15 12:58 .mozilla.keep lrwxr-xr-x 1 root wheel 54 Apr 7 22:55 flashplayer.xpt - /usr/local/lib/npapi/linux-flashplugin/flashplayer.xpt lrwxr-xr-x 1 root wheel 56 Apr 7 23:16 libflashplayer.so - /usr/local/lib/npapi/linux-flashplugin/libflashplayer.so lrwxr-xr-x 1 root wheel 60 Mar 15 12:20 libjavaplugin_oji.so - /usr/local/jdk1.5.0/jre/plugin/i386/ns7/libjavaplugin_oji.so lrwxr-xr-x 1 root wheel 59 Aug 27 2006 nppdf.so - /usr/X11R6/Adobe/Acrobat7.0/ENU/Browser/intellinux/nppdf.so There is one particular site (http://www.heise.de, a German computer news site) which causes firefox to freeze with this setup. Unfortunately the problem appears to be caused by ads that they run, so depending on what ads are scheduled you may not be able to reproduce the bug immediately (try again a day or two later). I haven't found a site yet which consistently shows this bug. Freezing can be suppressed by any of the following manipulations: 1) switch off JavaScript support in the browser settings 2) remove the libflashplayer.so link in the plugins directory 3) and finally the weird one: replace the libflashplayer.so link with the shared object that the link points to, which makes the browser_plugins entry look like this: -rw-r--r-- 1 root wheel 2181284 Apr 7 23:09 libflashplayer.so I was under the impression that symbolic links are handled transparently by the OS, but there seems to be something weird going on here. Any clues? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: dlopen(), atexit() on FreeBSD
Giorgos Keramidas writes: The __cxa_finalize() function is not called by atexit(), but by exit() I see. Since the program is going to exit and have all its dlopened shared objects be unmapped, it's probably ok to skip the dlclose() step in this example. Just for the record: I've tried disabling the dlclose() call in our code. It both prevents the segfaults and does not seem to inflict any harm. I'll use the code that way for the time being, any real solution for this problem notwithstanding. I think this is probably something that the freebsd-hackers list will be interested in. Can you post a description of the problem there too? I'll do. I haven't read up the full history of this problem, but apparently other programs and libraries are affected too, so it is probably fair to seek assistance. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
dlopen(), atexit() on FreeBSD
Hi, I bumped into a platform-specific problem when using the Firebird database client library in a dlopen()ed module on FreeBSD. libdbi (http://libdbi.sourceforge.net) is a database abstraction layer which dlopen()s available database drivers at runtime to provide connectivity to various database engines. This design works without problems on a variety of platforms and with a variety of database client libraries, but causes a segfault with Firebird on FreeBSD: #0 0x28514fe4 in ?? () #1 0x281507c3 in __cxa_finalize () from /lib/libc.so.6 #2 0x281503fe in exit () from /lib/libc.so.6 #3 0x0804a40f in main (argc=1, argv=0xbfbfe754) at test_dbi.c:419 The application crashes when exit() is called. Googling told me that __cxa_finalize () is invoked by atexit(). Our drivers and apps do not use this function, but the firebird client libraries do: [EMAIL PROTECTED]:~/prog/libdbi-drivers/tests# grep atexit /usr/local/lib/libfb* Binary file /usr/local/lib/libfbclient.so matches Binary file /usr/local/lib/libfbembed.so matches Googling also told me that the conflict between atexit() and dlopen() on FreeBSD is a known problem, see e.g.: http://www.imagemagick.org/pipermail/magick-developers/2006-March/002523.html Is there anything I can do about this from my end? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
threading and dlopen()
Hi, please bear with me if the issue below sounds familiar[1]. I've done some more experiments to find out why the firebird client library crashes on my box. If you're familiar with threads and dynamic linking, please read on. I'm trying to fix the libdbi driver for the firebird database engine on FreeBSD. libdbi is a C library that your application can link to. libdbi dlopen()s drivers which are shared objects themselves and provide the specific code to talk to the database client libraries. libdbi works across several platforms and supports a variety of database engines - with the notable exception of firebird on FreeBSD. I can create and access firebird databases using the isql command line tool. I can build and run a simple test program which creates a firebird database, opens it, and closes it again. Now, if I use pretty much the same code, compile it into a shared object, and dlopen() it, the firebird client library invariably crashes with the following gdb output: #0 0x28535b36 in ThreadData::restoreSpecific() from /usr/local/lib/libfbclient.so.2 #1 0x2852ceb4 in return_success () from /usr/local/lib/libfbclient.so.2 #2 0x28525179 in REM_attach_database () from /usr/local/lib/libfbclient.so.2 #3 0x2851386a in isc_attach_database () from /usr/local/lib/libfbclient.so.2 #4 0x284ece8e in _dbd_real_connect () from /usr/local/lib/dbd/libdbdfirebird.so #5 0x284eba19 in dbd_connect () from /usr/local/lib/dbd/libdbdfirebird.so #6 0x2808011d in dbi_conn_connect () from /usr/local/lib/libdbi.so.0 #7 0x0804982a in main (argc=1, argv=0xbfbfea2c) at test_dbi.c:91 My (limited) analysis makes me think this is some sort of a threading issue aggravated by the fact that the code is dlopen()ed (remember the same code works ok if compiled into a standalone app). BTW the firebird client library is the only library supported by libdbi which uses threads. All other drivers do not use threads and work ok. The firebird driver works ok on a variety of other platforms, including Linux and as weird ones as Windows. It somehow hurts my pride that it fails on FreeBSD. Does anyone out there have an idea how to fix this odd problem? System is 6.1-RELEASE, firebird2-client 2.0.3_1 was built as a port, libdbi and libdbi-drivers (both current cvs versions) were built from the sources (not as ports). regards, Markus [1] http://lists.freebsd.org/pipermail/freebsd-questions/2007-December/164571.html -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: threading and dlopen()
Heiko Wundram (Beenic) writes: Have you tried compiling your program with gcc -fpic -pthread ... ? I don't have any more insight into this problem, at least as I'm not using dbi and as such am not able to reproduce it, but I'd guess that if your program doesn't conform to the platform's required thread semantics (which are turned on by -fpic -pthread) but uses code that does require this, you're Thanks for the hint, but that didn't help. I've changed the appropriate acinclude.m4, Makefile.am, and configure.in files of both libdbi and libdbi-drivers to make sure that: - libdbi.so (the lib that dlopen()s the drivers) is built with -fpic -pthread - libdbifirebird.so (the firebird driver) is built with -fpic -pthread - test_dbi (linked against libdbi.so) is built with -fpic -pthread Still no luck, and the test app crashes at the very same point. Any other suggestions? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
firebird2 API woes
[apologies for cross-posting, but freebsd-database was not able to help] Hi, I'm having a hard time connecting to a firebird 2.0.3 server installed on FreeBSD 6.1 through the C API calls. The connection code is pretty much copy+pasted from the API documentation. However, all calls to isc_attach_database() cause the client library to segfault: #1 0x2852ceb4 in return_success () from /usr/local/lib/libfbclient.so.2 #2 0x28525179 in REM_attach_database () from /usr/local/lib/libfbclient.so.2 #3 0x2851386a in isc_attach_database () from /usr/local/lib/libfbclient.so.2 #4 0x284ece8e in _dbd_real_connect () from /usr/local/lib/dbd/libdbdfirebird.so #5 0x284eba19 in dbd_connect () from /usr/local/lib/dbd/libdbdfirebird.so #6 0x2808011d in dbi_conn_connect () from /usr/local/lib/libdbi.so.0 #7 0x0804982a in main (argc=1, argv=0xbfbfea2c) at test_dbi.c:91 The database I'm trying to attach to does exist, and I can attach to it using the isql command line utility. Is this a known issue? I have to admit that I'm not too familiar with firebird. All I'm trying to do is to upgrade the firebird driver of the libdbi-drivers project (http://libdbi-drivers.sourceforge.net, also available as a FreeBSD port). I'd greatly appreciate if someone more experienced in programming for firebird could give me a hint. I'll be happy to provide all sorts of debug information if needed. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: upgrading xfce4 fails
John Nielsen [EMAIL PROTECTED] was heard to say: There's a PR open for this with a working patch: http://www.freebsd.org/cgi/query-pr.cgi?pr=ports/108322 Download it and apply it in the ports/sysutils/xfce4-battery-plugin directory then rebuild the port. It fixed the problem for me. I'm sure it will be committed shortly. Thanks for the hint, I'll just wait until it is committed. And thanks to everyone else who helped me to figure out what went wrong. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
upgrading xfce4 fails
Hi all, I'm having a hard time upgrading xfce4 on a FreeBSD 6.1 box. I've tried portupgrade first, but as it bumped into a boatload of errors I eventually had to resort to upgrade individual packages. This left me with a mix of packages of xfce versions 4.2 and 4.4, and a build failure in xfce4-mcs-manager which I'm unable to resolve: /usr/local/lib/libxfcegui4.so: undefined reference to `xfce_desktop_entry_has_translated_entry' All relevant xfce4 libraries which are listed as a dependency of this package updated ok: [EMAIL PROTECTED]:/var/db/pkg# ls libxf* libxfce4gui-4.4.0: libxfce4mcs-4.4.0: libxfce4util-4.4.0: The other xfce4 packages look like this (some not upgraded due to their dependency on xfce4-mcs-manager): [EMAIL PROTECTED]:/var/db/pkg# ls xfce4* xfce4-appfinder-4.2.3: xfce4-artwork-0.0.4_5: xfce4-battery-plugin-0.2.0_8: xfce4-calendar-4.2.3_1: xfce4-clipman-plugin-0.4.1_3: xfce4-cpugraph-plugin-0.2.2_4: xfce4-desktop-4.2.3_1: xfce4-fm-4.2.3_1: xfce4-genmon-plugin-1.1_2: xfce4-icon-theme-4.4.0: xfce4-iconbox-4.2.3_1: xfce4-mcs-manager-4.2.3: xfce4-mcs-plugins-4.2.3_1: xfce4-mixer-4.2.3_1: xfce4-netload-plugin-0.3.2_1: xfce4-notes-plugin-0.9.7_7: xfce4-panel-4.2.3_1: xfce4-print-4.2.3_1: xfce4-session-4.2.3_1: xfce4-systemload-plugin-0.3.4_5: xfce4-systray-4.2.3_1: xfce4-toys-4.2.3_1: xfce4-trigger-launcher-4.2.3_1: xfce4-utils-4.2.3_1: xfce4-wm-4.2.3.2_1: xfce4-wm-themes-4.4.0: xfce4-xkb-plugin-0.3.3: All non-xfce4 packages which the xfce4 packages depend upon appear to have upgraded ok (or so portupgrade says). Is there anything I can do except waiting for prebuilt packages? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: upgrading xfce4 fails
John Nielsen writes: I second Warren's endorsement of the upgrade. :) Now that I managed to get it up and running, I fully agree with both of you :-) BTW turns out I had an old version of libxfcegui4.so in /usr/X11R6/lib which screwed up the build. After removing this library with its associated .a and .la files, and rebuilding some ports that use this library, the update finally worked ok. Even transparency and drop shadows seem to work ok with the i810 driver. There's one thing broken though: After the upgrade, panel plugins fail to display correct values. The battery monitor claims the battery of my laptop is at 0%. The network monitor plugin also sees the traffic at 0 kbyte/s no matter what I do. The volume plugin displays a value according to where I click in the plugin, but this does not translate to a higher or lower volume. Inversely, if I set the volume using xfce4-mixer, the value is not correctly displayed by the plugin. Did you notice similar problems? Did I miss to upgrade a package that all plugins rely on? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: upgrading xfce4 fails
Markus Hoenicka writes: There's one thing broken though: After the upgrade, panel plugins fail to display correct values. The battery monitor claims the battery of my laptop is at 0%. The network monitor plugin also sees the traffic at 0 kbyte/s no matter what I do. The volume plugin displays a value according to where I click in the plugin, but this does not translate to a higher or lower volume. Inversely, if I set the volume using xfce4-mixer, the value is not correctly displayed by the plugin. Did you notice similar problems? Did I miss to upgrade a package that all plugins rely on? Oops... The network monitor and the volume control plugins just required the network interfaces and device settings, respectively, and now work ok. Still, the battery monitor does not. Any clues? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
firefox starts up always in the first virtual screen in XFCE
Hi all, I don't know whether this qualifies as a bug. In any case it is only a mild annoyance, but if anyone knows how to fix it, please let me know. I upgraded to FreeBSD 6.1 from 5.4 with all ports updated to the latest about one week ago. My desktop is XFCE, which is configured with 6 virtual screens. If you start an X program, its window will pop up in the screen that was active at that time. This works for all programs except for firefox. The Firefox window will always pop up in the first virtual screen. I can then move it wherever I desire, so that's just an inconvenience, but its a clear regression to the behaviour before I upgraded to 6.1. Any clues? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: firefox starts up always in the first virtual screen in XFCE
Hi, your comment rang a bell. The Firefox window was apparently maximized when I upgraded the box. Firefox remembers the previous window size, and it seems like the (updated) XFCE window manager now thinks that the (updated) Firefox window is too big, and displays it on virtual screen 1 as a fallback strategy. I resized the window a little, closed Firefox, and opened it again - now it will open on the very screen which happens to be active. Problem solved. Poster embarrassed. Thanks a lot. regards, Markus Norberto Meijome writes: sounds like a session manager saved firefox on this desktop... not sure which one would do it when you start it up (usually they restore programs on startup...) -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
ppp works only as root
Hi, somehow I've managed to lock myself out of the ppp tool. I had it set up on my 5.3 box so that I could dial out as a regular user (I've added myself to the dialer group and added a allow users markus line to ppp.conf). This worked nicely until yesterday. Needless to say, I can't remember doing anything that might affect this peaceful state. I did install a few ports, but these were related to text processing, not to networking. Regardless, if I try to start ppp today from my user account, I'll never see the ppp command prompt. Instead, ppp eats all my cpu cycles by forking (the process list is full of ppp instances) until I kill ppp. ppp still works if I run it as root. However, I get this warning: Warning: ff02:4::/32: Change route failed: errno: Network is unreachable which I didn't get before either. As you can see, I can connect to the internet regardless, but I'd like to know how I managed to screw this up. Has anyone seen this before? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: mysql won't start on boot
Vonleigh Simmons [EMAIL PROTECTED] was heard to say: Nope, still no dice. Changed it back to mysql_enable but it still won't start up on boot. Anyone have any other ideas of what it could be? I can't help you any further if you want to understand what's going wrong. However, if you just want mysqld to show up in your process list, swallow your pride and install the package. Works for me. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: mysql won't start on boot
Vonleigh Simmons [EMAIL PROTECTED] was heard to say: After updating to 5.3 MySQL won't start at boot time. I read UPDATING and it says to include a line in /etc/rc.conf. But even after adding that line: # grep mysql /etc/rc.conf mysql_enable=YES It still won't start up on its own after I reboot. However doing: I'm sitting in front of a Windows box right now, so please excuse me that I reply from memory. AFAIK there is an example start script in /usr/local/etc/init.d/ which you have to rename to mysql.sh. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: mysql won't start on boot
Vonleigh Simmons writes: /usr/local/etc/init.d/ which you have to rename to mysql.sh. I do have a file in rc.d: Ah, sure. init.d is linuxish. It just so happened that I installed a 5.3 box from scratch in the last couple of days. MySQL wasn't installed yet, so I grabbed the packages mysql-server-4.1.5.tbz and mysql-client-4.1.5.tbz. After adding them, there were the following start scripts in /usr/local/etc/rc.d: 000.mysql-client.sh mysql-server.sh I did not edit any config file but went right ahead to reboot the box. mysqld came up as intended. Do you use ports or a different MySQL version, by any chance? What happens if you run /usr/local/etc/rc.d/mysql-server.sh start manually? Does the log file enlighten you (/var/db/mysql/[boxname].err)? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
configuration of XFree86
Leon writes: Hi, I'm first time installing FreeBSD. I'm trying to configure X server. And I have a problem with configuring of X server. I use the instructions for installation and configuration. Which says: use /stand/sysinstall Take Configure Then pick XFree86 Newer versions of FreeBSD (and Linux, and others) use X.Org instead of XFree86. Use that. If you start from scratch, use the X developer (or any other X) distro, and everything you'll need will be installed. Once done with sysinstall, type xorgconfig to run the X.Org setup tool (all from your root account). In your user account, create a suitable .xinitrc file in your home directory. One line like: startkde (if you install KDE) or afterstep (if you like it simple) should be enought. Then type startx to fire up the X Window system. The problem is that I can't find XFree86; it's not there. So, how can I configure X server? Can you send me a link,or instruction how can I do it? And the second question is the funny one. How can I see help on the FreeBSD? I have tried MAIN but it did not work. What command should I use? Help is spelled man on Unix (and yes, it's case-sensitive). Type man man to learn about it. regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: mysql won't start on boot
Joshua Lokken [EMAIL PROTECTED] was heard to say: On Wed, 8 Dec 2004 12:05:13 -0800, Vonleigh Simmons [EMAIL PROTECTED] wrote: What happens if you run /usr/local/etc/rc.d/mysql-server.sh start # /usr/local/etc/rc.d/mysql-server.sh start Starting mysql. Yet mysql never started up, here is the error log: --- 041208 14:07:26 mysqld started 041208 14:07:26 [ERROR] Can't find messagefile '/usr/local/share/mysql/english/errmsg.sys' 041208 14:07:26 [ERROR] Aborting 041208 14:07:26 mysqld ended --- The documentation specifies that the above is not the way to start mysql. That's interesting. Why on earth would the mysql package use exactly this way to start up the server (see my previous post) if it is all wrong? regards, Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: vinum raid5: newfs throws an error
Greg 'groggy' Lehey writes: There was once an error in the stripe size calculations that meant that there were holes in the plexes. Maybe it's still there (old Vinum is not being maintained). But you should have seen that in the console messages at create time. Vinum reports the disk sizes as 17500MB (da1) and 17359MB (da2, da3). The raid5 volume and plex have a size of 33GB. This looks like the kind of scenario where that could happen. Try this: 1. First, find a better stripe size. It shouldn't be a power of 2, but it should be a multiple of 16 kB. I'd recommend 496 kB. This won't fix the problem, but it's something you should do anyway 2. Calculate the length of an exact number of stripes, and create the subdisks in that length. Try again and see what happens. 3. Use gvinum instead of vinum and try both ways. Ok, I decreased the stripe size to 496, regardless of whether it has anything to do with my problem. Next I set the subdisk length to 17359m on all disks, and things started to work ok. No more newfs errors here. Before doing this I also had a brief encounter with gvinum. There is no manpage in 5.3BETA7, so I assumed it groks the same config files as vinum. However, this did not do me any good as it simply rebooted the box. I guess gvinum works better in RELEASE. Thanks a lot for your help. Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
vinum limits disk size to 255MB
Hi all, I'm trying to set up vinum on a freshly installed FreeBSD 5.3-BETA7 box. The system is installed on da0. I want to use three 18G SCSI drives to create a vinum volume. For some reason vinum believes the disks hold a mere 255MB. This is what vinum sets the subdisk length if I specify 0m as the length (meaning use all available space according to the manual). If I specify any larger size manually, vinum complains that there is No space left on device which strikes me odd. I *believe* that the fdisk and bsdlabel outputs reprinted below show that the disks are indeed 18G. According to my math, the appropriate size to specify in the vinum configuration would be 35566215s, that is the number of sectors of the smaller disks minus 265. I see only one minor problem: The disks are not identical, with da1 having a slightly larger capacity than da2 and da3. Can anyone throw me a ring here? Thanks, Markus fdisk da1 shows: *** Working on device /dev/da1 *** parameters extracted from in-core disklabel are: cylinders=2231 heads=255 sectors/track=63 (16065 blks/cyl) Figures below won't work with BIOS for partitions not in cyl 1 parameters to be used for BIOS calculations are: cylinders=2231 heads=255 sectors/track=63 (16065 blks/cyl) Media sector size is 512 Warning: BIOS sector numbering starts with sector 1 Information from DOS bootblock is: The data for partition 1 is: sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD) start 63, size 35840952 (17500 Meg), flag 80 (active) beg: cyl 0/ head 1/ sector 1; end: cyl 182/ head 254/ sector 63 The data for partition 2 is: UNUSED The data for partition 3 is: UNUSED The data for partition 4 is: UNUSED fdisk da2 or da3 show: *** Working on device /dev/da2 *** parameters extracted from in-core disklabel are: cylinders=2213 heads=255 sectors/track=63 (16065 blks/cyl) Figures below won't work with BIOS for partitions not in cyl 1 parameters to be used for BIOS calculations are: cylinders=2213 heads=255 sectors/track=63 (16065 blks/cyl) Media sector size is 512 Warning: BIOS sector numbering starts with sector 1 Information from DOS bootblock is: The data for partition 1 is: sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD) start 63, size 35551782 (17359 Meg), flag 80 (active) beg: cyl 0/ head 1/ sector 1; end: cyl 1023/ head 254/ sector 63 The data for partition 2 is: UNUSED The data for partition 3 is: UNUSED The data for partition 4 is: UNUSED bsdlabel -A da1 shows: # /dev/da1: type: unknown disk: amnesiac label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 2231 sectors/unit: 35843670 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 358436700unused0 0 # raw part, don't edit e: 358436700 vinum bsdlabel -A da2 or da3 show: # /dev/da2: type: unknown disk: amnesiac label: flags: bytes/sector: 512 sectors/track: 63 tracks/cylinder: 255 sectors/cylinder: 16065 cylinders: 2213 sectors/unit: 35566480 rpm: 3600 interleave: 1 trackskew: 0 cylinderskew: 0 headswitch: 0 # milliseconds track-to-track seek: 0 # milliseconds drivedata: 0 8 partitions: #size offsetfstype [fsize bsize bps/cpg] c: 355664800unused0 0 # raw part, don't edit e: 355664800 vinum The vinum config file reads: drive ibma device /dev/da1s1e drive ibmb device /dev/da2s1e drive ibmc device /dev/da3s1e volume raid5 setupstate plex org concat sd length 0m drive ibma sd length 0m drive ibmb sd length 0m drive ibmc vinum - create vinumconfig1 3 drives: D ibma State: up /dev/da1s1e A:0/256 MB (0%) D ibmb State: up /dev/da2s1e A:0/256 MB (0%) D ibmc State: up /dev/da3s1e A:0/256 MB (0%) 1 volumes: V raid5 State: up Plexes: 1 size: 767 MB 3 subdisks: S raid5.p0.s0 State: up D: ibma Size: 255MB S raid5.p0.s1 State: up D: ibmb Size: 255MB S raid5.p0.s2 State: up D: ibmc Size: 255MB uname -a shows: FreeBSD wutz.mininet 5.3-BETA7 FreeBSD 5.3-BETA7 #0: Sat Oct 2 21:01:00 UTC 2004 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC i386 -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: vinum limits disk size to 255MB
Greg 'groggy' Lehey writes: You don't say whether you're using vinum or gvinum. I've never seen this problem before, but if you're getting incorrect subdisk sizes, try specifying them explicitly: sd length 35840952s drive ibma I wonder whether the problem is related to specifying the size as 0m instead of 0. It shouldn't be. Actually I've never heard of gvinum, so I'm pretty sure we're looking at vinum here. Looks like this is an odd pilot error. If I label disks from within sysinstall and on the command line, one tool apparently doesn't know what the other is doing. This must be a gross misunderstanding on my side of how these tools work. Using bsdlabel to set the partition sizes apparently is not sufficient. To make a long story short, I went back to sysinstall, created a partition using the full disk, then went to bsdlabel -e to turn the partition into type vinum. Now I get the full capacity. I still can't make a raid5 out of these drives. I'm still fiddling and will get back to the list if I can't figure this out myself. Thanks anyway for your prompt answer. Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
vinum raid5: newfs throws an error
Hi all, now that I can use the full capacity of my disks, I'm stuck again. I'm trying to set up a raid5 from three SCSI disks (I know that a serious raid5 should use five disks or more, but I have to make do with three at the moment). The configuration is as follows: drive ibma device /dev/da1s1e drive ibmb device /dev/da2s1e drive ibmc device /dev/da3s1e volume raid5 setupstate plex org raid5 512k sd length 0m drive ibma sd length 0m drive ibmb sd length 0m drive ibmc This works ok. Then I run vinum init to initialize the drives. Trying to create a filesystem on this construct results in the error message: newfs: wtfs: 65536 bytes at sector 71130688: Input/output error Is that trying to tell me that my calculation of the group size is incorrect? Does it have to do anything with the fact that the three disks have slightly different capacities? Vinum reports the disk sizes as 17500MB (da1) and 17359MB (da2, da3). The raid5 volume and plex have a size of 33GB. BTW creating a concatenated volume on the same disks works ok, newfs does not throw an error here. Any help is appreciated. Markus -- Markus Hoenicka [EMAIL PROTECTED] (Spam-protected email: replace the quadrupeds with mhoenicka) http://www.mhoenicka.de ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
