With the 2.30.X gnome branch: When I initialize gnome from GDM, I get only the background image of gentoo (my gnome's background) and (if I'm lucky) the desktop (not every time). The panel doesn't appear, nor can I initialize it using "gnome-panel" from the shell. Other D.E. or W.M. work just fine (although don't know why gnome-panel gets into xfce as well). With the (masked) 2.32.0 gnome branch: When I initialize gnome from GDM, I get only the background and a recurring popup (say about 20 times in 10 seconds) with a writing "Panel" flashing in the top left corner of the screen. If I do not stop this by rebooting GDM (/etc/init.d/xdm restart) the computer freezes and I have to do a hardware reset. Reproducible: Always Steps to Reproduce: 1. Install gnome 2.30 branch (full set) and apply all recent updates 2. Login as a user from GDM to session "Gnome" 3. Can't get a stable panel (or, if there is, once you click it it gets unusable) OR: 1. Install the masked full 2.32 gnome branch and apply all recent updates 2. Login as a user from GDM to session "Gnome" 3. Only background appears. Can't get a terminal or a panel, not the desktop either. A popup flashing the screen appears (small popup with a writing "Panel"), about square shape. Actual Results: Gnome session and / or gnome panel don't work Expected Results: Gnome session should work fine and gnome panel should appear and be click-able
Created attachment 255585 [details] Processes running
Created attachment 255587 [details] Dmesg output
Tasks: 196 total, 3 running, 193 sleeping, 0 stopped, 0 zombie Cpu(s): 21.9%us, 3.8%sy, 0.0%ni, 73.4%id, 0.8%wa, 0.0%hi, 0.1%si, 0.0%st Mem: 8180592k total, 2027136k used, 6153456k free, 184448k buffers Swap: 8787548k total, 0k used, 8787548k free, 412408k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 4844 messageb 20 0 20144 2008 760 R 41 0.0 1:25.44 dbus-daemon 6453 root 20 0 43136 3320 2492 S 25 0.0 0:50.77 udisks-daemon dbus-daemon and udisks-daemon tends to use high cpu resources (not sure if this is really the problem). Furthermore actions performed in other D.E. (xfce or lxde) tends to make the computer crash or freeze.
gentoo stefano # emerge --info Portage 2.1.9.24 (default/linux/amd64/10.0/desktop, gcc-4.4.4, glibc-2.11.2-r3, 2.6.36-gentoo-r1 x86_64) ================================================================= System uname: Linux-2.6.36-gentoo-r1-x86_64-Intel-R-_Core-TM-_i7_CPU_870_@_2.93GHz-with-gentoo-1.12.14 Timestamp of tree: Sat, 27 Nov 2010 13:00:01 +0000 app-shells/bash: 4.1_p7 dev-java/java-config: 2.1.11-r1 dev-lang/python: 2.6.5-r3, 3.1.2-r4 dev-util/cmake: 2.8.1-r2 sys-apps/baselayout: 1.12.14-r1 sys-apps/sandbox: 2.3-r1 sys-devel/autoconf: 2.13, 2.65-r1 sys-devel/automake: 1.8.5-r4, 1.9.6-r3, 1.10.3, 1.11.1 sys-devel/binutils: 2.20.1-r1 sys-devel/gcc: 4.4.4-r2 sys-devel/gcc-config: 1.4.1 sys-devel/libtool: 2.2.10 sys-devel/make: 3.81-r2 virtual/os-headers: 2.6.30-r1 (sys-kernel/linux-headers) ACCEPT_KEYWORDS="amd64" ACCEPT_LICENSE="* -@EULA" CBUILD="x86_64-pc-linux-gnu" CFLAGS="-march=core2 -mtune=generic -O2 -pipe" CHOST="x86_64-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/share/X11/xkb /usr/share/config /usr/share/openvpn/easy-rsa /var/lib/hsqldb" CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/env.d /etc/env.d/java/ /etc/fonts/fonts.conf /etc/gconf /etc/php/apache2-php5/ext-active/ /etc/php/cgi-php5/ext-active/ /etc/php/cli-php5/ext-active/ /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo /etc/texmf/language.dat.d /etc/texmf/language.def.d /etc/texmf/updmap.d /etc/texmf/web2c" CXXFLAGS="-march=core2 -mtune=generic -O2 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="assume-digests binpkg-logs distlocks fixlafiles fixpackages news parallel-fetch protect-owned sandbox sfperms strict unknown-features-warn unmerge-logs unmerge-orphans userfetch" GENTOO_MIRRORS="http://mirror.switch.ch/ftp/mirror/gentoo/" LANG="it_CH.utf8" LDFLAGS="-Wl,-O1 -Wl,--as-needed" LINGUAS="it en" MAKEOPTS="-j9" PKGDIR="/usr/portage/packages" PORTAGE_CONFIGROOT="/" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --stats --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="X a52 aac acl acpi alsa amd64 apache2 bash-completion berkdb bluetooth branding bzip2 cairo cdr cli consolekit cracklib crypt css curl cxx dbus dri dts dvd dvdr emacs emboss encode exif extras fam firefox flac fortran gdbm gif gimp gnome gpm gtk gzip hal hddtemp iconv imagemagick ipv6 java jpeg kde lame lcms ldap libnotify lm_sensors mad mikmod mmx mng modules mono mp3 mp4 mpeg mudflap multilib mysql ncurses nls nptl nptlonly ogg opengl openmp pam pango pcre pdf perl php png ppds pppd python qt3support qt4 raw readline samba scanner sdl server session spell sql sqlite sqlite3 sse sse2 ssl startup-notification svg sysfs tcpd tiff truetype unicode usb vorbis webkit wifi x264 xcb xml xorg xulrunner xv xvid zlib" ALSA_CARDS="ali5451 als4000 atiixp atiixp-modem bt87x ca0106 cmipci emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel intel8x0 intel8x0m maestro3 trident usb-audio via82xx via82xx-modem ymfpci" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mmap_emul mulaw multi null plug rate route share shm softvol" APACHE2_MODULES="actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache cgi cgid dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias" COLLECTD_PLUGINS="df interface irq load memory rrdtool swap syslog" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm earthmate evermore fv18 garmin garmintxt gpsclock itrax mtk3301 nmea ntrip navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf superstar2 timing tsip tripmate tnt ubx" INPUT_DEVICES="keyboard mouse evdev" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LINGUAS="it en" PHP_TARGETS="php5-2" RUBY_TARGETS="ruby18" USERLAND="GNU" VIDEO_CARDS="fbdev glint intel mach64 mga neomagic nouveau nv r128 radeon savage sis tdfx trident vesa via vmware voodoo" XTABLES_ADDONS="quota2 psd pknock lscan length2 ipv4options ipset ipp2p iface geoip fuzzy condition tee tarpit sysrq steal rawnat logmark ipmark dhcpmac delude chaos account" Unset: CPPFLAGS, CTARGET, EMERGE_DEFAULT_OPTS, FFLAGS, INSTALL_MASK, LC_ALL, PORTAGE_BUNZIP2_COMMAND, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS
Created attachment 255589 [details] Kernel's messages log Might be related to the linux kernel as well, because it freezes with other windows managers as well. Another option might be the nvidia drivers. I attach here a portion (last 2 days) of /var/log/messages.txt as well. I hope you can help me.
Please try on a new created user account and, if still valid, attach ~/.xsession-errors file just after reproducing the problem and before restarting the session.
Created attachment 255617 [details] ~/.xsession_errors from a new created user Thank you for your answer Pacho Ramos. Here is the file you requested me to send here. I created a "test" account, tried to login from GDM to Gnome (gnome 2.32) and - as here the news is that I see the icons on the desktop - I cannot see the panel, neither click anything (this writing "Panel" which is a box of about 200 x 200 (or more) pixels. New symptom is the frequent crash of xfce / lxde when attempting to restore (in the sense of minimize -> restore) a firefox instance or a file manager (in this case thunar) instance. Thank you very much. I even tried to update my kernel but seems like it was not related (or maybe it is, simply the new kernel didn't fix the bug (gentoo-sources-2.6.36-r3).
Probably related with that: GVFS-RemoteVolumeMonitor-WARNING **: invoking List() failed for type GProxyVolumeMonitorGdu: org.freedesktop.DBus.Error.NoReply: Message did not receive a reply (timeout by message bus) errors, but I don't know how to solve them just now :-(, I have searched in google for them and looks like other distributions are also affected: http://www.google.com/search?sourceid=chrome&ie=UTF-8&q=GVFS-RemoteVolumeMonitor-WARNING+**:+invoking+List()+failed+for+type+GProxyVolumeMonitorGdu:+org.freedesktop.DBus.Error.NoReply:+Message+did+not+receive+a+reply+(timeout+by+message+bus)
Created attachment 255655 [details] ~/.xsession_errors from standard user / XFCE Mmm .. I'll dig into it soon. Just for more informations, I attach the .xsession-errors files of the standard user (the one I log in everyday) when from GDM I log in into XFCE (there is some instability as well, although it may not be inside the file [yet]). Edit: the instability happens [almost always] when listing files in a directory (in a graphical way, i.e. from a file manager like tunar / nautilus or from a browse files under firefox as well). That may have been recorded. Thank you !
Removed gvfs and now I can browse withouth any problems / crash / etc. Although in XFCE panels disappear, I have still to test if gnome is working (don't think so because the problem with gnome was the panel, other than the gvfs / browsing itself).
Your installation seems to be in a bad shape. > gnome-session[4809]: WARNING: Could not launch application '._cfg0000_gnome- > settings-daemon-helper.desktop': Unable to start application: Esecuzione del > processo figlio "/usr/libexec/gnome-session/helpers/gnome-settings-daemon- > helper" non riuscita (No such file or directory) according to this, you did not update your configuration files after updating your system. Please run dispatch-conf or etc-update and update your configuration. > GLib-GIO-ERROR **: Settings schema 'org.gnome.crypto.cache' is not installed > GLib-GIO-ERROR **: Settings schema 'org.gnome.Bluetooth' is not installed You installed masked gnome 2.32 packages, so this message is "normal". Please run the following command to resolve this problem: # glib-compile-schemas --allow-any-name /usr/share/glib-2.0/schemas > The program 'gnome-panel' received an X Window System error. > This probably reflects a bug in the program. > The error was 'BadMatch (invalid parameter attributes)'. > (Details: serial 923 error_code 8 request_code 7 minor_code 0) > (Note to programmers: normally, X errors are reported asynchronously; > that is, you will receive the error a while after causing it. > To debug your program, run it with the --sync command line > option to change this behavior. You can then get a meaningful > backtrace from your debugger if you break on the gdk_x_error() function.) This is really bad. The problem probably lies in your graphic card driver. Please make sure you driver is known to work with your kernel and xorg-server configuration. This is nothing gnome team can help you with so I'll CC nvidia-drivers maintainers.
*** Bug 348139 has been marked as a duplicate of this bug. ***
See bug 348240 for a (temporary) fix, patching gnome-panel to start on the first X screen only.
If this is really the same as bug 348240, I think this should be closed as a dupe of that one since, even being newer, it explains a bit where could be the problem on this and how we should handle it.
I don't think this is really the same as the bug you're discussing about, since here the problem occurs even with a single screen running. Furthermore now even XFCE doesn't show anything (no panels as gnome) and the computer is almost completely unusable (tried to update, but from gentoo the net is going at the speed of a 56K 1/100 times (and the rest of the time doesn't run at all)). I don't know if it's related, but what could be done to solve the problem ? Try to downgrade nvidia-drivers ?
(In reply to comment #15) > I don't think this is really the same as the bug you're discussing about, since > here the problem occurs even with a single screen running. > Furthermore now even XFCE doesn't show anything (no panels as gnome) and the > computer is almost completely unusable (tried to update, but from gentoo the > net is going at the speed of a 56K 1/100 times (and the rest of the time > doesn't run at all)). I don't know if it's related, but what could be done to > solve the problem ? Try to downgrade nvidia-drivers ? > OK. I think I solved it (the dirty way). I downgraded to gnome 2.30.X and nothing was running. Then I tried to re-upgrade as well. I updated all packages (there was a little problem with dbus in /usr/include/dbus-1.0/dbus/dbus-glib.h which I solved by symlink). Then nothing was running good, so I simply deleted all gnome configuration folders inside my home holder (.gconf , .gconfd , .gnome , .gnome2 , .gnome2_private , .config , .dbus , .gnome-color-chooser , .nautilus , .themes , .gnome-rdp.db) although some of them were probably not necessary. The I unmerged (again) the gvfs package and all seems to be fine. I cannot understand why each nautilus folder, browse button I click, panel item I click and so on make gnome / xfce / anything go dump if there is this gvfs installed. It is pulled in by force, but is it really necessary ? Thanks
It looks then to be a problem with an old configuration in your home directory (this bug is INVALID then). The next time, instead of removing all directories, I would suggest you to create a new user account and start copying your old configuration file to the new one until the problem reappears, knowing then what exact file is breaking things.
Knowing a few other GNOME distros, I should think it should upgrade the user/local config gracefully, so I don't agree with your bug resolution. It's still out in the open what exactly went wrong.
In that case, please tell us how to make a fresh configuration "break" and then, once we know the steps to reproduce the problem, we can report it to upstream.
(In reply to comment #19) > In that case, please tell us how to make a fresh configuration "break" and > then, once we know the steps to reproduce the problem, we can report it to > upstream. > Mmm ... basically I don't know what exactly was the problem. However it seems that during the upgrade / downgrade / update, some packages needed recompiling (in particular: metacity, gnome-session, gnome-panel and gvfs which needed each time unmerging). What I saw then was a *nice" gnome session which was running, desktop visible, could open a terminal doing right click -> open terminal and so on. However the panel was not visible and the window manager (metacity) did not appear. It sais something complaining about a theme which was missing and then said: "Assertion failed". I now have gnome 2.32 installed which works fine (although after each uptime I have to make sure that gvfs gets unmerged, otherwise it always freeze). I tried under another user profile and things went fine (even if with the same user the panel did NOT appear backwards, withouth any modification after the downgrade and another upgrade to gnome 2.32, it worked flawlessly). Still wondering why gnome panel wasn't running (it complianed it was already, but I couldn't see it). Is it still something related with metacity. Sorry if I caused some confusion: I can open windows, file manager and terminals, however they don't have a border. Thanks
Sorry I meant *after each update*. If you need some log files I can try to find them. Thanks
Created attachment 256981 [details] ~/.xsession_errors from standard user with gnome running Further symptom: the problems (with gvfs making the system hang) appears *only* after reboot (or login / logout). I don't really know what to do with it, because unfortunately I need it to mount a remote windows samba share. I don't know if the lofs I posted before say something about this. I updated the configuration files with dispatch-conf (with the new default files mostly) but that does not solve the problem. Any ideas where to find something else ? Thank you
Sorry, but I am a bit lost here, related with your last comment: - Is that obtained on a fully updated Gentoo system? - Does it occur on a fresh user account?
(In reply to comment #19) > In that case, please tell us how to make a fresh configuration "break" and > then, once we know the steps to reproduce the problem, we can report it to > upstream. > Ehm, I don't think your resolution to this bug is valid too. It makes no sense to me to create a new user to face the problem. It is clear that the upgrade to the latest gnome is not as smooth as it should be. If you not put effort to understand the source of the problem, you will keep having people CCing into this bug and post the same comments over and over again. The bug is definitely not invalid. If you are unable to find the root of the problem, it might be wise to add an ewarn/elog/ message and inform the users about possible breakages and which packages need to be rebuild. Please leave this bug open to avoid duplicates in the future
I think I've might narrowed it down to gnome-panel-(~)2.32.1. -Deleting all configuration files as stated at comment #16 from the common user and letting gnome create new ones didn't have any effect. Same with unmerging and re-emerging gnome-session-2.32.1, metacity-2.30.3, gvfs-1.6.6. Still same result as original poster's comment. > (comment #0) > With the (masked) 2.32.0 gnome branch: > When I initialize gnome from GDM, I get only the background and a recurring > popup (say about 20 times in 10 seconds) with a writing "Panel" flashing in the > top left corner of the screen. If I do not stop this by rebooting GDM > (/etc/init.d/xdm restart) the computer freezes and I have to do a hardware > reset. -Falling back to stable version gnome-panel-2.30.2 seems to correct the problem. Everything is working as it should. -Emerging gnome-panel-2.32.1 brings back the problem. Downgrading to 2.30.2, everything ok. -Bringing back, user configurations from backup, and still no problem. -Trying to update again to gnome-panel-2.32.1 and I'm having again the same problem. -As of now I'm with gnome-panel-2.30.2 and everything is ok. -gvfs-1.6.6 is also installed without problems.
This bug report is a mess. Could reporter provide a complete report of the situation on his box. I completely lost track of what is now ok and what is not.
(In reply to comment #24) > Ehm, I don't think your resolution to this bug is valid too. Maybe NEEDINFO would be better I guess > It makes no sense > to me to create a new user to face the problem. Creating a new user will allow reporter to see if it's a configuration problem or not. Then, he can try to start copying old config files to his new home directory to check when the problem reappears. Finally, we hope to be able to find what file is wrong and, then, we can report it to upstream to let them know the situation. The problem is that original reporter, if I don't misremember, simply removed his old configuration and, then, we cannot imagine what was wrong. >It is clear that the upgrade to > the latest gnome is not as smooth as it should be. No, it's clear that the upgrade to the latest gnome wasn't as smooth as it should be for *that specific configuration* (for example, I didn't suffered any problem like this on a mostly stable system, and I haven't seen so many duplicates with this same problem). > If you not put effort to > understand the source of the problem, you will keep having people CCing into > this bug and post the same comments over and over again. > I have put the effort to understand where could be the problem, I have also searched for similar problems in the web, but have failed to find the root cause and seeing it looks like a problem with some old config... > The bug is definitely not invalid. If you are unable to find the root of the > problem, it might be wise to add an ewarn/elog/ message and inform the users > about possible breakages and which packages need to be rebuild. Do we have any kind of "pattern" that could allow us to know what kind of systems are likely to break? We doesn't even know what packages are affected by this, what needs to be rebuilt and so. If you read above comments, you will see that there are information only from luckylinux and his setup. > Please leave > this bug open to avoid duplicates in the future > I won't close this for now, but it's obvious that more information (and probably a summary of the problem) is needed.
(In reply to comment #25) > I think I've might narrowed it down to gnome-panel-(~)2.32.1. > > -Deleting all configuration files as stated at comment #16 from the common user > and letting gnome create new ones didn't have any effect. Same with unmerging > and re-emerging gnome-session-2.32.1, metacity-2.30.3, gvfs-1.6.6. Still same > result as original poster's comment. > OK, then, attach your ~/.xsession-errors log file please :-)
As a side note, if you are suffering problems with gnome-panel and are using quick-longe-applet, you are probably hitting bug 348123
Created attachment 257041 [details] New user gives no more problems (In reply to comment #23) > Sorry, but I am a bit lost here, related with your last comment: > - Is that obtained on a fully updated Gentoo system? > - Does it occur on a fresh user account? > - Yes, the system be fully updated (at least as with emerge -uDn world - some packages have been un-keyworded) - No, it doesn't occours on a fresh user account anymore (before the dowgrade / re-upgrade procedure it was giving problems even on fresh accounts). Main account freezes, new account doesn't (tested 3 times login-logout: fresh account no problems, usual account freezes on login (no panel and really slow)).
(In reply to comment #26) > This bug report is a mess. Could reporter provide a complete report of the > situation on his box. I completely lost track of what is now ok and what is > not. > Sorry for the confusion. Summary: - At the beginning: after update of system (gnome 2.30.X -> 2.30.Y the gnome session was not stable anymore, i.e. I couldn't login) - Tried to update system once again -> XFCE / LXDE and all other window manager are definitively freezing due to an unknow reason beign GVFS. KDE works. - Tried to dowgrade to gnome-2.30.X (i ried to upgrade to gnome 2.32 but that made things worse) and then reupgrade (since now gnome 2.32 was marked as un-keyworded, not masked anymore) ===> NOW: all works OK, exceptly (if GVS is installed) everything freezes under the normal user: login takes a bunch of time, cannot open nautilus file browser, other file manager under XFCE and LXDE give also problems, ... although under a new created user (see my previous message) GVFS gives no problems. Deleting all gnome's configuration for the usual user did not solve the GVFS problem, although it solved others
>> The problem is that original reporter, if I don't misremember, simply removed his old configuration and, then, we cannot imagine what was wrong. Sorry but the fact there were many problems probably confonded you all. Problems appeared: 1) (Original problem) Gnome session fails to initialize - upgrade to 2.32 did not solve the problem, downgrade and then re-upgrade to 2.32 helped me identify the problem (which I still believe is GVFS). Now gnome session initializes (although I cannot say for certain that this is the same as no. 2, because the session was so slow that after 5 minutes I restarted X - gnome was not responding - only background shown and gnome panel popup blinking - see one of my firsts posts) 2) The GVFS problem affect indifferently gnome, xfce, lxde (maybe a gtk+ issue ? KDE seems to be unaffected). At session start of gnome, it freezes, causing panel to not appear at all and session interaction is slow. File manager, when opened, freezes for some minutes as well. THIS PROBLEM HAS NOT BEEN SOLVED YET 3) Metacity does not load up anymore, gnome-panel does not work (due to the fact metacity was looking for a theme which was not there anymore - DELETING CONFIGURATION SOLVED THIS)
(In reply to comment #30) > Created an attachment (id=257041) [details] > New user gives no more problems > > (In reply to comment #23) > > Sorry, but I am a bit lost here, related with your last comment: > > - Is that obtained on a fully updated Gentoo system? > > - Does it occur on a fresh user account? > > > - Yes, the system be fully updated (at least as with emerge -uDn world - some > packages have been un-keyworded) > - No, it doesn't occours on a fresh user account anymore (before the dowgrade / > re-upgrade procedure it was giving problems even on fresh accounts). Main > account freezes, new account doesn't (tested 3 times login-logout: fresh > account no problems, usual account freezes on login (no panel and really > slow)). > OK, then, I think that the procedure would be the following: 1. Have a fully updated system (looks like it's the current situation). 2. Verify you didn't forget to run any revdep-rebuild --library command (check /var/log/portage/elog/summary.log messages). 3. Have a new created user account -> things should work ok with it. 4. Start copying configuration files from your old home to your new one until problem reappears (probably it's something into .gconf). Once it reappears, try to go through subdirectories to try to find the offending file.
(In reply to comment #33) > (In reply to comment #30) > > Created an attachment (id=257041) [details] [details] > > New user gives no more problems > > > > (In reply to comment #23) > > > Sorry, but I am a bit lost here, related with your last comment: > > > - Is that obtained on a fully updated Gentoo system? > > > - Does it occur on a fresh user account? > > > > > - Yes, the system be fully updated (at least as with emerge -uDn world - some > > packages have been un-keyworded) > > - No, it doesn't occours on a fresh user account anymore (before the dowgrade / > > re-upgrade procedure it was giving problems even on fresh accounts). Main > > account freezes, new account doesn't (tested 3 times login-logout: fresh > > account no problems, usual account freezes on login (no panel and really > > slow)). > > > > OK, then, I think that the procedure would be the following: > 1. Have a fully updated system (looks like it's the current situation). > 2. Verify you didn't forget to run any revdep-rebuild --library command (check > /var/log/portage/elog/summary.log messages). > 3. Have a new created user account -> things should work ok with it. > 4. Start copying configuration files from your old home to your new one until > problem reappears (probably it's something into .gconf). Once it reappears, try > to go through subdirectories to try to find the offending file. > But the problem is: in post #16 I said I deleted the .gconf, .gnome, .gnome2, .gconfd directories to solve the metacity-thüme problem, so if that was the source of the problem shouldn't it have fixed after deleting the directory ? In fact as soon after it has merged gvfs, under the main user it freezes (instantly) - have to press ctrl + alt + F1, because the udisks-daemon and dbus-daemon processes are going crazy in cpu usage. That did not / does not / is not happening with the newly created user. Now I ask myself ... do I have to delete my documents folder to solve this problem ?? There are not many directories which were left and .gvfs was erased as well at that time.
Created attachment 257113 [details] emerge --info
Created attachment 257115 [details] .xsession-errors, new user, clean configs
Just to recap and help clear the situation following my comment #25 Fully updated system (uDN world). Run all revdep-rebuild --library that needed, as stated in the logs. Created a new test user. When I log into Gnome with I get this > (comment #0) > When I initialize gnome from GDM, I get only the background and a recurring > popup (say about 20 times in 10 seconds) with a writing "Panel" flashing in the > top left corner of the screen. If I do not stop this by rebooting GDM > (/etc/init.d/xdm restart) the computer freezes and I have to do a hardware > reset. After extensive searching and testing (unmerge-emerge various packages), I found that if I downgrade to gnome-panel-2.30.2 everything is working as it should, no problems at all, neither with the a new user nor with my existing one that I already had with all my old configuration files in it. If I upgrade to 2.32.1 I have the same problem I stated before, with both the new user account and my old existing one. (In reply to comment #29) > As a side note, if you are suffering problems with gnome-panel and are using > quick-longe-applet, you are probably hitting bug 348123 > No I don't have quick-lounge-applet package installed. Concerning original's poster conclusion for gvfs causing the problem, I cannot reproduce it or confirm it, as I have the latest gvfs package and: 1. With 2.32.1 it doesn't make any difference to have it installed or not. Problem persists. 2. With 2.30.2 it doesn't make any difference again whether it is installed or not. Gnome loads and it's stable. I have attached my "emerge --info" and ".xsession-errors" from the new user account. Let me know if you need anything else.
@Markellos, your issue isn't related to this bug, it seems you have a nvidia graphic card and also this problem has already been reported to us, there is not much we can do since the driver is probably at fault. You can search bugzilla for duplicates, the final word was that you should bring this issue upstream (either gtk+ or nvidia).
(In reply to comment #34) > But the problem is: in post #16 I said I deleted the .gconf, .gnome, .gnome2, > .gconfd directories to solve the metacity-thüme problem, so if that was the > source of the problem shouldn't it have fixed after deleting the directory ? > In fact as soon after it has merged gvfs, under the main user it freezes > (instantly) - have to press ctrl + alt + F1, because the udisks-daemon and > dbus-daemon processes are going crazy in cpu usage. That did not / does not / > is not happening with the newly created user. Please be a bit more patient and do the copying of old config files to the new home with ALL directories (.local or .config and maybe other directories or files could be involved on that gvfs issue) > Now I ask myself ... do I have to delete my documents folder to solve this > problem ?? There are not many directories which were left and .gvfs was erased > as well at that time. > I doubt documents are related but remember to copy them to the new (and working) home directory instead of removing them.
@luckylinux: from the two xsession-errors you pasted, it seems clear there is a problem with your dbus installation. In the log from your regular user (broken), you can see that a couple of gnome application try to connect to gvfs to get a list of drives but fails due to timeout (policy issue ?), then it tries to get that list in a loop, forever and this could cause your high CPU load with dbus + high latency due to applications waiting for dbus replies that never comes. I suggest rebuilding dbus and dbus-glib (please also paste emerge -pv of both) and eventually rebuild all reverse dependencies of dbus-glib.
(In reply to comment #40) > @luckylinux: from the two xsession-errors you pasted, it seems clear there is a > problem with your dbus installation. > > In the log from your regular user (broken), you can see that a couple of gnome > application try to connect to gvfs to get a list of drives but fails due to > timeout (policy issue ?), then it tries to get that list in a loop, forever and > this could cause your high CPU load with dbus + high latency due to > applications waiting for dbus replies that never comes. > > I suggest rebuilding dbus and dbus-glib (please also paste emerge -pv of both) > and eventually rebuild all reverse dependencies of dbus-glib. > Ouput requested (I cut the non-necessary lines): emerge -pv dbus gives: [ebuild R ] sys-apps/dbus-1.2.24-r2 USE="X -debug -doc (-selinux) -test" 0 kB emerge -pv dbus-glib gives: [ebuild R ] dev-libs/dbus-glib-0.92 USE="bash-completion debug static-libs test -doc" 0 kB Not sure it's related, but I post also the groups my users belongs to. gentoo stefano # groups stefano // standard user -> gives problems root wheel audio video users plugdev stefano gentoo stefano # groups test2 // test user -> gives no problems wheel audio video test2 Maybe adding my user to root was NOT a so good idea (maybe it's time to remove it).
Created attachment 257160 [details] revdep-rebuild I attach here the output of revdep-rebuild. 24 packages have to be rebuilded (though it seems they all are multimedia-related).
(In reply to comment #41) > Ouput requested (I cut the non-necessary lines): > > emerge -pv dbus gives: > [ebuild R ] sys-apps/dbus-1.2.24-r2 USE="X -debug -doc (-selinux) -test" 0 > kB > > emerge -pv dbus-glib gives: > [ebuild R ] dev-libs/dbus-glib-0.92 USE="bash-completion debug static-libs > test -doc" 0 kB > Please re-emerge them, maybe you are suffering some problems like bug 297483 > Not sure it's related, but I post also the groups my users belongs to. > gentoo stefano # groups stefano // standard user -> gives problems > root wheel audio video users plugdev stefano > > gentoo stefano # groups test2 // test user -> gives no problems > wheel audio video test2 > Try to add test2 user to plugdev group, maybe it will break then > Maybe adding my user to root was NOT a so good idea (maybe it's time to remove > it). > Well, I cannot remember any reason for having a normal user be in root group, and it could be a bit risky from a security point of view, I would vote from dropping that one from root group anyway ;-)
(In reply to comment #43) > (In reply to comment #41) > > Ouput requested (I cut the non-necessary lines): > > > > emerge -pv dbus gives: > > [ebuild R ] sys-apps/dbus-1.2.24-r2 USE="X -debug -doc (-selinux) -test" 0 > > kB > > > > emerge -pv dbus-glib gives: > > [ebuild R ] dev-libs/dbus-glib-0.92 USE="bash-completion debug static-libs > > test -doc" 0 kB > > > > Please re-emerge them, maybe you are suffering some problems like bug 297483 > > > Not sure it's related, but I post also the groups my users belongs to. > > gentoo stefano # groups stefano // standard user -> gives problems > > root wheel audio video users plugdev stefano > > > > gentoo stefano # groups test2 // test user -> gives no problems > > wheel audio video test2 > > > > Try to add test2 user to plugdev group, maybe it will break then > > > Maybe adding my user to root was NOT a so good idea (maybe it's time to remove > > it). > > > > Well, I cannot remember any reason for having a normal user be in root group, > and it could be a bit risky from a security point of view, I would vote from > dropping that one from root group anyway ;-) > Already reemerged them. Now I'm doing a revdep-rebuild. Unfortunately I have to study for some exams, so I'll let you know in about 24 hours. Thank you
(In reply to comment #44) > Already reemerged them. Now I'm doing a revdep-rebuild. > Unfortunately I have to study for some exams, so I'll let you know in about 24 > hours. > Thank you > No problem, I am in the same situation ;-)
(In reply to comment #45) > (In reply to comment #44) > > Already reemerged them. Now I'm doing a revdep-rebuild. > > Unfortunately I have to study for some exams, so I'll let you know in about 24 > > hours. > > Thank you > > > > No problem, I am in the same situation ;-) > OK sorry to keep you waiting :(. Found the problem (you'll think I'm crazy): - Removed the standard user "stefano" from the "users" group (gpasswd -d stefano users) -> when gvfs is installed the cpu usage is normal - Added the test users "test2" to the "users" group (before it was running well under this user - gpasswd -a test2 users) -> when gvfs is installed the cpu usage is above all limits and I can't get to see the gnome panel bar :S So it seems to be a group membership problem. Strange though, because normally any user should be added to the "users" group, shouldn't they ? Found the problem and a half-hotfix. Should there be a less critical solution than remove the user from the "users" group ? Thank you -
As I rebooted problems with gnome freezing came with both standard user and test user. CPU is now for both at 100% about. Before I had just restarted XORG / X11 and the problem wasn't coming for the standard user, but after I rebooted the problem came for him too. Any ideas ?
(In reply to comment #46) > OK sorry to keep you waiting :(. > Found the problem (you'll think I'm crazy): > - Removed the standard user "stefano" from the "users" group (gpasswd -d > stefano users) -> when gvfs is installed the cpu usage is normal > - Added the test users "test2" to the "users" group (before it was running well > under this user - gpasswd -a test2 users) -> when gvfs is installed the cpu > usage is above all limits and I can't get to see the gnome panel bar :S > > So it seems to be a group membership problem. > Strange though, because normally any user should be added to the "users" group, > shouldn't they ? > Found the problem and a half-hotfix. Should there be a less critical solution > than remove the user from the "users" group ? > Thank you > - > Please attach /etc/group , but I don't know what can cause this problem :-/ (In reply to comment #47) > As I rebooted problems with gnome freezing came with both standard user and > test user. CPU is now for both at 100% about. > Before I had just restarted XORG / X11 and the problem wasn't coming for the > standard user, but after I rebooted the problem came for him too. > Any ideas ? > Take a look on /var/log/messages, dmesg output and new ~/.xsession-errors contents
Created attachment 257938 [details] Dmesg output with GVFS installed and causing problems
Created attachment 257940 [details] /etc/group content
Created attachment 257941 [details] /var/log/messages without GVFS installed
Created attachment 257942 [details] /var/log/messages with GVFS installed and causing problems
Created attachment 257943 [details] /home/stefano/.xsession_errors without GVFS installed
Created attachment 257945 [details] /home/stefano/.xsession_errors with GVFS installed and causing problems Remark: strangely the thing about the "users" group only happened once and even know both users are subjected to the same problems (with or without reboot). Just ignore the "users" issue from now on (don't know what caused it, but seems like it was just a coincidence).
1. did you complete your revdep-rebuild run ? 2. did you rebuild dbus and dbus-glib ? 3. if you did not rebuild dbus-glib, please run the following command: # emerge -1av $(qfile -C $(find /etc/dbus-1/ -type f) | cut -f1 -d' ' |sort -u) $(qfile -C $(find /usr/share/dbus-1/ -type f) | cut -f1 -d' ' |sort -u) it will rebuild all applications providing dbus configuration files. You should then have lots of configuration files to update. Always pick the one provided by the newly compiled package (the default). Then restart your system, you should have a clean dbus installation and configuration, hopefully that should fix your problems.
(In reply to comment #55) > 1. did you complete your revdep-rebuild run ? > 2. did you rebuild dbus and dbus-glib ? > 3. if you did not rebuild dbus-glib, please run the following command: > # emerge -1av $(qfile -C $(find /etc/dbus-1/ -type f) | cut -f1 -d' ' |sort -u) > $(qfile -C $(find /usr/share/dbus-1/ -type f) | cut -f1 -d' ' |sort -u) > > it will rebuild all applications providing dbus configuration files. You should > then have lots of configuration files to update. Always pick the one provided > by the newly compiled package (the default). Then restart your system, you > should have a clean dbus installation and configuration, hopefully that should > fix your problems. > Strangely the dbus and dbus-glib packages were not pulled in by revdep-rebuild and dbus-glib was neither pulled in by the third command you provided, so I emerged it manually. The emerging is coming to an end. Ill let you know as soon as it is finished.
The third command you gave me did what it had to do. I rebooted, then tried to login under gnome -> same issues with both standard user and test user. I don-t know what to think now ...
I'm a bit lost a well as brute force didn't come around this bug. Was there any configuration file updates btw ? You didn't mention any. After doing a couple of google search, I found some interesting links though: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=573720 https://bbs.archlinux.org/viewtopic.php?pid=744904 http://us.generation-nt.com/answer/bug-602543-libgdu0-gvfs-gdu-volume-monitor-exits-when-plugging-removable-media-non-utf-8-system-help-200946691.html Afaics, you already are running with utf-8 locale. Could you make sure this is the case in your session by running the command "locale" in a terminal in your gnome session. Could you rebuild gvfs with samba disabled so we can check whether or not samba backend is to be blamed here ?
(In reply to comment #58) > Afaics, you already are running with utf-8 locale. Could you make sure this is > the case in your session by running the command "locale" in a terminal in your > gnome session. > Maybe it's unrelated but, when I have tried to read "messages" file uploaded by reporter in comment #51, gedit has failed to autodetect codification and I have needed to manually tell it to use ISO-8859 (while I have no problems opening my messages file). Then, maybe this is like the third problem pointed by Gilles :-D
(In reply to comment #58) > I'm a bit lost a well as brute force didn't come around this bug. > Was there any configuration file updates btw ? You didn't mention any. > > After doing a couple of google search, I found some interesting links though: > > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=573720 > https://bbs.archlinux.org/viewtopic.php?pid=744904 > http://us.generation-nt.com/answer/bug-602543-libgdu0-gvfs-gdu-volume-monitor-exits-when-plugging-removable-media-non-utf-8-system-help-200946691.html > > Afaics, you already are running with utf-8 locale. Could you make sure this is > the case in your session by running the command "locale" in a terminal in your > gnome session. > > Could you rebuild gvfs with samba disabled so we can check whether or not samba > backend is to be blamed here ? > Reporter, any news related with Gilles suggestions?
Sorry for the delay. I'm in the middle of an examination session which will end on 26th January, so I'm a bit busy. Trying out what Gilles said I get: --------------------------------------- stefano@gentoo ~/Desktop $ locale LANG=it_CH.utf8 LC_CTYPE=it_CH.UTF-8 LC_NUMERIC="it_CH.utf8" LC_TIME="it_CH.utf8" LC_COLLATE="it_CH.utf8" LC_MONETARY="it_CH.utf8" LC_MESSAGES="it_CH.utf8" LC_PAPER="it_CH.utf8" LC_NAME="it_CH.utf8" LC_ADDRESS="it_CH.utf8" LC_TELEPHONE="it_CH.utf8" LC_MEASUREMENT="it_CH.utf8" LC_IDENTIFICATION="it_CH.utf8" LC_ALL= --------------------------------------- So theoretically the locale should be utf8, as the locale command, nor env-update or source /etc/profile give errors (I had gotten many times errors while using UTF8 instead of utf8, or even utf-8) but with the current situation no errors, so it should be fine. Concerning the issues with /var/log/messages I'm in the same boat as you: with gedit the file won't open even if I specified the encoding manually, so I had to open it with kate :(. Should I try emerge -uDn world ? That's a dangerous command but if you think it should help [and not create more problems] I may give it a try. Thank you very much
Please take a look on: http://bugs.gentoo.org/show_bug.cgi?id=351567#c15 And: http://bugs.gentoo.org/show_bug.cgi?id=351567#c16
*** Bug 351567 has been marked as a duplicate of this bug. ***
>> Reporter, any news related with Gilles suggestions? I'm looking into it. In http://bugs.gentoo.org/show_bug.cgi?id=351567#c16 the user solved the problem by switching to UTF-8. I don't know how much it's related (since I tried all possible combinations until I got no error), but I'm using it_CH.utf8 and he's using it_IT.UTF-8. Does the little differente in "utf8" vs "UTF-8" be the cause of the problem ?
The "solution" proposed in the debian forum looks good, although I still don't understand which files have been deleted to "reset" the configuration :S. Maybe trying something like that will work ...
Tried the debian solution: tried to delete all files related to purely dbus in /etc/dbus-1/ with no result. Tried the archlinux solution proposed by Gilles, which worked for a few minutes under gnome with gvfs activated, but as soon as I tried to access the icon "Computer" on the gnome-desktop, it freezed again :S By the way ... gnome now opens folders using dolphin and not nautilus, could we change that ? Thank you very much
OK I see myself this bug it's becoming a mess. To clear it a little bit (I hope): - With two screens running, gnome 2.32, no gvfs -> no problems - With two screens running, gnome 2.32 with gvfs -> FREEZE AT LOGIN or at any basic task (see my last post concerning the "trigger") - With only one screen running, gnome 2.32 with gvfs -> no problems This should pull back the problem involving nvidia drivers. However I tried several versions of the drivers, including the latests, and that didn't fix the problem. I find it hard to track down because apparantly there are multiple "triggers" to activate problem once gvfs is installed: double click on an icon, try to access network share, or simply at login. All of this doesn't happen on KDE (but happens as well on xfce, lxde, ...) so this should lead to the conclusion that KDE doesn't suffer of this problem because it's shi**y nature to NOT support xinerama (double scren), whereas all other WM / DE support multiple screen. Now one question is: why does the fact that GVFS is installed trigger this problem ?? I don't know if this reassures you / worries you even more / confuse whatsoever, but assuming this as a new fact - not THE fact, because other causes may still be involved, some unknown and unidentified at the time. Sorry if i didn't precice this before. I didn't think it was related to gvfs (and yet, why should it be ?). Thank you very much
this is indeed really confusing.
(In reply to comment #68) > this is indeed really confusing. > Yeah, unfortunately :S Should I buy an ATI card to check this out :S:S:S ?
(In reply to comment #69) > (In reply to comment #68) > > this is indeed really confusing. > > > > Yeah, unfortunately :S > Should I buy an ATI card to check this out :S:S:S ? > You could probably try to use free driver for testing ("nv" I think) Anyway, you latest comments look contradictory with your comment #61 that looked to point to the UTF-8 problem. On the other hand, your latest problems look more like bug #348253
>> Anyway, you latest comments look contradictory with your comment #61 that looked to point to the UTF-8 problem. No contradiction really. I'm still using IT_CH.utf8 as locale, whereas the other user (in the other bug filled at bugs.gentoo.org reccomended by you) is using IT_IT.UTF-8. I didn't change anything, because I don't get any errors while (for example) updating the system or running source /etc/profile or env-update, so why should it be wrong ? Concerning the nvidia drivers I don't know, because it seems to me the free ones don't really support my card (460 GTX). No idea about the "nouveau" driver. Here they reccomend to use the VESA driver (????): http://www.phoronix.com/scan.php?page=article&item=nvidia_geforce_gtx460&num=4.
With nv driver I get: -------------------------------------- ... (EE) NV: The PCI device 0x10de0e22 ((null)) at 01@00:00:0 has a kernel module claiming it. (EE) NV: This driver cannotr operate until it has been unloaded (EE) NV: The PCI device 0x10de0e22 ((null)) at 01@00:00:0 has a kernel module claiming it. (EE) NV: This driver cannotr operate until it has been unloaded (EE) No devices detected Fatal server error: no screens found -------------------------------------- And if I give "modprobe -r nvidia" before trying to restart X: -------------------------------------- (EE) No devices detected Fatal server error: no screens found --------------------------------------
Now I'm completely lost. Even with **one** screen and gvfs installed the problem appears and gnome freezes :S Seems really to becoming a random-hard-to-track problem. Whatever, how should gvfs and nvidia drivers be related by the way ? I don't really want to stick with kde forever ...
I am completely out of ideas, maybe you are suffering a different problem now that original one :-/, maybe you should recheck ~/.xsession-errors as maybe messages are now different. About UTF8 setup, look at mine: $ cat /etc/env.d/02locale LANG="es_ES.UTF-8" LC_COLLATE="C" Maybe you could try with it (remember to run env-update after changing files under /etc/env.d I guess you have your /etc/hosts setup properly: https://wiki.archlinux.org/index.php/FAQ#Q.29_Why_is_Arch_so_slow.3F_Programs_open_slowly_or_do_not_run_at_all.21
(In reply to comment #74) > I am completely out of ideas, maybe you are suffering a different problem now > that original one :-/, maybe you should recheck ~/.xsession-errors as maybe > messages are now different. > > About UTF8 setup, look at mine: > > $ cat /etc/env.d/02locale > LANG="es_ES.UTF-8" > LC_COLLATE="C" > > Maybe you could try with it (remember to run env-update after changing files > under /etc/env.d > > I guess you have your /etc/hosts setup properly: > https://wiki.archlinux.org/index.php/FAQ#Q.29_Why_is_Arch_so_slow.3F_Programs_open_slowly_or_do_not_run_at_all.21 > OK thank you. I'll try that later when I get back from holiday (about in 2 weeks). Then I'll let you know. Seems strange anyway that it triggers "only" a gvfs problem. Thank you very much
@Pacho Ramos: I tried using your /etc/env.d/locale file (actually I hadn't one, so I just created it) but the funny thing is that it doesn't change anything. I rebooted and I get: ------ stefano@gentoo ~ $ locale LANG=it_CH LC_CTYPE="it_CH" LC_NUMERIC="it_CH" LC_TIME="it_CH" LC_COLLATE=C LC_MONETARY="it_CH" LC_MESSAGES="it_CH" LC_PAPER="it_CH" LC_NAME="it_CH" LC_ADDRESS="it_CH" LC_TELEPHONE="it_CH" LC_MEASUREMENT="it_CH" LC_IDENTIFICATION="it_CH" LC_ALL= ------ On the other side /etc/env.d/02locale ----- stefano@gentoo ~ $ cat /etc/env.d/02locale #LANG="it_IT.UTF-8" #LC_COLLATE="C" LANG="es_ES.UTF-8" LC_COLLATE="C" ----- Futhemore I don't think I have some custom .bashrc (I commented out the export LANG directives in my personal ~/.bashrc). The gvfs problem still appears. Last note (and hope) might be the fact that I once tried to enable vnc / XMBPP (or whatever it's called for gdm) access from remotes machines, although I never got to and never used it. However when sending /etc/init.d/xdm restart I get 2 messages kind of: ---------- Sending kill to process 7631 Sending kill to process 7841 ---------- (The numbers change every time, but they're different between the 2 lines) I rechecked the gdm configuration file, but it seems I disabled it now. Any thoughts ? As for the locale I remember I tried to set it elsewhere that /etc/env.d/02locale, but unfortunately I don't really remember where and the "find" command doesn't locate the file. Thank you very much
(In reply to comment #76) > @Pacho Ramos: I tried using your /etc/env.d/locale file (actually I hadn't one, > so I just created it) but the funny thing is that it doesn't change anything. > I rebooted and I get: > > ------ > stefano@gentoo ~ $ locale > LANG=it_CH > LC_CTYPE="it_CH" > LC_NUMERIC="it_CH" > LC_TIME="it_CH" > LC_COLLATE=C > LC_MONETARY="it_CH" > LC_MESSAGES="it_CH" > LC_PAPER="it_CH" > LC_NAME="it_CH" > LC_ADDRESS="it_CH" > LC_TELEPHONE="it_CH" > LC_MEASUREMENT="it_CH" > LC_IDENTIFICATION="it_CH" > LC_ALL= > ------ > > On the other side /etc/env.d/02locale > ----- > stefano@gentoo ~ $ cat /etc/env.d/02locale > #LANG="it_IT.UTF-8" > #LC_COLLATE="C" > > LANG="es_ES.UTF-8" > LC_COLLATE="C" > ----- > This is not normal, looks like you are defining your locale settings in other way, please check under /etc/env.d and so (maybe using "grep" and a lot of patience could help)
@Pacho Ramos I had a custom locale set in /etc/profile.d/lang.sh (a file I created). After disabling that now I get the same es_ES.UTF-8 as you in all fields of the locale command: ------------- stefano@gentoo ~ $ locale LANG=es_ES.UTF-8 LC_CTYPE="es_ES.UTF-8" LC_NUMERIC="es_ES.UTF-8" LC_TIME="es_ES.UTF-8" LC_COLLATE=C LC_MONETARY="es_ES.UTF-8" LC_MESSAGES="es_ES.UTF-8" LC_PAPER="es_ES.UTF-8" LC_NAME="es_ES.UTF-8" LC_ADDRESS="es_ES.UTF-8" LC_TELEPHONE="es_ES.UTF-8" LC_MEASUREMENT="es_ES.UTF-8" LC_IDENTIFICATION="es_ES.UTF-8" LC_ALL= -------------- Don't know if the problem is fixed though. For now, gvfs behaves well. I'll do some try and see if the problem is fixed. As for now I can login in gnome with gvfs installed (after reboot) and open basic things withouth freezing. I'll let you know. Thank you ;)
Thank you all for the support. I'm now closing this bug since the problem does not happen anymore. Here's a summary of the problem, the cause and the solution (for those not wanting to read the whole bug report which I admit is a "bit" confusing): - Problem: gnome panel fails to load when gvfs is installed. Overall system response is VERY slow. Cannot even get to open a terminal in gnome. The problem appears when using any other DE / WM than KDE (i.e. Gnome, XFCE, LXDE, ...). This problem appears in "top" as an increase to ~ 80% of the CPU usage of udisks-daemon or dbus-daemon. - Cause: mispelled (and badly placed locale configuration): the locale was set to "it_CH.utf8" inside "/etc/profile.d/lang.sh" - Solution: delete /etc/profile.d/lang.sh (or any other file you used to set your locale) and replace it by the "gentoo-way" using "it_CH.UTF-8" inside "/etc/env.d/02locale": "/etc/env.d/02locale" file contents: --------------------- LANG="it_CH.UTF-8" LC_COLLATE="C" --------------------- Hope it helps. Thank you very much to all of you (expecially - but not limited - to Pacho Ramos and Gilles Dartiguelongue). CLOSED / FIXED
Great to know were was the problem :-D Gnome team: would be anyway to check for "proper" locale when merging gvfs? Looks like it needs a "UTF-8" one...
(In reply to comment #79) > Thank you all for the support. I'm now closing this bug since the problem does > not happen anymore. Thanks you for your persistence in trying to fix the problem and the sum up. @pacho, we cannot rely on build time checks for verifying the locale as the package may be build in a build chroot with only the C locale for example. It might be worth writing this down somewhere though if someone comes up again with a weird gvfs bug like this, at least until upstream do something about it.
OK (reopening for proper resolution)
(In reply to comment #81) > (In reply to comment #79) > > Thank you all for the support. I'm now closing this bug since the problem does > > not happen anymore. > > Thanks you for your persistence in trying to fix the problem and the sum up. > > @pacho, we cannot rely on build time checks for verifying the locale as the > package may be build in a build chroot with only the C locale for example. > It might be worth writing this down somewhere though if someone comes up again > with a weird gvfs bug like this, at least until upstream do something about it. > Hi again. Do you need me to do further tests ? Until now (and after about 4 days of normal operating as well as 5-7 reboots) the problem didn't come up again. Still - if I remember correctly - the problem was engaged after and update (and "locale" didn't report any errors while using "utf8" instead of "utf-8"). Maybe something changed in gvfs which messed this up. If you need further tests please ask. Thank you for your time and suppotr.
*** Bug 360609 has been marked as a duplicate of this bug. ***
Thanks too, I had a temporary solution to this bug too, but I'm happy to see there are some steps moving to a solution. Cheers
