summaryrefslogtreecommitdiff
path: root/daemon
Commit message (Collapse)AuthorAgeFilesLines
* Fix setting of WINDOWPATH so it does not assume the format is 32 bit. FixesBrian Cameron2010-05-241-1/+1
| | | | bug #609272.
* Use g_remove to remove the XAUTH_DIR instead of g_unlink, which is safer.Brian Cameron2010-05-191-1/+1
|
* Reapply a better fix for bug #594818 that addresses the problem with theBrian Cameron2010-05-195-3/+76
| | | | previous commit.
* Revert fix for bug #594818 since it breaks the build.Brian Cameron2010-05-191-39/+3
|
* Check envrioment WINDOWPATH before setting it.Halton Huo2010-05-141-3/+5
| | | | Fixes https://bugzilla.gnome.org/show_bug.cgi?id=618510
* Fix bug #594818 so that each GDM session has separate GConf settings.Brian Cameron2010-05-111-3/+39
| | | | | | This way when a user enables or kills at AT program it will only start or exit on the display where the user made the request, and not on all displays.
* String fixes. Patch by Philip Withnall. Fixes bug #609178.Brian Cameron2010-05-036-16/+16
|
* Fix build failure when libxdmcp is not available. Fixes bug #616984.Brian Cameron2010-05-031-1/+2
|
* Sort session using file basenameFrederic Crozat2010-05-031-5/+23
| | | | | Ensure behaviour in default / fallback session is consistent with gdm <= 2.20. Fixes Mandriva bug 58501 / GNOME bug 617516.
* Make gdm_slave_setup_xhost_auth return voidRay Strode2010-04-271-1/+1
| | | | | It was doing that anyway, just the function signature was wrong.
* Add missing header (fix GNOME bug #616888)Frederic Crozat2010-04-271-0/+1
|
* Fix XDMCP so it shows the greeter again after logout. Patch by MichaelBrian Cameron2010-04-221-0/+15
| | | | Young. Part of fix for bug #606724.
* Fix XDMCP show it shows the greeter again after logout. Patch by Halton Huo.Brian Cameron2010-04-221-0/+37
| | | | Part of the fix for #606724.
* Set WINDOWPATH for user session. Fixes bug #609272.Brian Cameron2010-04-213-9/+96
|
* Drop xhost localuser:gdm and localuser:root after the user session starts.Brian Cameron2010-04-211-22/+35
| | | | Fixes bug #605350.
* Ensure Init script is called before starting autologin (fixes bug #614488).Brian Cameron2010-04-211-0/+2
|
* Fix duplicated and missing headers (fixes bug #614477).Brian Cameron2010-04-211-3/+1
|
* Bug 609179 — Date translation and duplicated stringsPhilip Withnall2010-04-212-2/+2
| | | | Merge two similar translatable strings. Closes: bgo#609179
* Fix bug #607738 since Vincent confirmed the fix for #607658 is going intoBrian Cameron2010-03-081-1/+1
| | | | | gnome-session. No longer kill gnome-session process group when the user logs in. Now kill just gnome-session and let it safely exi.
* s/"opened"/"conversation-started"/Ray Strode2010-02-091-1/+1
| | | | | It was missed in round up done in commit f5c918550dbabe04ece27e9f80234af93ef54514.
* Don't tear down greeter until pam_open_session finishesRay Strode2010-02-089-8/+289
| | | | | | Some PAM modules ask questions at that late stage of the game, and so we need a greeter to forward the questions on to the user.
* Rename session "Open" method to "StartConversation"Ray Strode2010-02-089-60/+60
| | | | | | | Open is very general. If we rename it, then we can use the term "open" for the state after PAM finishes checking identity, authorizing, and giving out credentials, before a session is started.
* Revert "init session pid to -1"Ray Strode2010-02-081-0/+1
| | | | | | This reverts commit da3d88004af9e0fc69a70282f261df409a008eb2. It's not correct and was accidentally pushed.
* init session pid to -1Ray Strode2010-02-081-1/+0
|
* The D-Bus connection between the worker and the GUI should never timeout,Brian Cameron2010-01-271-3/+6
| | | | so pass INT_MAX for the timeout rather than 10 minutes. Fixes bug #607861.
* remove remaining traces of HalMartin Pitt2010-01-263-292/+0
| | | | | | | | The code which used the Hal connection already was disabled, so gdm connected to Hal in vain (which just triggered Hal startup when using D-Bus activation). Remove all remaining traces of hal now. https://bugzilla.gnome.org/show_bug.cgi?id=593787
* Allow building without libxklavier. Fixes bug #606808.Brian Cameron2010-01-261-2/+6
|
* Don't double-free TryExec variableRay Strode2010-01-201-0/+1
| | | | | If the program could not be found the exec variable would get freed twice.
* Fix crash in getting system keyboard layoutMartin Pitt2010-01-181-11/+12
| | | | | | | | | | | In get_system_default_layout(), use a static variable for xkl_engine_get_instance() result, and don't close the X Display. This fixes the crash that happens at the second call of get_system_default_layout(): xkl_engine_get_instance() returns a singleton which saves the passed X Display instance, so we must never close it. https://launchpad.net/bugs/505972
* Fix typo highlighted in review of bug #606703.Brian Cameron2010-01-151-1/+1
|
* Only run PostLogin script post loginRay Strode2010-01-131-1/+4
| | | | | | | | We don't want to run it when stopping the greeter for other reasons. Spotted by Halton. See https://bugzilla.gnome.org/show_bug.cgi?id=602403#c6
* Run PostLogin script as normal userBrian Cameron2010-01-132-4/+5
| | | | | | | | | | | | | | | | | 1) In daemon/gdm-slave.c it now sets PATH to GDM_SESSION_DEFAULT_PATH so that the scripts have the system normal PATH when run. This is much better than hardcoding the PATH in the scripts themselves, especially since the Init, PreSession, PostSession, and PostLogin scripts are designed to run per-display and it makes things just complicated if users need to add the common system path to any per-display script they might want to use. 2) The Init and PreSession scripts are fixed to set the PATH to "@X_PATH:$PATH" and not add "/bin:/usr/bin" since this isn't neeed since PATH is already set to GDM_SESSION_DEFAULT_PATH before running the script. See https://bugzilla.gnome.org/show_bug.cgi?id=602403
* Fix up XDMCP for SolarisBrian Cameron2010-01-131-5/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After doing further testing, I notice that XDMCP still does not work on Solaris. This commit fixes: - In gdm-address.c, there was a problem where a failure in gdm_address_get_hostname could cause an infinite loop since it would call gdm_address_debug, which would then call gdm_address_get_hostname again. I fixed this by making gdm_address_debug call a private _gdm_address_debug function and gdm_address_get_hostname calls _gdm_adress_debug rather than gdm_address_debug. - Two calls in gdm-address.c were using "sizeof (struct sockaddr_storage)" and changing this to "(int) gdm_sockaddr_len (address->ss)" works better. This is the same issue as fixed in the previous patch. I just missed these spots before. - In daemon/gdm-xdmcp-display-factory.c in the on_hostname_selected() function it is necessary to set hints.ai_socktype to "SOCK_DGRAM" or else I would get this error: WARNING: Unable get address: service name not available for the specified socket type - In daemon/gdm-xdmcp-display-factory.c in the decode_packet() function and in gui/simple-chooser/gdm-host-chooser-widget.c it is necessary to set ss_len to "gdm_sockaddr_len (&clnt_ss)" instead of "sizeof (clnt_ss)". - In gui/simple-chooser/gdm-host-chooser-widget.c in find_broacast_address it is also necessary to check for "(errno != ENXIO)". See https://bugzilla.gnome.org/show_bug.cgi?id=494817
* Don't try to print NULL valuesBrian Cameron2010-01-131-44/+64
| | | | | | This causes crashes on Solaris. See https://bugzilla.gnome.org/show_bug.cgi?id=494817
* Detect default layout with xklavierMartin Pitt2010-01-122-3/+39
| | | | | | Try to read the default layout by connecting to the X server. https://bugzilla.gnome.org/show_bug.cgi?id=572765
* Don't reveal which user names are valid without authenticatingBrian Cameron2010-01-121-0/+8
| | | | | | | | | | | | | | | | | | | If you enter an invalid account name, then hit return for the password, it pops up "No account present for user" GDM should not allow users to find out which account names are valid or invalid without authenticating. I understand that the face browser does expose some users, but this is still an issue for systems that do not have the face browser enabled. Also, even with the Face Browser, certain users are filtered or excluded, and users should not be able to find out information about such users. To fix, this problem, simply change the PAM_USER_UNKNOWN return code code from pam_authenticate to PAM_AUTH_ERR. See https://bugzilla.gnome.org/show_bug.cgi?id=606703
* Reload i18n file automaticallyTakao Fujiwara2009-12-212-13/+103
| | | | | | This was fixed in gdm 2.20 by bug #384603. http://bugzilla.gnome.org/show_bug.cgi?id=599273
* Fix bug #603756, ensure runtime dir is always set.Brian Cameron2009-12-041-0/+3
|
* Fix X-GDM-BypassXSession to X-GDM-BypassXsession.Brian Cameron2009-11-242-3/+4
|
* Allow session scripts to bypass Xsession launcherNiall Power2009-11-183-2/+62
| | | | | | | | | | | It is occasionally useful to launch sessions that aren't run through Xsession. This is good, for starting a bare xterm session for instance. We now allow a session to specify that it shouldn't be run through the Xsession script with the key file entry: X-GDM-BypassXsession=true
* Fix GDM debugging so that it works. Now debugging is enabled via theBrian Cameron2009-11-096-31/+102
| | | | debug/Enable key in the custom.conf file. See bug #596831.
* Lock down /var/run/gdmRay Strode2009-11-051-7/+7
| | | | | We don't need it so open now that screenshots are written to their own directory, and having it open has implications for quota abuse.
* Create screenshot dir at runtime if not availableRay Strode2009-11-053-0/+34
| | | | | We want the screenshot dir to be owned by the GDM user, so the greeter can write screenshots to it.
* Improve strings for translation. Partially fix bug #582891.Brian Cameron2009-10-307-11/+11
|
* Port fix for bug #494817 to the new GDM rewrite. This fixes the code soBrian Cameron2009-10-281-10/+10
| | | | | | | that the length of the sockaddr is set to the correct length for IPv4 or IPv6. This problem affects Solaris and FreeBSD. Linux only require that the length be greater or equal to the expected length, so you would not see this problem on Linux.
* Don't send auth-failed when worker diesRay Strode2009-10-221-10/+2
| | | | | | Authentication hasn't failed, it just got aborted before it could. This prevents a crash that happens when switching runlevels while the login screen is up.
* Drop "stopped" signal from worker-job classRay Strode2009-10-213-26/+0
| | | | It was unused, dead code.
* Drop duplicated entry introspection outputRay Strode2009-10-211-1/+0
|
* Make /var/log/gdm if it doesn't exist. Fixes bug #589862.Brian Cameron2009-10-191-1/+5
|
* Don't add localuser auth entries for XDMCPRay Strode2009-10-131-1/+4
| | | | | | | | | | | | We add peer credential verified access to the X server for local X servers. This makes no sense for remote connections and will result in BadAccess, which we weren't trapping. This commit avoids calling XAddHosts for remote connections. This should address bug 598142 which was investigated and filed by Richard Evans <richard.evans@datanomic.com>
View Lorry Controller Status