[systemd-devel] systemd prevents pulseaudio from shutting down
Hello, i'm using "ecryptfs" to encrypt my home directory and "pam_mount" to have it automatically mounted/unmounted at login/logout. The unmounting never worked and i discoverd that a pulseaudio process of my user was keept running although my user was already logged out. This process had some files opened in "~./pulse" which is why i think my home dir is not unmounted. The only client that was accessing my pulseaudio process was the module systemd-login.c [pulseaudio] main.c: 1 client(s) logged in. [pulseaudio] main.c: index: 0 [pulseaudio] main.c: driver: [pulseaudio] main.c: owner module: 19 [pulseaudio] main.c: properties: [pulseaudio] main.c: application.name = "Login Session 2" [pulseaudio] main.c: systemd-login.session = "2" I already tried using "KillUserProcesses=yes" in "logind.conf",to to have systemd kill all user process on logout, but it didn't help. So is it true that systemd keeps my pusleaudio alive? If so how can i avoid that? And is this behavior intended? Thanks in advance, Robert ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] systemd prevents pulseaudio from shutting down
Dear Robert, Am Sonntag, den 03.06.2012, 11:35 +0200 schrieb Robert Buhren: > i'm using "ecryptfs" to encrypt my home directory and "pam_mount" to > have it automatically > mounted/unmounted at login/logout. The unmounting never worked and i > discoverd that a pulseaudio process of my user was keept running > although my user was already logged out. This process had some files > opened in "~./pulse" which is why i think my home dir is not unmounted. > The only client that was accessing my pulseaudio process was the module > systemd-login.c > > [pulseaudio] main.c: 1 client(s) logged in. > [pulseaudio] main.c: index: 0 > [pulseaudio] main.c: driver: > [pulseaudio] main.c: owner module: 19 > [pulseaudio] main.c: properties: > [pulseaudio] main.c: application.name = "Login Session 2" > [pulseaudio] main.c: systemd-login.session = "2" > > I already tried using "KillUserProcesses=yes" in "logind.conf",to to > have systemd kill all user process on logout, but it didn't help. > > So is it true that systemd keeps my pusleaudio alive? If so how can i > avoid that? And is this behavior intended? I will not be able to help you, but you can give more information. What distribution do you use? What version of Linux, PulseAudio, systemd? Maybe even attach the PulseAudio’s unit file for reference. Thanks, Paul signature.asc Description: This is a digitally signed message part ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] systemd prevents pulseaudio from shutting down
Am 03.06.2012 18:46, schrieb Paul Menzel: >> i'm using "ecryptfs" to encrypt my home directory and "pam_mount" to >> have it automatically >> mounted/unmounted at login/logout. The unmounting never worked and i >> discoverd that a pulseaudio process of my user was keept running >> although my user was already logged out. This process had some files >> opened in "~./pulse" which is why i think my home dir is not unmounted. > > I will not be able to help you, but you can give more information. What > distribution do you use? What version of Linux, PulseAudio, systemd? > Maybe even attach the PulseAudio’s unit file for reference. pulseaudio process is not stopped on Fdora 15/16 as example after logout i generally hate the idea of sound.daemon depeding on user-sessions because it prevents things like mpd from running completly in background and if i hear music on my desktop and switch with CTRL+F2 to a terminal it is simply idiotic that music stops to play signature.asc Description: OpenPGP digital signature ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
[systemd-devel] using native *.mount units instead of /etc/fstab - unpredictable state in case of many disks
Hello. I've hit by a strange issue on a machine with 8 hard disks. if I list them all in /etc/fstab they mounts fine. If I provide native *.mount files then almost every reboot one or several disks failed to mount (with 32 error, e.g. "code=exited, status=32", which means that no hadrwae is available at the moment of mount attempt, wrong fs, etc). If I login and run "systemctl restart media-mypathX.mount" it mounts just fine. Here is how these *.mount files looks like (no RAID, no LVM, etc - just plain alone single-partitioned hard disk): === [Unit] Description=Random Stuff Directory #After=media.mount systemd-udev-settle.service dev-disk-by\x2dlabel-heap.device After=media.mount Before=nfs-server.service [Mount] #What=LABEL="heap" What=/dev/disk/by-label/heap Where=/media/heap Type=xfs Options=defaults,noatime,nodiratime === You can see - I tried to start it after systemd-udev service as well as after appropriate *.device but w/o success. Also I tried to mount using labels and using corresponding paths. No success either. Could someone point me out what did I miss? I'm sure there is some specific *.service or *.target file I must add it as a dependency. Fedora 18 if it matters. -- With best regards, Peter Lemenkov. ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] systemd prevents pulseaudio from shutting down
On 03.06.2012 18:46, Paul Menzel wrote: Dear Robert, Am Sonntag, den 03.06.2012, 11:35 +0200 schrieb Robert Buhren: i'm using "ecryptfs" to encrypt my home directory and "pam_mount" to have it automatically mounted/unmounted at login/logout. The unmounting never worked and i discoverd that a pulseaudio process of my user was keept running although my user was already logged out. This process had some files opened in "~./pulse" which is why i think my home dir is not unmounted. The only client that was accessing my pulseaudio process was the module systemd-login.c [pulseaudio] main.c: 1 client(s) logged in. [pulseaudio] main.c: index: 0 [pulseaudio] main.c: driver: [pulseaudio] main.c: owner module: 19 [pulseaudio] main.c: properties: [pulseaudio] main.c: application.name = "Login Session 2" [pulseaudio] main.c: systemd-login.session = "2" I already tried using "KillUserProcesses=yes" in "logind.conf",to to have systemd kill all user process on logout, but it didn't help. So is it true that systemd keeps my pusleaudio alive? If so how can i avoid that? And is this behavior intended? I will not be able to help you, but you can give more information. What distribution do you use? What version of Linux, PulseAudio, systemd? Maybe even attach the PulseAudio’s unit file for reference. Thanks, Paul Dear Paul, i'm using archlinux with gnome3.4. Pulseaudio 2.0, systemd 184, and Linux 3.4. Pulseaudio has no unitfile, as it is started by gnome itself. Regards, Robert ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] using native *.mount units instead of /etc/fstab - unpredictable state in case of many disks
On Sun, Jun 3, 2012 at 8:18 PM, Peter Lemenkov wrote: > Could someone point me out what did I miss? I'm sure there is some > specific *.service or *.target file I must add it as a dependency. > Fedora 18 if it matters. I'd try "systemctl show .mount" when it is based on fstab, and when it is a mount unit, and compare the two. It will probably tell you what you are missing. -t ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] using native *.mount units instead of /etc/fstab - unpredictable state in case of many disks
On Sun, Jun 3, 2012 at 8:18 PM, Peter Lemenkov wrote: > I've hit by a strange issue on a machine with 8 hard disks. if I list > them all in /etc/fstab they mounts fine. If I provide native *.mount > files then almost every reboot one or several disks failed to mount > (with 32 error, e.g. "code=exited, status=32", which means that no > hadrwae is available at the moment of mount attempt, wrong fs, etc). > [Unit] > Description=Random Stuff Directory > #After=media.mount systemd-udev-settle.service > dev-disk-by\x2dlabel-heap.device > > After=media.mount > Before=nfs-server.service > > [Mount] > #What=LABEL="heap" > What=/dev/disk/by-label/heap > Where=/media/heap > Type=xfs > Options=defaults,noatime,nodiratime > > === > > You can see - I tried to start it after systemd-udev service as well > as after appropriate *.device but w/o success. Also I tried to mount > using labels and using corresponding paths. No success either. The unit media.mount does not exist anymore in the recent system version. After=*.device would only work if the device is already there, and can be included in the transaction, but then the After= should have no effect. > Could someone point me out what did I miss? I'm sure there is some > specific *.service or *.target file I must add it as a dependency. > Fedora 18 if it matters. Maybe this works: After=local-fs-pre.target and hook it into /usr/lib/systemd/system/local-fs.target.wants Kay ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] systemd prevents pulseaudio from shutting down
Am 03.06.2012 20:22, schrieb Robert Buhren: > i'm using archlinux with gnome3.4. Pulseaudio 2.0, systemd 184, and Linux 3.4. > Pulseaudio has no unitfile, as it is started by gnome itself. however the question remains: why is pulseaudio still running after logout signature.asc Description: OpenPGP digital signature ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
[systemd-devel] Automount units: intermittent failure when the real mount triggers another mount first
Hi, I stumbled accross this issue when trying to get this construct working: mnt-portage.automount is an enabled unit mnt-portage.mount mounts an nfs4 export and depends, via Requires=, and After= on some other units that in turn depend on var-lib-nfs-rpc_pipefs.mount The result is that the very first access to /mnt/portage fails with -ENODEV while any subsequent access works just fine and all units in the whole dependency chain have been started successfully. To get any weird NFS-interaction out of the way, I reduced the whole thing to a minimal testcase: = tmp-test.automout, enable this one [Unit] Description=automount test automount point After=local-fs.target [Automount] Where=/tmp/test [Install] WantedBy=multi-user.target = tmp-test.mount [Unit] Description=automount test mount point Requires=tmp-dummy.mount After=tmp-dummy.mount [Mount] What=/ Where=/tmp/test Options=bind = tmp-dummy.mount [Unit] Description=automount test secondary mount point [Mount] What=none Where=/tmp/dummy Type=tmpfs First access to /tmp/test results in "no such device", next access will succeed. To test things further, I changed tmp-test.mount to Requires=, After= tmp-dummy.service instead of tmp-dummy.mount: = tmp-dummy.service [Unit] Description=automount test secondary mount point as a service [Service] Type=oneshot RemainAfterExit=true ExecStartPre=/bin/mkdir -p /tmp/dummy ExecStart=/bin/mount -t tmpfs none /tmp/dummy ExecStop=/bin/umount /tmp/dummy This has the same effect. The correspondin log entry: Jun 03 21:26:30 test systemd[1]: Sending failure: No such device This comes from here: #0 mount_notify_automount (m=0x1076a90, status=-19) at src/core/mount.c:616 #1 0x00422133 in mount_set_state (m=0x1076a90, state=MOUNT_DEAD) at src/core/mount.c:659 #2 0x00424f05 in mount_fd_event (m=0x10413d0, events=10) at src/core/mount.c:1613 #3 0x00410a80 in process_event (ev=0x7fffa4d2ba10, m=0x10413d0) at src/core/manager.c:1392 #4 manager_loop (m=0x10413d0) at src/core/manager.c:1497 #5 0x0040b2f3 in main (argc=1, argv=0x7fffa4d2c238) at src/core/main.c:1619 The BP on mount.c:616 is hit like this six times in a row. Misc info: Distro: Gentoo Kernel: 3.4.0 systemd: 7ff5404be1bad93cb8facbcae0bc78f77f9e067d Dunno if this is Gentoo-specific or triggered by some broken kernel configuration I used, but I'll gladly help figure this out if maybe someone could push me into the right direction. Thanks, Malte ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
[systemd-devel] [PATCH] Punt duplicate definition of InhibitWhat
From: Malte Starostik Trivial fix for: src/login/logind-inhibit.h:37:3: error: redefinition of typedef 'InhibitWhat' src/login/logind-inhibit.h:26:26: note: previous declaration of 'InhibitWhat' was here Signed-off-by: Malte Starostik --- src/login/logind-inhibit.h |1 - 1 files changed, 0 insertions(+), 1 deletions(-) diff --git a/src/login/logind-inhibit.h b/src/login/logind-inhibit.h index 4377f00..e72536f 100644 --- a/src/login/logind-inhibit.h +++ b/src/login/logind-inhibit.h @@ -23,7 +23,6 @@ ***/ typedef struct Inhibitor Inhibitor; -typedef enum InhibitWhat InhibitWhat; #include "list.h" #include "util.h" -- 1.7.3.4 ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] [PATCH] Add "set-default-target" to systemctl
'Twas brillig, and har...@redhat.com at 23/04/12 08:23 did gyre and gimble: > From: Harald Hoyer Ping! any comments on this one? I'd personally like to see it in if there are no objections. Col -- Colin Guthrie gmane(at)colin.guthr.ie http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/ ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] systemd prevents pulseaudio from shutting down
'Twas brillig, and Reindl Harald at 03/06/12 20:08 did gyre and gimble: > > > Am 03.06.2012 20:22, schrieb Robert Buhren: >> i'm using archlinux with gnome3.4. Pulseaudio 2.0, systemd 184, and Linux >> 3.4. >> Pulseaudio has no unitfile, as it is started by gnome itself. > > however > > the question remains: why is pulseaudio still running after logout That's the intended behaviour. PulseAudio will spawn itself automatically when needed and shut itself down when it's not needed any more, after a suitable timeout (defaults to 20). As startup is quite intensive (it requires a lot of probing of the hardware to see what modes it uses) we take two general precautions to avoid excessive restarts: 1) We implement an "exit-idle-time". Simply set this to 0 if you want PA to exit by itself immediately after it becomes unused (man pulse-daemon.conf) 2) When logging into an X11 session, we load a special X11 module that ensures that PA is not exited until the X11 session is finished. So in this setup you have two main options to get the results you want. The first is to simply set the exit-idle-time=0 in daemon.conf. This should make PA behave generally a bit more gracefully, but does nothing to help any other apps that may behave in a similar way. The second option (and IMO the better one) is to configure pam_systemd to kill the processes of the user session when it's done. This should be a matter of setting the kill-session-processes= (and optionally the kill-only-users=) options in your pam configuration. See man pam_systemd. HTHs Col -- Colin Guthrie gmane(at)colin.guthr.ie http://colin.guthr.ie/ Day Job: Tribalogic Limited http://www.tribalogic.net/ Open Source: Mageia Contributor http://www.mageia.org/ PulseAudio Hacker http://www.pulseaudio.org/ Trac Hacker http://trac.edgewall.org/ ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
Re: [systemd-devel] Path to a mount point that begins with causes problems
On Thu, May 31, 2012 at 02:14:14PM +0200, Peter Lindgren wrote: > Hi, > > I'm using systemd-44 and experiencing some problems with my mount unit. > > I have an /etc/systemd/system/.aufs-normal.mount file > - > [Unit] > Description=normal mount > > [Mount] > What=/dev/mmcblk0p5 > Where=/.aufs/normal > Type=ext2 > - > > I am able to run 'systemctl start .aufs-normal.mount' and 'systemctl stop > .aufs-normal.mount' to mount/unmount /dev/mmcblk0p5 correctly. > > I then have created an /etc/systemd/system/dummy.service file > -- > [Unit] > Description=Dummy service > Requires=.aufs-normal.mount > After=.aufs-normal.mount > > [Service] > ExecStart=/bin/dummy > Type=oneshot > RemainAfterExit=yes > - > > When I then run 'systemctl start dummy.service' systemd refuse to mount my > .aufs/normal directory. > > If I rename the '.aufs' directory to 'aufs' in the example above it works > fine. > I have tried to escape the without any success. > > Any ideas what's causing this? > > Thanks, > Peter > This is present in 184 as well since fstab is just a source for more units. Best I can tell, unit files with a leading '.' are simply ignored. d ___ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
