[opensuse-factory] Is it true that after an update we don't need to reboot?
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, After the last YOU update of dbus things, a run of this code: lsof | grep -E 'RPMDELETE|;|path inode=' detects a long list of applications and libraries that are still running the old non updated library; a sample: COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME console-k 2962 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) hald 3027 haldaemon mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) hald-runn 3028 root mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) hald-addo 3074 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) avahi-dae 4559 avahi mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) powersave 4837 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) login 5313 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gdm 5551 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-ses 21496 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-scr 21562 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-pan 21573 kar mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) nautilus 21582 kar mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) beagled 21585 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) resapplet 21591 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-cup 21592 kar mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) nm-applet 21602 kar mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) gnome-vfs 21604 kar mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) gnome-pow 21610 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-vol 21611 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) main-menu 21638 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) tomboy 21640 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) mixer_app 21642 kar mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-ses 29456 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) nautilus 29531 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-vfs 29537 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) resapplet 29556 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-scr 29557 cer mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) gnome-cup 29568 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) opensuse- 29595 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) apparmora 29620 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) nm-applet 29623 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) seahorse- 29648 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gweather- 29707 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) mini_comm 29709 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) deskbar-a 29713 cer mem REG 22,70 1381793 /usr/lib/python2.5/site-packages/_dbus_bindings.so (path inode=1577045) stickynot 29715 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) evolution 29930 cer mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) This is known, is not new: but I believe YOU should warn users of this situation and advise to restart services/programs and list them (yes, there is a Bugzilla, ignored) Obviously, a lot of those are reloaded after login out of the X11 sessions. But not all: gdm remains, needing an "init 3; init 5". Then, there are still some pending: COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME console-k 2962 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) console-k 2962 root mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) hald 3027 haldaemon mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) hald 3027 haldaemon mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) hald-runn 3028 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) hald-runn 3028 root mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) hald-addo 3074 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) hald-addo 3074 root mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) hald-addo 3088 haldaemon mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) hald-addo 3435 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) hald-addo 3435 root mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) avahi-dae 4559 avahi mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) powersave 4837 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) powersave 4837 root mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) login 5313 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) login 5315 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-vfs 29537 cer mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) gnome-vfs 29537 cer mem REG 22,70 1560834 /usr/lib/libdbus-glib-1.so.2.1.0 (path inode=1560833) Notice that gnome-vfs did not die after closing login out, nor after killing X11. That's another bug, reported, I think. I know that after login out of gnome I have to manually kill some apps like this: cer@nimrodel:~> beagle-shutdown ; killall beagled gnome-vfs-daemon gnome-vfs-daemon dbus-daemon dbus-daemon gpg-agent ; sleep 5; killall -9 beagled (yes, beagle normally refuses to die...) Some services have to be restarted: nimrodel:~ # rcpowersaved restart Shutting down powersaved done Starting powersaved: done nimrodel:~ # rchal restart Shutting down HAL daemon done Starting HAL daemon done Loading CPUFreq modules (CPUFreq not supported) nimrodel:~ # rcconsolekit restart Shutting down ConsoleKit done Starting ConsoleKit done nimrodel:~ # nimrodel:~ # rcavahi-daemon restart Shutting down Avahi daemon done Starting Avahi daemon done Still one more remaining: COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME login 5313 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) login 5315 root mem REG 22,70 1349371 /lib/libdbus-1.so.3.2.0 (path inode=1007980) which clears after I log out of all text sessions. So... in the end, a reboot was not required, but after running YOU and doing a security update, the machine is still at risk because the update is not really applied till many services and applications are restarted. Or recomend a "init 1 ; init 5". But being silent about this problem may leave many at risk, because the hole plugged by the security update is very open if the unsuspecting user does nothing. IMO, of course :-) - -- Cheers, Carlos Robinson -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4-svn0 (GNU/Linux) iD8DBQFH0Z9vtTMYHG2NR9URAsGvAJ4iBPB1nID+2jkQbzw2fLeWcE2g7gCcCqJ3 U4JCtpn1XvGEktw1+pnheqo= =4dRe -----END PGP SIGNATURE----- --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org
On Fri, Mar 07, 2008 at 09:02:47PM +0100, Carlos E. R. wrote:
Hi,
After the last YOU update of dbus things, a run of this code:
I think for this problem it was sufficient that dbus daemons got restarted. It was however, well, not a critical problem. We will publish a kernel update for your reboot convience some time. Ciao, Marcus --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The Friday 2008-03-07 at 23:25 +0100, Marcus Meissner wrote:
On Fri, Mar 07, 2008 at 09:02:47PM +0100, Carlos E. R. wrote:
Hi,
After the last YOU update of dbus things, a run of this code:
I think for this problem it was sufficient that dbus daemons got restarted.
It was however, well, not a critical problem.
That's not my point. I just took this update as an example that YOU does not say what services or programs should be restarted to fully apply the last update just made. In this case they were a lot, even if the security implication is minimal.
We will publish a kernel update for your reboot convience some time.
X'-) - -- Cheers, Carlos E. R. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4-svn0 (GNU/Linux) iD8DBQFH0dBstTMYHG2NR9URAjypAJ0dqqci2KPeLVTSz4oHfMAcfLgAUgCbBdCI PhDfsPMNXIBjzPQKVyK460E= =z+mM -----END PGP SIGNATURE----- --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org
Marcus Meissner escribió:
I think for this problem it was sufficient that dbus daemons got restarted.
I think so too, anyway dbus-1.rpm does not %restart_on_update and according to the changelog it is on purpose... -- "Morality is merely an interpretation of certain phenomena — more precisely, a misinterpretation." - Friedrich Nietzsche Cristian Rodríguez R. Platform/OpenSUSE - Core Services SUSE LINUX Products GmbH Research & Development http://www.opensuse.org/ --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org
Am Samstag, 8. März 2008 schrieb Cristian Rodríguez:
Marcus Meissner escribió:
I think for this problem it was sufficient that dbus daemons got restarted.
I think so too, anyway dbus-1.rpm does not %restart_on_update and according to the changelog it is on purpose...
Yes, because it would crash half your session. Greetings, Stephan --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org
On Fri, Mar 07, 2008 at 08:36:46PM -0300, Cristian Rodríguez wrote:
Marcus Meissner escribió:
I think for this problem it was sufficient that dbus daemons got restarted.
I think so too, anyway dbus-1.rpm does not %restart_on_update and according to the changelog it is on purpose...
Actually the DBUS developers requested that we add RebootNeeded: Yes before, we just forgot it for our security patch. Ciao, Marcus --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org
Samstag, 8. März 2008 Marcus Meissner:
Actually the DBUS developers requested that we add RebootNeeded: Yes before, we just forgot it for our security patch.
So, is this the public announcement that people who never reboot might appreciate? Wolfgang --------------------------------------------------------------------- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org
participants (5)
-
Carlos E. R.
-
Cristian Rodríguez
-
Marcus Meissner
-
Stephan Kulow
-
Wolfgang Woehl