https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c zhang jiajun changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |jzhang@suse.com AssignedTo|bnc-team-screening@forge.pr |systemd-maintainers@suse.de |ovo.novell.com | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c3 Rainer Krienke changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |krienke@uni-koblenz.de --- Comment #3 from Rainer Krienke 2014-02-17 10:11:26 UTC --- I installed systemd-208-549.1.x86_64 from the repos given above. I also manually added "# X-Start-Before: autofs" to the installed nfs init script into the LSB header. The nfs header looks like this now: ### BEGIN INIT INFO # Provides: nfs # Required-Start: $network $portmap # Required-Stop: $network $portmap # Default-Start: 3 5 # Default-Stop: 0 1 2 6 # Short-Description: NFS client services # Description: All necessary services for NFS clients # X-Systemd-RemainAfterExit: true # X-Start-Before: autofs ### END INIT INFO Unfortunately the behaviour is still the same. 9 in 10 tries to reboot the system it hangs and does not shut down. In one case the system rebooted as it should. In only about 5 sec the shutdown part had been completed in this case. Thanks Rainer -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c5 --- Comment #5 from Rainer Krienke 2014-02-17 11:29:56 UTC --- Can you point me to the version of sysconfig-network I should try? I guess you are not talking about updates from stadard 13.1 updade repository? Those are already installed. Thanks Rainer -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c7 --- Comment #7 from Rainer Krienke 2014-02-17 12:20:57 UTC --- The output is: * Do Jan 23 2014 mt@suse.de - A systemctl status is quite expensive as it (by default) parses logs, etc. Use is-active where we need the code only (bnc#859360). [+ 0009-Use-systemctl-quiet-is-active-instead-status.patch] Seems the patch has not yet arrived on all update servers. I did a zypper refresh and a zypper update but the change comment you wrote above is not yet visible. So I will wait some days and try again. As soon as I have this update I will report if it works for me. Thanks Rainer -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c9 --- Comment #9 from Rainer Krienke 2014-02-20 08:09:01 UTC --- Just installed the patch sysconfig-network-0.81.5-18.1. The behaviour now is very different depending on where I tell the system what to do. - In the kde display manager kdm (on console I did an ls on my home, so at least one NFS mount is active now): A "shutdown system" (/sbin/halt -p) does work always (did not work before) a "reboot system" (/sbin/reboot) does never work, it hangs - When I log in (my home is on NFS) and have a running KDE session neither "shutdown" not "reboot" work. The system then always hangs. Thanks Rainer -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c11 --- Comment #11 from Dr. Werner Fink 2014-02-20 09:08:30 UTC --- Just mounted my working partition as NFS share at an uptodate 13.1 virtual system with the setup I've described in my last comment. I've added a line in /etc/fstab with the mount options relatime,rsize=32768,wsize=32768,intr,soft,noauto as I mount this partition on demand. Just tried several times: first mount it which triggers the start of rpc.statd and then just do a shutdown -r +0 and it works flawless. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c13 Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- InfoProvider|krienke@uni-koblenz.de |nfbrown@suse.com --- Comment #13 from Dr. Werner Fink 2014-02-20 10:04:52 UTC --- (In reply to comment #12) Thanks a lot, you made my day :) @ Neil : it seems that there should be a further update of the nfs-utils but be aware the the current systemd of openSUSE 13.1 does not know about the `+' of the line # X-Start-Before: +autofs -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c15 --- Comment #15 from Dr. Werner Fink 2014-02-20 13:55:16 UTC --- (In reply to comment #14) Guess: an automated update had happen -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c17 --- Comment #17 from Rainer Krienke 2014-02-20 15:28:42 UTC --- The "Before" line you mentioned is there. I also veryfied the versions of systemd and nfs-client and did a rpm -q --verify on the rpms to see if anything might have changed because of any updates or manual interventions by myself. But nothing. Next I did the following which I think shows the base of the problem: $ systemctl list-dependencies --after autofs autofs.service ├─nfs.service ├─system.slice ├─systemd-journald.socket ├─ypbind.service ├─basic.target ... The output shows to my knowledge which services are started *after* autofs has been started. Service nfs is in this list however it should have been started *before* autofs. In /etc/init.d/nfs the LFS comments are correct as far as I can see: ### BEGIN INIT INFO # Provides: nfs # Required-Start: $network $portmap # Required-Stop: $network $portmap # X-Start-Before: +autofs # X-Start-Before: autofs # Default-Start: 3 5 # Default-Stop: 0 1 2 6 # Short-Description: NFS client services # Description: All necessary services for NFS clients # X-Systemd-RemainAfterExit: true ### END INIT INFO So why is nfs started after autofs and not before? Any idea? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- Blocks| |865018 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c20 --- Comment #20 from Andrey Borzenkov 2014-02-21 09:38:19 UTC --- (In reply to comment #19)

The difference to the successful test you performed using a nfs entry in fstab and our setup is perhaps the higher complexity of our solution with more services involved. Our NFS mounts are done only via automounter, there are no NFS entries in fstab.

Yes. The problem is, if no filesystem that requires network is present in /etc/fstab, network-online.target IS NOT STARTED at all, so no synchronization point exists during shutdown. That is the weakest link here. The reasoning is - if nothing needs network, do not delay startup needlessly. But this contradicts "event based" systemd paradigm - we never know when network comes and goes. Currently if we unconditionally enable network-online.target this will pull in NetworkManager-wait-online.service for those that are using NM. By default it does not wait for anything, so it could be acceptable. Please try adding Wants: network-online.target to network@.service. This will ensure that when ifup is started it also starts network-online.target. NFS mounts should get dependency on it automatically. For ifup it has zero cost. You will need to restart network of course. But please consider that it is not strictly speaking systemd issue. Even assuming that all needed units are started - what happens when network *does* fail? There should be generic solution that covers situation of remote server being unavailable. But of course we need to fix the case of orderly shutdown. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c22 Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- InfoProvider|nfbrown@suse.com |krienke@uni-koblenz.de --- Comment #22 from Dr. Werner Fink 2014-02-21 16:32:09 UTC --- @ Rainer : Please add a line Wants=network-online.target before the line Before=network-online.target in /usr/lib/systemd/system/network@.service ... @ Andrey : Beside this ... ``not strictly speaking systemd issue'' ... I know that I've done a lot of shell code for SysVinit to avoid such NFS hangs at shutdown even if the network is down. Also it was possible to close any remote sessions before network was down. And both had worked. The question rises: does systemd has the possibility to avoid such hangs? Also it is very simple to let systemd hang at boot if one choose a conflicting dependencies ... there is a warning but then nothing happens where I expect a emergency shell. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c24 --- Comment #24 from Andrey Borzenkov 2014-02-21 17:44:00 UTC --- (In reply to comment #22)

Beside this ... ``not strictly speaking systemd issue'' ... I know that I've done a lot of shell code for SysVinit to avoid such NFS hangs at shutdown even if the network is down. Also it was possible to close any remote sessions before network was down. And both had worked. The question rises: does systemd has the possibility to avoid such hangs?

At least in 13.1 we still have the same script behind nfs.service. So probably there are conditions when it does not work? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c26 --- Comment #26 from Rainer Krienke 2014-02-24 10:17:29 UTC --- Hello, first thanks very much to Werner and to Andrey for your proposals over the weekend. This morning I tried the Wants=network-online.target configuration in network@.service. I copied network@.service to /etc/systemd/system and edited this file. Next I tested what happens upon a reboot. The results were not reliable just as it happened earlier. Sometimes I could reboot a couple of times with no problems then I removed the /etc/systemd/system/network@.service file tested again, next put the file in place again (with the additional Wants and Before lines in place ) but now a reboot did suddenly hang. So there is something else wrong and I thought about what is nonstandard in my tweaked installation of opensuse. All desktop installation I do have one thing in common: A non local, nfs mounted /usr/local directory that contains centrally installed software for all linux clients and is also used for some central administration tasks. For this purpose /usr/local on each client is not a directory but a symbolic link to a directory /import/sw. /import/sw again is a key for a automount map that causes a NFS mount of the servers /export/sw directory to /import/sw on the client. So accessing /usr/local triggers a NFS mount via automounter. What I now tried is simply to remove the local symbolic link and to replace it by a the original "local" on disk installation directory (which after all contains only empty subdirectories like sbin, man, ...). In this configuration I was able to reboot the system successfully even if a user with NFS home directory has a running KDE session. It seems to work each and every time I tried, at least up to now it did not fail. I could even remove the modified /etc/systemd/system/network@.service file and it still worked fine. If I replace the directory /usr/local by the symbolic link "local" pointing to the automount mount point the system hung again in several tries. So my thoughts now are that some systemd-service uses /usr/local, at least it accesses this directory and thus triggers a NFS mount, which might lead to a hanging mount when shutting down this service, because the dependencies of this service are not designed to be NFS dependent cause usually /usr/local is not a NFS directory. I'll go on searching in this direction, and see what more I can find out. Rainer -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c28 --- Comment #28 from Rainer Krienke 2014-02-25 09:42:29 UTC --- My idea about services using /usr/local did not prove correct. I removed several services, that might have accessed /usr/local (like: systemd-binfmt, systemd-tmpfiles-clean, ...) but this did not really help. Then I started from scratch with a new suse13.1 installation (done via autoyast, so its the same then the previous installation was) including latest versions of systemd and nfs-client and sysconfig-network. I then only copied network@.service in /etc/systemd/system and tried to remove the "IgnoreOnIsolate=yes" option from the Units configuration. I also simply tried and removed /etc/init.d/{autofs, ypbind} since they shouldn't be used anyway because there are corresponding .service files. Then I rebooted the system and afterwards everything seemd fine. I could shut down the system without any hangs. I did this about 10 times, shut down from kdm, as well as from a running kde session. Worked wonderful. Next I tried and restored the autofs and ypbind files into /etc/init.d and rebooted once again. The next reboot from kdm did then hang again. To crosscheck, I deleted the autofs,ypbind init files again, rebooted and tried then again to run a shutdown. Well it hung again. The real problem is that this bug shows a kind of "random" behaviour. Sometimes it works sometimes it doesn't and it is not understandable to me why and makes debugging quite hard. You never know if any change you make really helps or if you simply had good luck when trying to shut down and next time it won't work again. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c30 --- Comment #30 from Dr. Werner Fink 2014-02-26 13:15:26 UTC --- openSUSE_13.1 i586 succeeded openSUSE_13.1 x86_64 succeeded Now please test this out. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c32 --- Comment #32 from Dr. Werner Fink 2014-02-26 15:11:48 UTC --- Hmmm ... ntpd is also listen on network interfaces linux:~ # systemctl status -l ntp.service ntp.service - LSB: Network time protocol daemon (ntpd) Loaded: loaded (/etc/init.d/ntp) Drop-In: /run/systemd/generator/ntp.service.d `-50-insserv.conf-$time.conf Active: active (running) since Wed 2014-02-26 14:12:16 CET; 1h 47min ago Process: 1743 ExecStart=/etc/init.d/ntp start (code=exited, status=0/SUCCESS) CGroup: /system.slice/ntp.service `-1777 /usr/sbin/ntpd -p /var/run/ntp/ntpd.pid -g -u ntp:ntp -i /var/lib/ntp -c /etc/ntp.conf Feb 26 14:12:11 linux ntpd[1777]: proto: precision = 3.632 usec Feb 26 14:12:11 linux ntpd[1777]: ntp_io: estimated max descriptors: 1024, initial socket boundary: 16 Feb 26 14:12:11 linux ntpd[1777]: Listen and drop on 0 v4wildcard 0.0.0.0 UDP 123 Feb 26 14:12:11 linux ntpd[1777]: Listen and drop on 1 v6wildcard :: UDP 123 Feb 26 14:12:11 linux ntpd[1777]: Listen normally on 2 lo 127.0.0.1 UDP 123 Feb 26 14:12:11 linux ntpd[1777]: Listen normally on 3 enp0s3 10.120.67.7 UDP 123 Feb 26 14:12:11 linux ntpd[1777]: Listen normally on 4 lo ::1 UDP 123 Feb 26 14:12:11 linux ntpd[1777]: Listen normally on 5 enp0s3 2620:113:80c0:8080:a00:27ff:fe8b:cee4 UDP 123 Feb 26 14:12:11 linux ntpd[1777]: Listen normally on 6 enp0s3 2620:113:80c0:8080:987c:1d5c:181c:aeae UDP 123 Feb 26 14:12:11 linux ntpd[1777]: Listen normally on 7 enp0s3 fe80::a00:27ff:fe8b:cee4 UDP 123 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c34 --- Comment #34 from Rainer Krienke 2014-02-26 15:24:51 UTC --- Tried and replaced +autofs by only autofs according to Neils comment. The first 3 tries were fine again, but the next one showed the old behaviour. In this case kde is stopped (console is visible) the splash screen does not even appear and no more shutdown messages appear on the screen, the system hangs. Changing consoles by ALT-<n> is still possible but nothing else. Resetted the system 5 times now, rebooted and shut down again and always got the same behaviour. Probably tomorrow I will try your ntp.list proposal. Do you use "+autofs" or simply "autofs" in your nfs script ? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c36 --- Comment #36 from Dr. Werner Fink 2014-02-27 08:35:11 UTC --- The script shows X-Start-Before: autofs which should be OK. What me worries is the fact that it looks like ntpd is able to stop the shutdown. I've never seen that. The /etc/init.d/ntp includes the LSB head ### BEGIN INIT INFO # Provides: ntp ntpd xntpd # Required-Start: $network $remote_fs $syslog $named # Required-Stop: $network $remote_fs $syslog # Should-Start: network-remotefs # Should-Stop: network-remotefs # Default-Start: 3 5 # Default-Stop: 0 1 2 6 # Short-Description: Network time protocol daemon (ntpd) # Description: Start network time protocol daemon (NTPD). ### END INIT INFO which ensures that ntpd is off before network and any remote NFS share will down. And systemd flags not only NFS but every remote share with the $remote_fs dependency. Are you using a chroot environment for your ntp? That is that NTPD_RUN_CHROOTED in /etc/sysconfig/ntp is set to "yes" ... How does your /etc/sysconfig/ntp looks like? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c38 --- Comment #38 from Dr. Werner Fink 2014-02-27 11:40:50 UTC --- (In reply to comment #37) Hmmm ... the scripts /sbin/ifup and /sbin/ifdown may or may not start a daemon service. That is here the script /sbin/ifup starts dhcpcd and avahi-autoipd: werner@linux:~> systemctl status network@enp0s3.service network@enp0s3.service - ifup managed network interface enp0s3 Loaded: loaded (/usr/lib/systemd/system/network@.service; static) Active: active (running) since Wed 2014-02-26 16:08:15 CET; 20h ago Process: 984 ExecStart=/sbin/ifup %i (code=exited, status=0/SUCCESS) CGroup: /system.slice/system-network.slice/network@enp0s3.service ├─1566 avahi-autoipd: [enp0s3] sleeping ├─1567 avahi-autoipd: [enp0s3] callout dispatcher └─1649 /sbin/dhcpcd --netconfig -L -E -c /etc/sysconfig/network/scripts/dhcpcd-hook -t 0 -h linux enp0s3 Now the question rises: ¿ What does the network@enp0s3.service start on your system ? please show the output of the status command for your system. That is run route and choose the interface of your network card and then run systemctl status network@<Iface>.service with <Iface> replaced with the real interface name. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c40 Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |systemd-maintainers@suse.de AssignedTo|systemd-maintainers@suse.de |mt@suse.com --- Comment #40 from Dr. Werner Fink 2014-02-27 12:59:34 UTC --- I'm out of scope :( Nevertheless this *is* a network issue -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Marius Tomaschewski changed: What |Removed |Added ---------------------------------------------------------------------------- Depends on| |853845 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c42 --- Comment #42 from Marius Tomaschewski 2014-02-27 15:19:21 UTC --- Can you modify /etc/sysconfig/network/scripts/extradebug: - test -d $RUN_FILES_BASE && dir=$RUN_FILES_BASE || dir=/ + dir=/ set DEBUG=EXTRA in /etc/sysconfig/network/config reboot and on after boot provide all the $dir/exdeb.* files? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c44 --- Comment #44 from Rainer Krienke 2014-02-28 08:13:36 UTC --- @Marius attached you find the exdeb debug files you requested. Before the test I removed avahi-autoipd from the system Rainer -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c46 --- Comment #46 from Rainer Krienke 2014-02-28 08:15:36 UTC --- Created an attachment (id=580469) --> (http://bugzilla.novell.com/attachment.cgi?id=580469) exdeb debuf info file -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c48 --- Comment #48 from Rainer Krienke 2014-02-28 08:16:13 UTC --- Created an attachment (id=580470) --> (http://bugzilla.novell.com/attachment.cgi?id=580470) exdeb debuf info file -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c50 --- Comment #50 from Rainer Krienke 2014-02-28 08:17:05 UTC --- Created an attachment (id=580472) --> (http://bugzilla.novell.com/attachment.cgi?id=580472) exdeb debug info file -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c52 Rainer Krienke changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #580468|0 |1 is obsolete| | Attachment #580469|0 |1 is obsolete| | Attachment #580470|0 |1 is obsolete| | Attachment #580471|0 |1 is obsolete| | Attachment #580472|0 |1 is obsolete| | Attachment #580473|0 |1 is obsolete| | --- Comment #52 from Rainer Krienke 2014-02-28 08:19:58 UTC --- Created an attachment (id=580474) --> (http://bugzilla.novell.com/attachment.cgi?id=580474) exdeb.ifdown_1929.1801._.enp2s0.-o.dhcp -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c54 --- Comment #54 from Rainer Krienke 2014-02-28 08:20:46 UTC --- Created an attachment (id=580477) --> (http://bugzilla.novell.com/attachment.cgi?id=580477) exdeb.ifup-dhcp_1051.952._.enp2s0.enp2s0 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c56 --- Comment #56 from Rainer Krienke 2014-02-28 08:22:37 UTC --- Created an attachment (id=580480) --> (http://bugzilla.novell.com/attachment.cgi?id=580480) exdeb.ifup_392.293.1964_.enp2s0.-o.hotplug -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c57 Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #580470|application/octet-stream |text/plain mime type| | --- Comment #57 from Rainer Krienke 2014-02-28 08:22:57 UTC --- Created an attachment (id=580481) --> (http://bugzilla.novell.com/attachment.cgi?id=580481) exdeb.ifup_779.455._.lo.-o.rc.onboot -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c58 --- Comment #58 from Rainer Krienke 2014-02-28 08:23:44 UTC --- Created an attachment (id=580482) --> (http://bugzilla.novell.com/attachment.cgi?id=580482) xdeb.ifup_924.455._.enp2s0.-o.rc.onboot -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #580474|application/octet-stream |text/plain mime type| | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #580476|application/octet-stream |text/plain mime type| | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #580478|application/octet-stream |text/plain mime type| | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #580481|application/octet-stream |text/plain mime type| | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #580484|application/octet-stream |text/plain mime type| | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Dr. Werner Fink changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #580482|application/octet-stream |text/plain mime type| | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c62 --- Comment #62 from Dr. Werner Fink 2014-02-28 13:37:25 UTC --- IMHO the order at boot should be something like network `-> remote-fs-pre.target `-> { nfs.service, autofs.service, ... } `-> remote-fs.target and at shutdown in reverse order ... but with the Before= done by the insserv generator the nfs.service will be listed Before remote-fs-pre.target in the above chain which may interfere with network -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c64 --- Comment #64 from Andrey Borzenkov 2014-02-28 13:47:02 UTC --- (In reply to comment #62) To add ro Fred reply. All remote mounts are done *after* remote-fs-pre and *before* remote-fs. So NFS service must come *before* remote-fs-pre to ensure that NFS is started when remote mounts are done. Starting NFS service after remote-fs-pre will start it concurrently with NFS mounts leading to races and intermittent failures to mount. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c66 --- Comment #66 from Frederic Crozat 2014-02-28 15:20:09 UTC --- hmm, /etc/init.d/nfs shouldn't try to do the mounting itself (this is probably a leftover of the sysvinit days) but let systemd handles it. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c68 --- Comment #68 from Dr. Werner Fink 2014-03-04 11:38:03 UTC --- (In reply to comment #27) After the discussion in the last few comments this is true but it does not harm. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c70 --- Comment #70 from Rainer Krienke 2014-03-10 09:45:28 UTC --- Just for your interest: I tried nfs-client 1.2.8-131.1 from factory and systemd 210-573.2 Still the system does not shut down (or only after an hour). If there is anything else I could try, just let me know. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=861489 https://bugzilla.novell.com/show_bug.cgi?id=861489#c Ludwig Nussel changed: What |Removed |Added ---------------------------------------------------------------------------- Blocks|865018 | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=861489 http://bugzilla.novell.com/show_bug.cgi?id=861489#c74 --- Comment #74 from Rainer Krienke ---

you may give the last systemd from

http://download.opensuse.org/repositories/Base:/System/openSUSE_13.1/

a try as I've found and removes a nasty race between automounted home directories and the user services systemd-exit.service below

/usr/lib/systemd/user/systemd-exit.service

which tries to stop the systemd user manager process started at login for the session handling by using a chroot into the not existing user home before sending the signal to stop manager process

Thanks a lot Werner. In between we are only using openSuSE13.2 installations with NFS and automounter. In 13.2 the shutdown problem sometimes still shows up, but most of the time it simply works. Compared to 13.1 13.2 is much more reliable in this respect and in general. Rainer -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=861489 Swamp Workflow Management changed: What |Removed |Added ---------------------------------------------------------------------------- Whiteboard| |obs:running:4416:moderate -- You are receiving this mail because: You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=861489 http://bugzilla.novell.com/show_bug.cgi?id=861489#c75 --- Comment #75 from Swamp Workflow Management --- openSUSE-RU-2016:0320-1: An update that has 146 recommended fixes can now be installed. Category: recommended (moderate) Bug References: 737690,742774,750845,818044,838475,841544,849870,852015,852021,852232,853293,854884,856389,856392,856858,857204,858864,859072,859365,860574,860937,861316,861489,863217,864745,864904,865834,866732,866933,867128,867663,867664,867840,868019,868230,868439,868931,869142,869603,872929,873432,873444,874665,875502,876587,876694,877021,877674,878525,880438,880732,881125,881559,881942,882393,882714,883565,884271,884403,885232,885288,886211,886599,886852,888178,888215,888612,889297,889357,890977,892096,892162,892300,893797,895087,896664,897799,897801,897803,898233,898240,898432,900558,901481,902240,902901,903009,903963,904214,904517,904828,905550,906709,906900,907318,907393,908476,909358,910643,911347,912030,912334,913517,916420,918118,919095,920195,921831,921898,921920,926169,927250,927457,928265,931388,932284,933365,933512,933521,933533,934077,934901,937512,937900,938908,939571,940264,941576,944132,944799,945282,947212,948458,948555,948705,949574,949683,949739,950510,951265,951663,953241,9543 36,954781,955635,961576 CVE References: Sources used: openSUSE 13.1 (src): systemd-210-40.1, systemd-mini-210-40.1 -- You are receiving this mail because: You are on the CC list for the bug.