I also think it's about networking issues. Maybe some caches or tmp directories hold bad / wrong information, that has happened here several times over the last years. 1) Maybe I missed it, but which display-manager are you using? Kdm? SDDM? Try switching to the simplest possible and see if the problem persists. 2) Does the problem also occur when you start a single X session from a console? Try (as root) service display-manager stop (kills all X sessions!) Then login as unpriv. user and type startx and see what happens Repeat the same as root (I know, but for testing permission problems that's helpful). Am Montag, 15. Juli 2019, 10:10:50 CEST schrieb Roger Oberholtzer:

I have a strange thing in KDE with CIFS as a client. If a share is not unmounted, but the server shuts down, large parts of KDE hang for a long time (e.g., the status bar at the bottom). I think it is trying to monitor the directories that are mounted. And when the server goes away, the monitoring hangs. Could your KDE issues be because the NFS server is not reachable? Or at least access to the share hangs, causing parts of KDE to hang as well? If is hanging because of an error on the server or client is a different matter. But it could explain the KDE part of your problem.

On Mon, Jul 15, 2019 at 9:58 AM wrote:

...

Hi Together,

since some month I've a lot of trouble with my about 20 Years old Home network, which was all running on Tumbleweed:

Server (NFS 4 Server which is sharing /home config via YaST): |- Client1 (/home mounted via & user via NIS) |- Client2 (/home mounted via & user via NIS)

...

|- Client5 (/home mounted via & user via NIS)

Remark: Server and Client user are synced via NIS/YP configurated via YaST at Server and Client (basic setup was before about 20 years - last installation due to a root FS crash before about 1 year based on the Tumbleweed installation ISO).

On the most Clients KDE Plasma Crashes many times after login

If they are starting - it takes about 10 up to 30 minutes!!!! Afterwords the system hangs from time to time with normal load and only the tracker extract info below. What can I do? One client I installed the last ISO (from 8th July) but this one don't mount /home well, so I need to unmount it and mount again via:

umount -f /home ; systemctl start home.mount ; systemctl status home.mount

# cat /etc/fstab (extract) 192.168.2.1:/mnt/nfs4/home /home nfs defaults 0 0

# cat /etc/fstab (extract - also tested with nfs4) 192.168.2.1:/mnt/nfs4/home /home nfs4 defaults 0 0

# journalctl -xa (extract) Jul 13 10:33:23 client1 tracker-extract[6746]: If the error above is recurrent for the same item/ID, consider running "tracker-extract" in the terminal with the TRACKER_VERBOSITY=3 environment variable, and filing a bug with the additional information Jul 13 10:33:23 client1 tracker-extract[6746]: Could not insert metadata for item "file:///usr/share/applications/YaST2/sysconfig.desktop": Subject `urn:uuid:xyz' is not in domain `nie:DataObject' of property `nie:dataSource'

If I call tracker manually, there will be the following output: # tracker extract -v detailed /usr/share/applications/YaST2/groups/ ** Message: 19:36:49.371: Starting tracker-extract 2.2.2 ** Message: 19:36:49.371: General options: ** Message: 19:36:49.371: Verbosity ............................ 2 ** Message: 19:36:49.371: Sched Idle ........................... 1 (tracker-extract:4939): dconf-DEBUG: 19:36:49.371: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) ** Message: 19:36:49.371: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'inode/directory' (from GIO) file:///usr/share/applications/YaST2/groups: Es wurden keine Metadaten gefunden oder keine Entdecker, die mit dieser Datei umgehen können ulfi@baldo:~> tracker extract -v detailed /usr/share/applications/YaST2/groups/* ** Message: 19:36:57.536: Starting tracker-extract 2.2.2 ** Message: 19:36:57.536: General options: ** Message: 19:36:57.536: Verbosity ............................ 2 ** Message: 19:36:57.536: Sched Idle ........................... 1 ** Message: 19:36:57.536: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 (tracker-extract:4955): dconf-DEBUG: 19:36:57.537: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST nie:title "X-SuSE-YaST" ;

a nfo:SoftwareCategory .

urn:theme-icon:yast-hardware-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/hardware.desktop nie:title "Hardware" ;> nfo:softwareCmdLine "/sbin/yast2" ; nie:isLogicalPartOf urn:software-category:Qt , urn:software-category:X-SuSE-YaST ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-hardware-group .

** Message: 19:36:57.554: Starting tracker-extract 2.2.2 ** Message: 19:36:57.554: General options: ** Message: 19:36:57.554: Verbosity ............................ 2 ** Message: 19:36:57.554: Sched Idle ........................... 1 (tracker-extract:4969): dconf-DEBUG: 19:36:57.554: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) ** Message: 19:36:57.554: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST nie:title "X-SuSE-YaST" ;

a nfo:SoftwareCategory .

urn:theme-icon:yast-high_availability-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/high_availability.desktop nie:title "Hochverfügbarkeit" ;> nfo:softwareCmdLine "/sbin/yast2" ; nie:isLogicalPartOf urn:software-category:Qt , urn:software-category:X-SuSE-YaST ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-high_availability-group .

** Message: 19:36:57.569: Starting tracker-extract 2.2.2 ** Message: 19:36:57.569: General options: ** Message: 19:36:57.570: Verbosity ............................ 2 ** Message: 19:36:57.570: Sched Idle ........................... 1 ** Message: 19:36:57.570: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 Loading extractor rules... (/usr/share/tracker-miners/extract-rules) (tracker-extract:4983): dconf-DEBUG: 19:36:57.570: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST nie:title "X-SuSE-YaST" ;

a nfo:SoftwareCategory .

urn:theme-icon:yast-misc-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/misc.desktop nie:title "Verschiedenes" ;> nfo:softwareCmdLine "/sbin/yast2" ; nie:isLogicalPartOf urn:software-category:Qt , urn:software-category:X-SuSE-YaST ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-misc-group .

** Message: 19:36:57.588: Starting tracker-extract 2.2.2 (tracker-extract:4997): dconf-DEBUG: 19:36:57.588: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) ** Message: 19:36:57.588: General options: ** Message: 19:36:57.588: Verbosity ............................ 2 ** Message: 19:36:57.588: Sched Idle ........................... 1 ** Message: 19:36:57.588: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST nie:title "X-SuSE-YaST" ;

a nfo:SoftwareCategory .

urn:theme-icon:yast-network-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/network_services.desktop nie:title "Netzwerkdienste" ;> nfo:softwareCmdLine "/sbin/yast2" ; nie:isLogicalPartOf urn:software-category:Qt , urn:software-category:X-SuSE-YaST ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-network-group .

** Message: 19:36:57.604: Starting tracker-extract 2.2.2 ** Message: 19:36:57.604: General options: ** Message: 19:36:57.604: Verbosity ............................ 2 ** Message: 19:36:57.604: Sched Idle ........................... 1 ** Message: 19:36:57.604: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 (tracker-extract:5011): dconf-DEBUG: 19:36:57.604: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST nie:title "X-SuSE-YaST" ;

a nfo:SoftwareCategory .

urn:theme-icon:yast-security-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/security.desktop nie:title "Sicherheit und Benutzer" ;> nfo:softwareCmdLine "/sbin/yast2" ; nie:isLogicalPartOf urn:software-category:Qt , urn:software-category:X-SuSE-YaST ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-security-group .

(tracker-extract:5025): dconf-DEBUG: 19:36:57.621: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) ** Message: 19:36:57.621: Starting tracker-extract 2.2.2 ** Message: 19:36:57.621: General options: ** Message: 19:36:57.621: Verbosity ............................ 2 ** Message: 19:36:57.621: Sched Idle ........................... 1 ** Message: 19:36:57.621: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST nie:title "X-SuSE-YaST" ;

a nfo:SoftwareCategory .

urn:theme-icon:yast-software-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/software.desktop nie:title "Software" ;> nfo:softwareCmdLine "/sbin/yast2" ; nie:isLogicalPartOf urn:software-category:Qt , urn:software-category:X-SuSE-YaST ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-software-group .

** Message: 19:36:57.637: Starting tracker-extract 2.2.2 (tracker-extract:5039): dconf-DEBUG: 19:36:57.637: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) ** Message: 19:36:57.637: General options: ** Message: 19:36:57.637: Verbosity ............................ 2 ** Message: 19:36:57.637: Sched Idle ........................... 1 ** Message: 19:36:57.637: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST nie:title "X-SuSE-YaST" ;

a nfo:SoftwareCategory .

urn:theme-icon:yast-support-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/support.desktop nie:title "Unterstützung" ;> nfo:softwareCmdLine "/sbin/yast2" ; nie:isLogicalPartOf urn:software-category:Qt , urn:software-category:X-SuSE-YaST ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-support-group .

** Message: 19:36:57.656: Starting tracker-extract 2.2.2 ** Message: 19:36:57.656: General options: ** Message: 19:36:57.656: Verbosity ............................ 2 ** Message: 19:36:57.656: Sched Idle ........................... 1 ** Message: 19:36:57.656: Max bytes (per file) ................. 1048576 Setting scheduler policy to SCHED_IDLE (tracker-extract:5053): dconf-DEBUG: 19:36:57.656: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) Setting priority nice level to 19 Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST nie:title "X-SuSE-YaST" ;

a nfo:SoftwareCategory .

urn:theme-icon:yast-system-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/system.desktop nie:title "System" ;> nfo:softwareCmdLine "/sbin/yast2" ; nie:isLogicalPartOf urn:software-category:Qt , urn:software-category:X-SuSE-YaST ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-system-group .

** Message: 19:36:57.671: Starting tracker-extract 2.2.2 ** Message: 19:36:57.671: General options: ** Message: 19:36:57.671: Verbosity ............................ 2 ** Message: 19:36:57.671: Sched Idle ........................... 1 ** Message: 19:36:57.671: Max bytes (per file) ................. 1048576 (tracker-extract:5067): dconf-DEBUG: 19:36:57.671: watch_established: "/org/freedesktop/tracker/extract/" (establishing: 1) Setting scheduler policy to SCHED_IDLE Setting priority nice level to 19 Loading extractor rules... (/usr/share/tracker-miners/extract-rules) Extractor rules loaded MIME type guessed as 'application/x-desktop' (from GIO) @prefix nie: http://www.semanticdesktop.org/ontologies/2007/01/19/nie# . @prefix nfo: http://www.semanticdesktop.org/ontologies/2007/03/22/nfo# . @prefix rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# .

urn:software-category:Settings nie:title "Settings" ;

a nfo:SoftwareCategory .

urn:software-category:System nie:title "System" ;

a nfo:SoftwareCategory .

urn:software-category:Qt nie:title "Qt" ;

a nfo:SoftwareCategory .

urn:software-category:X-SuSE-YaST-Virtualization nie:title "X-SuSE-YaST-Virtualization" ;> a nfo:SoftwareCategory .

urn:theme-icon:yast-vm-group a nfo:Image .

file:///usr/share/applications/YaST2/groups/virtualization.desktop nie:title "Virtualisierung" ;> nfo:softwareCmdLine "yast2" ; nie:isLogicalPartOf urn:software-category:Settings , urn:software-category:System , urn:software-category:Qt , urn:software-category:X-SuSE-YaST-Virtualization ; a nfo:SoftwareApplication , nfo:Executable ; nfo:softwareIcon urn:theme-icon:yast-vm-group .

-- Regards Ulf -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

-- Best Regards - Mit freundlichen Grüßen, Markus Feilner ------------------------- _"Zu sagen was ist, bleibt die revolutionärste Tat."_ ------------------------- Leitender Redakteur IX - www.heise.de Karl-Wiechert-Alle 10, 30625 Hannover fon: +49 511 53 52 388 mail: mfe@ix.de PGP: 40A3C306F96133067C11CFD9A958A906268C9F0A http://www.feilner-it.net/files/MFpub.asc Xing: http://www.xing.com/profile/Markus_Feilner LinkedIn: https://www.linkedin.com/in/markusfeilner ------------------------- Mindestinformationen nach §37a HGB: Heise Medien GmbH & Co. KG Registergericht: Amtsgericht Hannover HRA 26709 Persoenlich haftende Gesellschafterin: Heise Medien Geschaeftsfuehrung GmbH Registergericht: Amtsgericht Hannover, HRB 60405 Geschaeftsfuehrer: Ansgar Heise, Dr. Alfons Schraeder

Hi Markus,

Gesendet: Montag, 15. Juli 2019 um 10:40 Uhr; Von: "Markus Feilner"

I also think it's about networking issues. Maybe some caches or tmp directories hold bad / wrong information, that has happened here several times over the last years.

So the issue is with a new installed client (mount of /home and NIS on the client via YaST) very similar. Old ones mount with same /etc/fstab stable new ones hangs and need to unmount and mount manually again before I can start the user! This is OK for me - but my Family can't start a PC anymore with proper working system.

1) Maybe I missed it, but which display-manager are you using? Kdm? SDDM? Try switching to the simplest possible and see if the problem persists.

So it looks like tumbleweed uses sddm if KDE was selected during installation. Inside them I select "Plasma" as desktop environment (if I select "Plasma Wayland" the client hangs anyway hours on a blank screen).

2) Does the problem also occur when you start a single X session from a console? Try (as root) service display-manager stop (kills all X sessions!) Then login as unpriv. user and type startx and see what happens Repeat the same as root (I know, but for testing permission problems that's helpful).

seems that the system hangs as unpriv. user but I will wate . But during parallel using of a second client, I see that the system hangs (nicer a plasma action nor a keyboard entry possible) if I do an action on firefox :-( - I don't know if this relays to the problem but also in journalctl -x I can find kwin_x11[2931]: qt.qpa.xcb: Unhandled client message: "_NET_CURRENT_DESKTOP" ??? which tells me nothing. Regards Ulf -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

On Mon, Jul 15, 2019 at 11:03 AM wrote:

Hi Roger,

...

Gesendet: Montag, 15. Juli 2019 um 10:10 Uhr; Von: "Roger O."

I have a strange thing in KDE with CIFS as a client. If a share is not unmounted, but the server shuts down, large parts of KDE hang for a long time (e.g., the status bar at the bottom). I think it is trying to monitor the directories that are mounted. And when the server goes away, the monitoring hangs. Could your KDE issues be because the NFS server is not reachable? Or at least access to the share hangs, causing parts of KDE to hang as well? If is hanging because of an error on the server or client is a different matter. But it could explain the KDE part of your problem.

You can expect, I checked a lot (about 100h poor working time). Yes in the past, the server was also an issue, but this was fixed since about 2 month. Afterwords it works mainly, but still some crazy issues. But since about update to 5.1 on Server and Clients either the booting hangs on a crashed "akonadictl" or they boots slowly. Since the last update the duration was dramatically increased (since update to new KDE/Plasma 5.16?!) like described below. For me it looks like an issue from update of some configs. Especially akonadi have issues since about 5 Years, to migrate my 20 years old kmail folders.

To evaluate, the server issue, I started the user on the server, but the issue seams to be similar.

We also have the KDE issue on boot. Our CIFS shares are mounted via autofs as defined in /etc/fstab. For example: noauto,x-systemd.automount,x-systemd.idle-timeout=60,nofail,_netdev The interesting thing is that the mounts never go away when not used. I think this is because KDE is checking them. Even when they are not the current directory in any KDE component (that we can locate). When you log in again, KDE remembers the previous directory layout and tries to access the directories. This will hang until the server boots, or some KDE timeout is reached (we think). All of which makes the autofs a bit useless. Ok, the system does not hang waiting. But since KDE does, the system could just as well hang anyway... -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

* ub22@gmx.net [07-15-19 06:56]:

Hi,

...

Gesendet: Montag, 15. Juli 2019 um 11:15 Uhr; Von: "Roger Oberholtzer"

We also have the KDE issue on boot. Our CIFS shares are mounted via autofs as defined in /etc/fstab. For example:

noauto,x-systemd.automount,x-systemd.idle-timeout=60,nofail,_netdev

In the moment at all systems the automounter was disabled.

...

The interesting thing is that the mounts never go away when not used. I think this is because KDE is checking them. Even when they are not the current directory in any KDE component (that we can locate).

When you log in again, KDE remembers the previous directory layout and tries to access the directories. This will hang until the server boots, or some KDE timeout is reached (we think). All of which makes the autofs a bit useless. Ok, the system does not hang waiting. But since KDE does, the system could just as well hang anyway...

Hmm.. I dedicated disabled the automounter again. Look what happens and enable them in a second test to verify.

But the strange thing is, the behavior refers on different user and clients. The whorst one is my own - but with the benefit that the kwallet password entry starts like normal (but entry will not shown and the dialog a closed after some minutes - if password was enter this will be accepted). Entry on the konsole is possible but also is delayed about 1minute.

you could try making your mount with sshfs as (make one line): echo "<passwd>" | sshfs -C -o password_stdin,reconnect,ServerAliveInterval=15,ServerAliveCountMax=3,allow_other <user>@<machine>:/http://en.opensuse.org openSUSE Community Member facebook/ptilopteri Photos: http://wahoo.no-ip.org/piwigo paka @ IRCnet freenode -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

Am Montag, 15. Juli 2019, 16:16:50 CEST schrieb ub22@gmx.net:

Hi Patrick,

...

Gesendet: Montag, 15. Juli 2019 um 13:52 Uhr; Von: "Patrick Shanahan"

I find nfs somewhat qwerky in some respects but sshfs is also qwerky but in other respects and seem to have reached a balance using both.

In the moment I stay with NFS - so it is also used on TV, Smartphone and RasPi

Regards Ulf Smartphone with NFS? What app or which phone OS are you using?

-- Best Regards - Mit freundlichen Grüßen, Markus Feilner ------------------------- _"Zu sagen was ist, bleibt die revolutionärste Tat."_ ------------------------- Leitender Redakteur IX - www.heise.de Karl-Wiechert-Alle 10, 30625 Hannover fon: +49 511 53 52 388 mail: mfe@ix.de PGP: 40A3C306F96133067C11CFD9A958A906268C9F0A http://www.feilner-it.net/files/MFpub.asc Xing: http://www.xing.com/profile/Markus_Feilner LinkedIn: https://www.linkedin.com/in/markusfeilner ------------------------- Mindestinformationen nach §37a HGB: Heise Medien GmbH & Co. KG Registergericht: Amtsgericht Hannover HRA 26709 Persoenlich haftende Gesellschafterin: Heise Medien Geschaeftsfuehrung GmbH Registergericht: Amtsgericht Hannover, HRB 60405 Geschaeftsfuehrer: Ansgar Heise, Dr. Alfons Schraeder

Hi Manfred,

Gesendet: Montag, 15. Juli 2019 um 14:36 Uhr; Von: "Manfred Hollstein"

On Mon, 15 Jul 2019, 12:53:12 +0200, ub22@gmx.net wrote:

...

Hi,

...

Gesendet: Montag, 15. Juli 2019 um 11:15 Uhr; Von: "Roger Oberholtzer"

We also have the KDE issue on boot. Our CIFS shares are mounted via autofs as defined in /etc/fstab. For example:

noauto,x-systemd.automount,x-systemd.idle-timeout=60,nofail,_netdev

In the moment at all systems the automounter was disabled.

you shouldn't, autofs will not solve the issues, but you can make them somewhat more acceptable; keyword is "AUTOFS_OPTIONS="-t 30" (or something like this) in /etc/sysconfig/autofs

OK - done on the main client - if it works fine (must be checked after reboot) - I give you a short feed back tomorrow.

/etc/sysconfig/nfs: ## Path: Network/File systems/NFS server ## Description: Grace time for NFSv4 leases ## Type: integer ## Default: "" # # Set the grace time for the NFSv4 and NLM (for NFSv2 and # NFSv3). New file open requests (NFSv4) and new file locks # (NLM) will not be allowed until after this time has passed # to allow clients to recover state. NFSV4GRACETIME="10"

/etc/nfs.conf.local: [nfsd] grace-time=$NFSV4GRACETIME

/etc/sysctl.d/90-nfs.conf: fs/nfs/nlm_grace_period = 10

NFSV4GRACETIME (new) in /etc/sysconfig/nfs also defaults to 90 seconds, which makes your access to any such NFS share unavailable during that time.

As you can see, I reduce it to 10 seconds here, because I do trust my own network...

Still, rebooting or shutting down a 15.0 or 15.1 system hangs for too long (up to 3 minutes) from time to time... Didn't find the actual reason for it yet, though.

Done, hope it helps Ulf -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

On Mon, 15 Jul 2019, 20:06:29 +0200, ub22@gmx.net wrote:

Update:

...

...

Gesendet: Montag, 15. Juli 2019 um 14:36 Uhr; Von: "Manfred Hollstein"

On Mon, 15 Jul 2019, 12:53:12 +0200, ub22@gmx.net wrote:

...

Hi,

...

Gesendet: Montag, 15. Juli 2019 um 11:15 Uhr; Von: "Roger Oberholtzer"

We also have the KDE issue on boot. Our CIFS shares are mounted via autofs as defined in /etc/fstab. For example:

noauto,x-systemd.automount,x-systemd.idle-timeout=60,nofail,_netdev

In the moment at all systems the automounter was disabled.

you shouldn't, autofs will not solve the issues, but you can make them somewhat more acceptable; keyword is "AUTOFS_OPTIONS="-t 30" (or something like this) in /etc/sysconfig/autofs

OK - done on the main client - if it works fine (must be checked after reboot) - I give you a short feed back tomorrow.

...

/etc/sysconfig/nfs: ## Path: Network/File systems/NFS server ## Description: Grace time for NFSv4 leases ## Type: integer ## Default: "" # # Set the grace time for the NFSv4 and NLM (for NFSv2 and # NFSv3). New file open requests (NFSv4) and new file locks # (NLM) will not be allowed until after this time has passed # to allow clients to recover state. NFSV4GRACETIME="10"

/etc/nfs.conf.local: [nfsd] grace-time=$NFSV4GRACETIME

/etc/sysctl.d/90-nfs.conf: fs/nfs/nlm_grace_period = 10

Seems that this is not helping. The login is without remount like next line not possible.

====== (systemctl status home.mount ; umount -f /home ; systemctl start home.mount ; systemctl status home.mount) ====== ...

Sorry, should have been more clear. In order to actually benefit from using autofs, you need to get rid of these static mounts in /etc/fstab! Enable autofs.service, and then use /net/<hostname>/exported/directory.

Tanks Ulf

HTH, cheers. l8er manfred

On Mon, Jul 15, 2019 at 10:08 PM wrote:

??? So you mean I should remove the entries in the /etc/fstab ?

You have two options in this respect: 1.Remove the items from /etc/fstab and configure autofs. This is the older traditional way. It is more work... 2. Add a bit of text to the existing items in /etc/fstab that you want automounted, and let systemd set it all up. I use method 2. Add the text I put in my first reply in this thread. Add it to the other mount options; noauto,x-systemd.automount,x-systemd.idle-timeout=60,nofail,_netdev There is a command to get systemd to reread the /etc/fstab file after it is edited. Unfortunately I don't recall what it is. So you will need to reboot the client. I know that this works back to openSUSE 42.3. Probably earlier. But that's my experience. You still need to enable autofs for either method. -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

On Tue, Jul 16, 2019 at 10:11 AM wrote:

OK, I've done it like you with methode 2

PS: The issue isn't present anymore with a new user - so I plane to generate new users and move the contend

So maybe autofs causes the mount to be redone as soon as it is detected that it has gone away (by KDE if nothing else). So nothing hangs. Why it goes away is of course a question. But as autofs brings it back before it is missed, the problem gets solved. My problem where the server has gone away is, unfortunately, not helped by autofs. -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

Hi, Everyone I kindly request that you move all future replies to this thread to opensuse-support@opensuse.org, this is clearly not a general development discussion On 16/07/2019 05:38, ub22@gmx.net wrote:

Hi Manfred,

...

Gesendet: Montag, 15. Juli 2019 um 20:25 Uhr; Von: "Manfred Hollstein"

Sorry, should have been more clear. In order to actually benefit from using autofs, you need to get rid of these static mounts in /etc/fstab!

??? So you mean I should remove the entries in the /etc/fstab ?

...

Enable autofs.service, and then use /net/<hostname>/exported/directory.

But how to be configured this?

===== # systemctl status autofs.service ● autofs.service - Automounts filesystems on demand Loaded: loaded (/usr/lib/systemd/system/autofs.service; enabled; vendor preset: disabled) Active: active (running) since Mon 2019-07-15 21:13:51 CEST; 42min ago Main PID: 2394 (automount) Tasks: 3 (limit: 4915) Memory: 51.8M CGroup: /system.slice/autofs.service └─2394 /usr/sbin/automount --systemd-service --dont-check-daemon

Jul 15 21:13:51 baldo systemd[1]: Starting Automounts filesystems on demand... Jul 15 21:13:51 baldo systemd[1]: Started Automounts filesystems on demand. =====

In my case I think you mean to add in ===== /etc/auto.master ===== /net/192.168.1.1/home =====

===== /etc/fstab ===== #192.168.2.1:/mnt/nfs4/home /home nfs defaults 0 0 =====

Thanks Ulf

-- Simon Lees (Simotek) http://simotek.net Emergency Update Team keybase.io/simotek SUSE Linux Adelaide Australia, UTC+10:30 GPG Fingerprint: 5B87 DB9D 88DC F606 E489 CEC5 0922 C246 02F0 014B -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

On 16/07/2019 17:39, ub22@gmx.net wrote:

Hi Simon,

...

Gesendet: Dienstag, 16. Juli 2019 um 00:25 Uhr; Von: "Simon Lees"

I kindly request that you move all future replies to this thread to opensuse-support@opensuse.org, this is clearly not a general development discussion

I forwarded my mail to the Support but no answer till now.

But the main reason to post this here, was the more and more issues on the tumbleweed build which makes working also for a Linux user with 20 years experiance no more possible to still use it. So my network was unchanged since about 15 years, but during the last weeks - the development seems to be going in the wrong direction!

So in the moment in our LUG and the half yearly LPDs I promote Tumbleweed - but this I don't do anymore - so the contious usage of a SoHo Network isn't possible anymore!!

PS: Excuse my but since more than about 20 years with openSUSE I'm a little bit frustated (only once this was before)!

Well regardless of whether it should have been posted here, they have been a bunch of responses and hopefully those people can continue the discussion with you on that more appropriate list -- Simon Lees (Simotek) http://simotek.net Emergency Update Team keybase.io/simotek SUSE Linux Adelaide Australia, UTC+10:30 GPG Fingerprint: 5B87 DB9D 88DC F606 E489 CEC5 0922 C246 02F0 014B -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org

On Mon, Jul 15, 2019 at 5:35 PM AW wrote:

Am Montag, 15. Juli 2019, 14:36:40 CEST schrieb Manfred Hollstein:

...

Still, rebooting or shutting down a 15.0 or 15.1 system hangs for too long (up to 3 minutes) from time to time... Didn't find the actual reason for it yet, though.

Yes, here as well. Leap 15.1, using QT 5.13.

In our use, a file system server for a mounted system goes away. In our mixed environment, we have a main computer running openSUSE, and a number of support ones running Windows 10 (the equipment suppliers only support Windows...). Both machines mount filesystems from each other. The Windows machines are accessed via our own network protocol that manages the data collection on these systems. They do not have a display. When they need to be shut down, an rdesktop session is started, and they are shutdown. This is initiated from within KDE. When the Windows file server goes away (which it will because it is always turned off first), parts of KDE 'freeze'. Every time. Guaranteed. Unmounting the shares is problematic because we use autofs (which makes the system startup much better as the Windows machines are either late in starting or perhaps not powered on - life in the real world). System startup works fine. It is problematic at shutdown because even if one unmounts the file systems, KDE will access them again and they are remounted. Usually rather quickly. So they will probably still be mounted when the Windows server goes away. The ability to use the system when the Windows machines are not powered on is a requirement. So we have set things up accordingly. But it has made system shutdown when the Windows machines are used a pita. I had a thread about this in the General list a while back. We have never solved this KDE behavior. I suspect the consensus will be that KDE is just trying to keep the desktop information up-to-date. I'm fine with that. It just shouldn't freeze when a file system is no longer mounted. Maybe it is stuck in the system call doing the actual access. If so, then that should perhaps be done in a thread that does not effect KDE responsiveness. I do not think this is new KDE behavior. I see it as of 42.3. -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org