December 2018 - openSUSE Users - openSUSE Mailing Lists

[opensuse] Anyway to preserve currently installed packages with 'zypper clean (cc)'
by David C. Rankin 20 Dec '18

20 Dec '18

All, Is there any way to use a zypper or rpm command to clean the current package and metadata cache of all packages EXCEPT the currently installed working set of packages? From the man page, it looks like you either clean all, packages or metadata or you are out of luck. For years I've used a script to move older packages out of the package cache (off the root partition) and over to a directory under /home while leaving the currently installed packages in /var/cache/zypp/packages. For example under /home/admin/dups I have: 22:08 wizard:/home/admin/dups> tree -d . ├── adobe ├── cinelerra ├── gnomeapps │ ├── noarch │ └── x86_64 ├── kde │ ├── noarch │ └── x86_64 ├── mozilla │ └── x86_64 ├── nvidia │ └── x86_64 ├── packman │ ├── Essentials │ │ └── x86_64 │ ├── Extra │ │ └── x86_64 │ └── Multimedia │ └── x86_64 ├── publish │ ├── noarch │ └── x86_64 ├── update │ ├── noarch │ └── x86_64 ├── videolan │ ├── noarch │ └── x86_64 ├── x11utl │ └── x86_64 └── x11wm ├── noarch └── x86_64 Which after a few days/weeks of insuring there are no surprises with the new packages, I can just delete. The remaining packages in /var/cache/zypp/packages can then be rsynced to a local update repo, etc.. (not nearly as important with broadband as it was on dialup...) The point being, I always like to keep a "working set" of rpms. Is there a way that zypper or rpm can do the purge all but the current rpms from the cache? If not, possible feature? Arch does this with 'pacman -Sc', which is quite convenient. (providing 'pacman -Scc' to remove all) It seems like zypper should be able to do this too. Or, am I just stuck with the scripts? -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

4 4

[opensuse] apparmor / nfs: RPC call returned error 13
by Per Jessen 19 Dec '18

19 Dec '18

I have just 30mins ago updated a webserver which is on Leap15 on nfs root on real iron. Now I'm seeing an increasing number of $SUBJ. Kernel is the latest, 4.12.14-lp150.12.28-default. A simple thing such as : # less srv003057/logs/access-log-20181206.gz /usr/bin/lessopen.sh: line 31: mktemp: command not found "srv003057/logs/access-log-20181206.gz" may be a binary file. See it anyway? or # less srv003057/logs/access-log-20181206.gz /usr/bin/lessopen.sh: line 14: /usr/bin/grep: Permission denied /usr/bin/lessopen.sh: line 31: mktemp: command not found grep and mktemp are both present, but $SUBJ gets in the way. The NFS server was not changed, it's quite ancient. It looks like this happens with kernel 4.12.14-lp150.12.28 and 4.12.14-lp150.12.25, but with 4.12.14-lp150.12.16 there is no problem. Apparmor ? Yes indeed - type=AVC msg=audit(1545211515.082:140): apparmor="DENIED" operation="sendmsg" profile="/usr/bin/lessopen.sh" pid=4077 comm="lessopen.sh" laddr=10.42.8.240 lport=980 faddr=10.42.8.254 fport=2049 family="inet" sock_type="stream" protocol=6 type=AVC msg=audit(1545211515.082:141): apparmor="DENIED" operation="sendmsg" profile="/usr/bin/lessopen.sh" pid=4077 comm="lessopen.sh" laddr=10.42.8.240 lport=980 faddr=10.42.8.254 fport=2049 family="inet" sock_type="stream" protocol=6 type=AVC msg=audit(1545210738.216:264): apparmor="DENIED" operation="sendmsg" profile="/usr/bin/lessopen.sh" pid=5074 comm="grep" laddr=10.42.8.240 lport=795 faddr=10.42.8.254 fport=2049 family="inet" sock_type="stream" protocol=6 type=AVC msg=audit(1545210738.216:265): apparmor="DENIED" operation="sendmsg" profile="/usr/bin/lessopen.sh" pid=5074 comm="grep" laddr=10.42.8.240 lport=795 faddr=10.42.8.254 fport=2049 family="inet" sock_type="stream" protocol=6 Just to be complete, a few of these too: type=AVC msg=audit(1545209710.736:49): apparmor="DENIED" operation="open" profile="/usr/sbin/nscd" name="/etc/my.cnf" pid=1068 comm="nscd" requested_mask="r" denied_mask="r" fsuid=0 ouid=0 type=AVC msg=audit(1545209710.740:50): apparmor="DENIED" operation="open" profile="/usr/sbin/nscd" name="/etc/my.cnf" pid=1068 comm="nscd" requested_mask="r" denied_mask="r" fsuid=0 ouid=0 I honestly can't think of why nscd should want to read /etc/my.cnf. I guess this "sendmsg" restriction is new in the most recent kernels? -- Per Jessen, Zürich (2.9°C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

1 1

[opensuse] apparmor - any reason nscd should be reading /etc/my.cnf ?
by Per Jessen 19 Dec '18

19 Dec '18

Seen on leap15 with latest kernel: type=AVC msg=audit(1545212627.324:49): apparmor="DENIED" operation="open" profile="/usr/sbin/nscd" name="/etc/my.cnf" pid=1019 comm="nscd" requested_mask="r" denied_mask="r" fsuid=0 ouid=0 type=AVC msg=audit(1545212627.324:50): apparmor="DENIED" operation="open" profile="/usr/sbin/nscd" name="/etc/my.cnf" pid=1019 comm="nscd" requested_mask="r" denied_mask="r" fsuid=0 ouid=0 type=AVC msg=audit(1545212644.324:113): apparmor="DENIED" operation="open" profile="/usr/sbin/nscd" name="/etc/my.cnf" pid=1019 comm="nscd" requested_mask="r" denied_mask="r" fsuid=477 ouid=0 -- Per Jessen, Zürich (3.7°C) http://www.cloudsuisse.com/ - your owncloud, hosted in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

1 0

[opensuse] how to renew a dhcp lease with wicked?
by Per Jessen 19 Dec '18

19 Dec '18

I was just doing some tests wrt the questions Bjoern asked, and I cannot figure out how to renew/refresh a dhcp lease with wicked? -- Per Jessen, Zürich (0.2°C) http://www.dns24.ch/ - your free DNS host, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

3 3

[opensuse] has anyone used lvmcache on leap15 ?
by Per Jessen 18 Dec '18

18 Dec '18

I have created the cache LV and the metadata LV and created a cache pool. When I try to associate the new cachepool with the LV: # lvconvert --type cache --cachepool /dev/Group1/music_cache /dev/Group1/music Do you want wipe existing metadata of cache pool Group1/music_cache? [y/n]: y device-mapper: reload ioctl on (253:1) failed: Invalid argument Failed to lock logical volume Group1/music. >From dmesg: device-mapper: ioctl: can't change device type after initial table load. This is a stand-alone storage server, just in the process of being set up. -- Per Jessen, Zürich (5.1°C) http://www.dns24.ch/ - free dynamic DNS, made in Switzerland. -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

1 2

[opensuse] Using open source projects to spread malware
by Antonio Ojea 17 Dec '18

17 Dec '18

Hi all, this is just a heads up about a new social-engineering vector for malware using open source projects, more details in the following article [1] As the article mention: >This is apparently what happened to event-stream, a widely used tool that was compromised by a crypto-currency stealing attacker who gained commit access, >poisoned an update, and then locked the project's owner out. > I don't know what to say. #116 [Dominic Tarr/Github] [2] [1] https://boingboing.net/2018/11/26/candy-from-strangers.html [2] https://github.com/dominictarr/event-stream/issues/116 -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

1 0

[opensuse] Corrupted video display (intel graphics)
by Carlos E. R. 16 Dec '18

16 Dec '18

Hi, I have another situation with my laptop: unbearable video flickering (Intel) occasionally (but fatal). It happens sometimes after using kodi, sometimes while using kodi in window mode, sometimes it starts later. Once it starts it gets worse and worse till I have to reboot. It starts by the video flickering on the top quarter of the window. If I maximize kodi it stops. Go back to window mode, continues. Then it starts to affect other windows on the system. Yesterday, I used kodi for some minutes, then stopped it, then in Chrome I started to view a video on Amazon Prime Video, and it was heavily affected, but also in window mode, not in full screen. If I casted the display to my Chromecast, this was not affected, but video mode degraded to low res. Then I noticed that "Web Content", the children from firefox (all of them) were at 100% cpu, so I existed firefox, then killed the children. Then the xfce terminals began to have terrible flickering and distortion, it was impossible to read logs. I was trying to see if there was a GPU reset event. Looking now, there wasn't this time. The machine is running now leap 15.0. There was no problem with 42.3 a month ago. minas-tirith:~ # hwinfo --gfxcard 16: PCI 02.0: 0300 VGA compatible controller (VGA) [Created at pci.378] Unique ID: _Znp.rCqQQ9NkN2D SysFS ID: /devices/pci0000:00/0000:00:02.0 SysFS BusID: 0000:00:02.0 Hardware Class: graphics card Model: "Intel Mobile 4 Series Chipset Integrated Graphics Controller" Vendor: pci 0x8086 "Intel Corporation" Device: pci 0x2a42 "Mobile 4 Series Chipset Integrated Graphics Controller" SubVendor: pci 0x103c "Hewlett-Packard Company" SubDevice: pci 0x3069 Revision: 0x07 Driver: "i915" Driver Modules: "i915" Memory Range: 0xd0000000-0xd03fffff (rw,non-prefetchable) Memory Range: 0xc0000000-0xcfffffff (ro,non-prefetchable) I/O Ports: 0x4110-0x4117 (rw) Memory Range: 0x000c0000-0x000dffff (rw,non-prefetchable,disabled) IRQ: 16 (192437 events) I/O Ports: 0x3c0-0x3df (rw) Module Alias: "pci:v00008086d00002A42sv0000103Csd00003069bc03sc00i00" Driver Info #0: Driver Status: i915 is active Driver Activation Cmd: "modprobe i915" Config Status: cfg=no, avail=yes, need=no, active=unknown 26: PCI 02.1: 0380 Display controller [Created at pci.378] Unique ID: ruGf.MfB3K+fOff2 SysFS ID: /devices/pci0000:00/0000:00:02.1 SysFS BusID: 0000:00:02.1 Hardware Class: graphics card Model: "Intel Mobile 4 Series Chipset Integrated Graphics Controller" Vendor: pci 0x8086 "Intel Corporation" Device: pci 0x2a43 "Mobile 4 Series Chipset Integrated Graphics Controller" SubVendor: pci 0x103c "Hewlett-Packard Company" SubDevice: pci 0x3069 Revision: 0x07 Memory Range: 0xd2400000-0xd24fffff (rw,non-prefetchable) Module Alias: "pci:v00008086d00002A43sv0000103Csd00003069bc03sc80i00" Config Status: cfg=no, avail=yes, need=no, active=unknown Primary display adapter: #16 minas-tirith:~ # minas-tirith:~ # hwinfo --monitor 22: None 00.0: 10002 LCD Monitor [Created at monitor.125] Unique ID: rdCR.eLo9pOY7oOA Parent ID: _Znp.rCqQQ9NkN2D Hardware Class: monitor Model: "N156B3-L02 LCD Monitor" Vendor: CMO "N156B3-L02" Device: eisa 0x1558 Resolution: 1366x768@60Hz Size: 344x193 mm Year of Manufacture: 2007 Week of Manufacture: 40 Detailed Timings #0: Resolution: 1366x768 Horizontal: 1366 1382 1416 1466 (+16 +50 +100) -hsync Vertical: 768 770 776 788 (+2 +8 +20) -vsync Frequencies: 69.31 MHz, 47.28 kHz, 60.00 Hz Config Status: cfg=no, avail=yes, need=no, active=unknown Attached to: #10 (VGA compatible controller) minas-tirith:~ # -- Cheers / Saludos, Carlos E. R. (from 42.3 x86_64 "Malachite" at Telcontar)

4 9

[opensuse] identifying the toolkit an application uses
by Istvan Gabor 16 Dec '18

16 Dec '18

Hello: I don't know if the subject term correctly describes what I want. How can I identify the toolkit an application uses? For example how can I find out whether my pulseaudio or audacious program uses gt2 or gtk3, or my vlc uses qt2/qt3/qt4? For example if I want to change the look of pulseaudio sliders I have to know which theme to modify, gtk2/gtk3 etc. Thanks, Istvan -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

4 3

[opensuse] Q: Unusable Teamviewer V 13 & 14
by Andrei Verovski 14 Dec '18

14 Dec '18

Hi, I have almost unusable Teamviewer on a number of SuSE Leap 15.0 and Tumblweed PCs with MATE desktop. It freezes connections on random intervals (usually very short). As far as I remember, same happens with KDE. Disabled UDP connections as suggested by someone on the forum, didn't help. Tried versions 13 and 14, none work reliably. Tested different configurations, its seems happens ONLY when connecting to another SuSE desktop, in other situation, for example connecting from SuSE to Windows everything is fine. This is somehow related to SuSE, because with Debian 9 & MATE this problem doesn't exists. With old Teamviewer everything worked, unfortunately, Teamviewer 11 and older are incompatible with Leap starting from 42.3 or 42.2, I don't remember exactly (TW GUI won't start at all). Anyone have any clue how to fix this? Thanks in advance. Andrei ******************** This is the last Leap 15.0 system (HP Proliant DL380) where Teamviewer behaves absolutely horrible: hwinfo --gfxcard 09: PCI 103.0: 0300 VGA compatible controller (VGA) [Created at pci.378] Unique ID: kX1R.o_JrRNWqxE3 Parent ID: 6NW+.RcDfn3vgRg7 SysFS ID: /devices/pci0000:00/0000:00:1e.0/0000:01:03.0 SysFS BusID: 0000:01:03.0 Hardware Class: graphics card Model: "ATI ES1000 515E" Vendor: pci 0x1002 "ATI Technologies Inc" Device: pci 0x515e "ES1000 515E" SubVendor: pci 0x103c "Hewlett-Packard Company" SubDevice: pci 0x31fb Revision: 0x02 Driver: "radeon" Driver Modules: "radeon" Memory Range: 0xe8000000-0xefffffff (ro,non-prefetchable) I/O Ports: 0x2000-0x2fff (rw) Memory Range: 0xf1cf0000-0xf1cfffff (rw,non-prefetchable) Memory Range: 0x000c0000-0x000dffff (rw,non-prefetchable,disabled) IRQ: 23 (no events) I/O Ports: 0x3c0-0x3df (rw) Module Alias: "pci:v00001002d0000515Esv0000103Csd000031FBbc03sc00i00" Driver Info #0: XFree86 v4 Server Module: radeon Config Status: cfg=no, avail=yes, need -- To unsubscribe, e-mail: opensuse+unsubscribe(a)opensuse.org To contact the owner, e-mail: opensuse+owner(a)opensuse.org

2 1

[opensuse] Plymouth animation with Nvidia's early KMS
by Oleksii Vilchanskyi 13 Dec '18

13 Dec '18

Out-of-the box after the installation of > nvidia-gfxG05-kmp-default | NVIDIA graphics driver kernel module for GeForce 400 series and newerthe Plymouth boot animation doesn't work. That package owns two important files, > % cat /etc/modprobe.d/nvidia.conf > options nvidia_drm modeset=1 and > % cat /etc/dracut.conf.d/50-nvidia-default.conf > install_items+=" /usr/bin/chmod /usr/bin/mknod /usr/bin/cat /usr/bin/echo /usr/bin/chown " > add_drivers+="nvidia nvidia-drm nvidia-modeset nvidia-uvm" > omit_dracutmodules+="plymouth" So I proceed to enable early KMS via the kernel parameter, since the required dracut modules are already specified: > % cat /etc/modprobe.d/nvidia.conf > options nvidia_drm modeset=1 All's fine, KMS works, but the > omit_dracutmodules+="plymouth" line obviously prevents the plymouth animation. If I do (notice the 51-*.conf, so should(?) override 50-*.conf): > % cat /etc/dracut.conf.d/51-nvidia-default.conf > add_dracutmodules+="plymouth" then dracut -f still says that the module will be omitted. So, I ended up commenting this line out from 50-*.conf, and plymouth works just fine. Questions: 1. Why is plymouth disabled by default? Since the required nvidia modules are already added by this config, what can prevent plymouth from working? Aside from the nvidia-drm.modeset=1 kernel parameter? 2. How do I re-add dracut modules after they were omitted (instead of touching the default config)? I would like to use a separate file, because I assume this one gets overridden with every package upgrade, and it's generally sane to separate custom configuration from vendor configuration. -- Thank you, Alex

2 2