[SOLVED][KDE, intel] Screen goes gray when switching windows

Question

Hello,
on my EEE netbook, since the xorg update, I guess --noticed today, as I use it for school only --, the screen goes gray after I switch windows via alt+tab. This does not happen every time, but eventually it occures. In that case the whole screen has color of not yet painted window, and the DE is seemingly not responding to anything, I can still move my mose though and cursor moves too. Switching to text console works, but xserver restart is needed to get things back to normal.
I have been using graphics system raster, but change to default didn't solved the issue. The same goes for -lcd font filters. System is up-to-date and uses some SELinux packages from the AUR if that matters, but it should not, as they are not graphics nor DE related.
Would you have any sugestions? I should not be doing any potentially risky operations till thursday, as I need the computer for school and don't have any other available.
Thank you.
Will happily provide additional information, if you specify it.
EDIT: The situation is simmilar as when I run a program, first a fullscreen window is painted gray, and when everything is loaded its content is painted. This lasts few second in case of big applications like Kontact. In the case of mentioned bug it is like the content of window never painted.
EDIT: So I recompiled bespin-svn, as I remembered, that this package likes to cause trobles, when it is older, and I wanted to mark this solved, as it behaved during the day, but just few minutes ago, the problem happend again.
EDIT: I don't understand it. On monday it happed several times during day. On tuesday it happend once. Since wednesday it didn't happend. Sould I start to beleave in ghosts?
Last edited by Nicky726 (2010-12-09 16:02:37)

Answer 1

Ok, happend again several times today. Didn't happen at any other PC, but I do not have other Intel based, in case it is graphics related.
As for log (not like I have an idea what to look for) I have error loading fbdev module and some touchpad errors in xorg.log (not imediatelly after the problem), and some Networkmanager warnings.
Damn. Now I failed three times in a row to switch from konsole to rekonq in order to copy those logs. Anyway this is tail -n 50 everything.log (right now X11 screen is just gray and mouse cursor). SELinux can be ignored as it is set permissive.
Oct 13 20:39:46 localhost kernel: ata3.00: configured for UDMA/133
Oct 13 20:39:46 localhost kernel: ata3: EH complete
Oct 13 20:39:46 localhost kernel: EXT4-fs (dm-3): re-mounted. Opts: commit=0
Oct 13 20:39:46 localhost kernel: EXT4-fs (dm-5): re-mounted. Opts: commit=0
Oct 13 20:39:46 localhost kernel: type=1400 audit(1286995186.522:24): avc: denied { execstack } for pid=16129 comm="ksmserver" scontext=user_u:user_r:insmod_t tcontext=user_u:user_r:insmod_t tclass=process
Oct 13 20:39:46 localhost kernel: EXT4-fs (dm-2): re-mounted. Opts: commit=0
Oct 13 20:39:46 localhost kernel: EXT4-fs (dm-4): re-mounted. Opts: commit=0
Oct 13 20:39:47 localhost NetworkManager[22388]: <warn> invalid connection: 'NMSettingWireless' / 'mac-address' invalid: 1
Oct 13 20:39:47 localhost NetworkManager[22388]: <warn> invalid connection: 'NMSettingWireless' / 'mac-address' invalid: 1
Oct 13 20:39:53 localhost kernel: composite sync not supported
Oct 13 20:39:54 localhost kernel: SELinux: initialized (dev fuse, type fuse), uses genfs_contexts
Oct 13 20:39:54 localhost kernel: type=1401 audit(1286995194.036:25): security_compute_sid: invalid context user_u:user_r:mount_t for scontext=user_u:user_r:insmod_t tcontext=system_u:object_r:mount_exec_t tclass=process
Oct 13 20:39:54 localhost kernel: type=1400 audit(1286995194.036:26): avc: denied { read write } for pid=16308 comm="mount" path="socket:[1171927]" dev=sockfs ino=1171927 scontext=user_u:user_r:mount_t tcontext=user_u:user_r:insmod_t tclass=unix_stream_socket
Oct 13 20:39:54 localhost kernel: type=1400 audit(1286995194.036:27): avc: denied { read write } for pid=16308 comm="mount" path="/dev/fuse" dev=devtmpfs ino=4086 scontext=user_u:user_r:mount_t tcontext=system_u:object_r:fuse_device_t tclass=chr_file
Oct 13 20:39:54 localhost kernel: type=1400 audit(1286995194.039:28): avc: denied { append } for pid=16308 comm="mount" name="mtab" dev=dm-3 ino=37749 scontext=user_u:user_r:mount_t tcontext=system_u:object_r:etc_t tclass=file
Oct 13 20:39:56 localhost kernel: type=1400 audit(1286995196.412:29): avc: denied { execmod } for pid=16256 comm="knotify4" path="/usr/lib/codecs/drvc.so" dev=dm-3 ino=828208 scontext=user_u:user_r:insmod_t tcontext=system_u:object_r:lib_t tclass=file
Oct 13 20:39:59 localhost kernel: type=1400 audit(1286995199.259:30): avc: denied { execmod } for pid=16256 comm="knotify4" path="/usr/lib/libx264.so.104" dev=dm-3 ino=16190 scontext=user_u:user_r:insmod_t tcontext=user_u:object_r:lib_t tclass=file
Oct 13 20:42:02 localhost dhcpcd[22391]: eth0: renewing lease of 192.168.1.200
Oct 13 20:42:02 localhost dhcpcd[22391]: eth0: acknowledged 192.168.1.200 from 192.168.1.254 `router'
Oct 13 20:42:02 localhost dhcpcd[22391]: eth0: leased 192.168.1.200 for 300 seconds
Oct 13 20:43:09 localhost kernel: type=1401 audit(1286995389.304:31): security_compute_sid: invalid context user_u:user_r:mount_t for scontext=user_u:user_r:insmod_t tcontext=system_u:object_r:mount_exec_t tclass=process
Oct 13 20:43:12 localhost kdm[3254]: X server for display :0 terminated unexpectedly
Oct 13 20:43:12 localhost acpid: client 15799[0:0] has disconnected
Oct 13 20:43:12 localhost acpid: client connected from 20364[0:0]
Oct 13 20:43:12 localhost acpid: 1 client rule loaded
Oct 13 20:43:13 localhost kernel: composite sync not supported
Oct 13 20:43:13 localhost kernel: composite sync not supported
Oct 13 20:43:20 localhost kernel: ata3.00: configured for UDMA/133
Oct 13 20:43:20 localhost kernel: ata3: EH complete
Oct 13 20:43:20 localhost kernel: EXT4-fs (dm-3): re-mounted. Opts: commit=0
Oct 13 20:43:21 localhost kernel: EXT4-fs (dm-5): re-mounted. Opts: commit=0
Oct 13 20:43:21 localhost kernel: EXT4-fs (dm-2): re-mounted. Opts: commit=0
Oct 13 20:43:22 localhost kernel: EXT4-fs (dm-4): re-mounted. Opts: commit=0
Oct 13 20:43:22 localhost NetworkManager[22388]: <warn> invalid connection: 'NMSettingWireless' / 'mac-address' invalid: 1
Oct 13 20:43:22 localhost NetworkManager[22388]: <warn> invalid connection: 'NMSettingWireless' / 'mac-address' invalid: 1
Oct 13 20:43:24 localhost kernel: SELinux: initialized (dev fuse, type fuse), uses genfs_contexts
Oct 13 20:43:24 localhost kernel: type=1401 audit(1286995404.994:32): security_compute_sid: invalid context user_u:user_r:mount_t for scontext=user_u:user_r:insmod_t tcontext=system_u:object_r:mount_exec_t tclass=process
Oct 13 20:43:25 localhost kernel: type=1400 audit(1286995405.008:33): avc: denied { write } for pid=20701 comm="mount" name="mtab" dev=dm-3 ino=37746 scontext=user_u:user_r:mount_t tcontext=system_u:object_r:etc_t tclass=file
Oct 13 20:43:25 localhost kernel: type=1400 audit(1286995405.008:34): avc: denied { append } for pid=20701 comm="mount" name="mtab" dev=dm-3 ino=37746 scontext=user_u:user_r:mount_t tcontext=system_u:object_r:etc_t tclass=file
Oct 13 20:43:25 localhost kernel: composite sync not supported
Oct 13 20:44:32 localhost dhcpcd[22391]: eth0: renewing lease of 192.168.1.200
Oct 13 20:44:32 localhost dhcpcd[22391]: eth0: acknowledged 192.168.1.200 from 192.168.1.254 `router'
Oct 13 20:44:32 localhost dhcpcd[22391]: eth0: leased 192.168.1.200 for 300 seconds
Oct 13 20:45:44 localhost ntpd[22509]: skew change 53.711 exceeds limit
Oct 13 20:47:02 localhost dhcpcd[22391]: eth0: renewing lease of 192.168.1.200
Oct 13 20:47:02 localhost dhcpcd[22391]: eth0: acknowledged 192.168.1.200 from 192.168.1.254 `router'
Oct 13 20:47:02 localhost dhcpcd[22391]: eth0: leased 192.168.1.200 for 300 seconds
Oct 13 20:49:32 localhost dhcpcd[22391]: eth0: renewing lease of 192.168.1.200
Oct 13 20:49:32 localhost dhcpcd[22391]: eth0: acknowledged 192.168.1.200 from 192.168.1.254 `router'
Oct 13 20:49:32 localhost dhcpcd[22391]: eth0: leased 192.168.1.200 for 300 seconds
This should cover the time of at least one of the failures. Can somebody see something in it?
Last edited by Nicky726 (2010-10-13 18:58:16)