I am running CentOS 5/2 (latest updates) with the GNOME DE on a 32-bit machine (at work).
I have k3b installed, and I was trying to copy a DVD earlier this morning, but k3b said it couldn't read encrypted DVDs.
So, I installed libdvdcss from rpmforge and restarted k3b. It hung the system. I rebooted, and / had been damaged. After running e2fsck from the repair prompt, I rebooted and a whole slew of errors revolving around various /var directories that did not ecist occurred.
I have been trying to repair /var, and so far with a fair modicum of success, but I've hit an interesting wall - two, actually.
1) The gdm refuses to come up. It claims that "Server Authorization directory (daemon/ServAuthDir) is set to /var/gdm, but this does not exist...."
However:
# ll -d /var/gdm drwxrwx--T 2 root gdm 4096 Dec 11 10:31 /var/gdm # ll /var/gdm total 8 -rw-r----- 1 root root 45 Nov 26 10:47 :0.Xauth -rw-r--r-- 1 root root 63 Dec 11 09:14 :0.Xservers
This is identical to my backup system (which is not surprising - I set up the dir and copied the files from here - was that a bad idea?).
2) The following daemons fail to start: auditd, NFS statd, avahi and HAL. I've tried to pin down why the avahi daemon won't start because it keeps logging permissions errors trying to create the pid file /var/run/avahi-daemon//pid, but the setup of /var /var/run and /var/run/avahi-daemon are all identical to this (backup) machine. Any suggestions?
Or is there a better, more comprehensive repair facility available?
BTW, OT: Does anyone know why this might have happened? I have all this installed at home, no problems whatsoever (libdvdcss works seamlessly with all my DVD tools, including k3b).
Thanks!
mhr
MHR wrote:
I am running CentOS 5/2 (latest updates) with the GNOME DE on a 32-bit machine (at work).
I have k3b installed, and I was trying to copy a DVD earlier this morning, but k3b said it couldn't read encrypted DVDs.
So, I installed libdvdcss from rpmforge and restarted k3b. It hung the system. I rebooted, and / had been damaged. After running e2fsck from the repair prompt, I rebooted and a whole slew of errors revolving around various /var directories that did not ecist occurred.
I have been trying to repair /var, and so far with a fair modicum of success, but I've hit an interesting wall - two, actually.
- The gdm refuses to come up. It claims that "Server Authorization
directory (daemon/ServAuthDir) is set to /var/gdm, but this does not exist...."
However:
# ll -d /var/gdm drwxrwx--T 2 root gdm 4096 Dec 11 10:31 /var/gdm # ll /var/gdm total 8 -rw-r----- 1 root root 45 Nov 26 10:47 :0.Xauth -rw-r--r-- 1 root root 63 Dec 11 09:14 :0.Xservers
A common cause of failures like that is damaged directory permissions near the top of the filesystem hierarchy, often as a result of unpacking a tar archive into the root directory.
On Thu, Dec 11, 2008 at 2:19 PM, Robert Nichols rnicholsNOSPAM@comcast.net wrote:
MHR wrote:
I am running CentOS 5/2 (latest updates) with the GNOME DE on a 32-bit machine (at work).
I have k3b installed, and I was trying to copy a DVD earlier this morning, but k3b said it couldn't read encrypted DVDs.
So, I installed libdvdcss from rpmforge and restarted k3b. It hung
<snip>
A common cause of failures like that is damaged directory permissions near the top of the filesystem hierarchy, often as a result of unpacking a tar archive into the root directory.
Mark: If you trace it back to the libdvdcss package from rpmforge, notify them about it. Strange that it works OK for you at home and you hosed your Workstation at work, but that's Murphy's Law. Good luck getting it working! Possibly try this again, under VMWare, when you are relaxed and want to see if it can be recreated. Lanny
On Thu, Dec 11, 2008 at 11:35 AM, Lanny Marcus lmmailinglists@gmail.com wrote:
On Thu, Dec 11, 2008 at 2:19 PM, Robert Nichols rnicholsNOSPAM@comcast.net wrote:
<snip> > A common cause of failures like that is damaged directory permissions > near the top of the filesystem hierarchy, often as a result of > unpacking a tar archive into the root directory.
Hm - well, _I_ never do that, and I rather doubt that yum did, either, but I suppose that would depend on what's in libdvdcss. I find it hard to believe that it wasn't something else more subtle with k3b, but, again, who knows?
Mark: If you trace it back to the libdvdcss package from rpmforge, notify them about it. Strange that it works OK for you at home and you hosed your Workstation at work, but that's Murphy's Law. Good luck getting it working! Possibly try this again, under VMWare, when you are relaxed and want to see if it can be recreated. Lanny
1) I did ask on the rpmforge list. Waiting to hear back from there, too.
2) I am just now beginning to really appreciate virtualization. Still, past experience told me this would not be a problem. I guess that would best be described as naive....
mhr
I managed to get a partial solution running. I rebooted from my installation DVD and started to run an upgrade, but when it got to the reboot part, I booted back to the DVD and ran 'linux rescue'. I'm not entirely sure what it did besides mangle my /etc/fstab (lost the nfs /home mount), but I just happened to have a backup, so I remounted /home and now everything seems to be working.
EXCEPT
The system is deathly slow in some areas. OOo and Evolution take forever to do the simplest things, although Seamonkey and the xterms are working just fine. I'm going to reboot again to see if that makes any difference.
Bar.
mhr
On Thu, Dec 11, 2008 at 12:29 PM, MHR mhullrich@gmail.com wrote:
I managed to get a partial solution running. I rebooted from my installation DVD and started to run an upgrade, but when it got to the reboot part, I booted back to the DVD and ran 'linux rescue'. I'm not entirely sure what it did besides mangle my /etc/fstab (lost the nfs /home mount), but I just happened to have a backup, so I remounted /home and now everything seems to be working.
EXCEPT
The system is deathly slow in some areas. OOo and Evolution take forever to do the simplest things, although Seamonkey and the xterms are working just fine. I'm going to reboot again to see if that makes any difference.
It doesn't. It seems, though, that it is GNOME that is painfully slow to start any thing up, and when it is doing so, it hampers everything else to some extent. Slow as in it takes minutes to load programs instead of milliseconds....
mhr
On Thu, Dec 11, 2008 at 3:01 PM, MHR mhullrich@gmail.com wrote:
On Thu, Dec 11, 2008 at 12:29 PM, MHR mhullrich@gmail.com wrote:
I managed to get a partial solution running. I rebooted from my installation DVD and started to run an upgrade, but when it got to the reboot part, I booted back to the DVD and ran 'linux rescue'. I'm not entirely sure what it did besides mangle my /etc/fstab (lost the nfs /home mount), but I just happened to have a backup, so I remounted /home and now everything seems to be working.
EXCEPT
The system is deathly slow in some areas. OOo and Evolution take forever to do the simplest things, although Seamonkey and the xterms are working just fine. I'm going to reboot again to see if that makes any difference.
It doesn't. It seems, though, that it is GNOME that is painfully slow to start any thing up, and when it is doing so, it hampers everything else to some extent. Slow as in it takes minutes to load programs instead of milliseconds....
Are you sure your hard disk is healthy?
MHR wrote: . . .
It doesn't. It seems, though, that it is GNOME that is painfully slow to start any thing up, and when it is doing so, it hampers everything else to some extent. Slow as in it takes minutes to load programs instead of milliseconds....
Gnome at one time (RH9 days I think) was painfully slow to start after a hostname change, until it udpated itself in all places - or whatever it was doing. Is your hostname & /etc/hosts still intact? If it's been up for a while, I would expect it to resolve itself already. That's how it used to was anyway.
Nothing in the log files? top show anything of interest?
On Thu, Dec 11, 2008 at 1:32 PM, Toby Bluhm tkb@alltechmedusa.com wrote:
MHR wrote:
Gnome at one time (RH9 days I think) was painfully slow to start after a hostname change, until it udpated itself in all places - or whatever it was doing. Is your hostname & /etc/hosts still intact? If it's been up for a while, I would expect it to resolve itself already. That's how it used to was anyway.
All that looks ok so far.
Nothing in the log files? top show anything of interest?
Just this, in /var/log/messages:
Dec 11 14:38:07 swordfish kernel: statd: server localhost not responding, timed out Dec 11 14:38:07 swordfish kernel: lockd: cannot monitor 10.24.15.48 Dec 11 14:38:07 swordfish kernel: lockd: failed to monitor 10.24.15.48 Dec 11 14:38:37 swordfish kernel: statd: server localhost not responding, timed out Dec 11 14:38:37 swordfish kernel: lockd: cannot monitor 10.24.15.48 Dec 11 14:38:37 swordfish kernel: lockd: failed to monitor 10.24.15.48 Dec 11 14:42:06 swordfish kernel: statd: server localhost not responding, timed out Dec 11 14:42:06 swordfish kernel: lockd: cannot monitor 10.24.15.48 Dec 11 14:42:06 swordfish kernel: lockd: failed to monitor 10.24.15.48
This happens a lot. 10.24.15.48 is our internal /home nfs machine, so that might be involved (slow access to /home can cause a lot of problems). I don't know why localhost would be a problem. Here's my /etc/hosts and ifconfig output:
[mrichter@swordfish ~]$ sudo cat /etc/hosts # Do not remove the following line, or various programs # that require network functionality will fail. 127.0.0.1 localhost localhost.localdomain ::1 localhost.localdomain localhost 192.168.9.225 swordfish swordfish.sjhtca.com 192.168.9.92 t-mrichter-08 t-mrichter-08.sjhtca.com [mrichter@swordfish ~]$ ifc eth0 Link encap:Ethernet HWaddr 00:1C:C0:1F:5E:38 inet addr:192.168.9.225 Bcast:192.168.9.255 Mask:255.255.255.0 inet6 addr: fe80::21c:c0ff:fe1f:5e38/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:31680 errors:0 dropped:0 overruns:0 frame:0 TX packets:28897 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:100 RX bytes:25685445 (24.4 MiB) TX bytes:15109432 (14.4 MiB) Memory:92200000-92220000
lo Link encap:Local Loopback inet addr:127.0.0.1 Mask:255.0.0.0 inet6 addr: ::1/128 Scope:Host UP LOOPBACK RUNNING MTU:16436 Metric:1 RX packets:363 errors:0 dropped:0 overruns:0 frame:0 TX packets:363 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:33380 (32.5 KiB) TX bytes:33380 (32.5 KiB)
sit0 Link encap:IPv6-in-IPv4 NOARP MTU:1480 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:0 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
vmnet8 Link encap:Ethernet HWaddr 00:50:56:C0:00:08 inet addr:192.168.240.1 Bcast:192.168.240.255 Mask:255.255.255.0 inet6 addr: fe80::250:56ff:fec0:8/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:82 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
During startup now, the only process that fails is NFS statd, which probably fits right into the above, but I know zip about this area.
Any suggestions? This wasn't happening before, and it doesn't happen on my backup machine at all (surprise!).
Thanks.
mhr
On Thu, 2008-12-11 at 14:46 -0800, MHR wrote:
<snip>
[mrichter@swordfish ~]$ sudo cat /etc/hosts # Do not remove the following line, or various programs # that require network functionality will fail. 127.0.0.1 localhost localhost.localdomain ::1 localhost.localdomain localhost 192.168.9.225 swordfish swordfish.sjhtca.com 192.168.9.92 t-mrichter-08 t-mrichter-08.sjhtca.com
/etc/hosts permissions OK? I ask because it should be worl-readable and you sudo'd it.
<snip>
During startup now, the only process that fails is NFS statd, which probably fits right into the above, but I know zip about this area.
IIRC (it's been a long time), the RPC stuff needs to be running for nfs locks, status, etc. Are they? I can't confirm this because I don't have any nfs stuff running.
But I don't recall whether server/clioent needs any/all on one or both.
Anyway, that and portmap? Not sure. /etc/hosts/{deny,allow} come into play if portmap is in use.
I don't recall dinking with the nfs service, so I ps'd it.
rpc 2284 1 0 05:36 ? 00:00:00 portmap rpcuser 2310 1 0 05:36 ? 00:00:00 rpc.statd root 2356 1 0 05:36 ? 00:00:00 rpc.idmapd
Maybe that gives a clue?
Showmounts might help you out (server side only?).
Any suggestions? This wasn't happening before, and it doesn't happen on my backup machine at all (surprise!).
<snip sig stuff>
HTH
On Thu, Dec 11, 2008 at 3:29 PM, William L. Maltby CentOS4Bill@triad.rr.com wrote:
/etc/hosts permissions OK? I ask because it should be worl-readable and you sudo'd it.
It is - I was overdoing it....
IIRC (it's been a long time), the RPC stuff needs to be running for nfs locks, status, etc. Are they? I can't confirm this because I don't have any nfs stuff running.
But I don't recall whether server/clioent needs any/all on one or both.
Anyway, that and portmap? Not sure. /etc/hosts/{deny,allow} come into play if portmap is in use.
I don't recall dinking with the nfs service, so I ps'd it.
rpc 2284 1 0 05:36 ? 00:00:00 portmap rpcuser 2310 1 0 05:36 ? 00:00:00 rpc.statd root 2356 1 0 05:36 ? 00:00:00 rpc.idmapd
Maybe that gives a clue?
I get (psg = ps -ef | grep -i):
[mrichter@swordfish ~]$ psg rpc rpc 5786 1 0 14:32 ? 00:00:00 portmap root 5937 11 0 14:32 ? 00:00:00 [rpciod/0] root 5938 11 0 14:32 ? 00:00:00 [rpciod/1]
Showmounts might help you out (server side only?).
I'm listed there (it's 'showmount' though :-).
I am now certain that this is an nfs issue, but I haven't a clue where to look. I noticed that my backup system had a /var/lock/subsys/nfslock file, so I touched it over here. I think it's running a little faster, but it's still like tar whenever one of my home dir files gets touched (which is a lot, since most apps use their .<appname> files for local storage).
Any nfs gurus here?
:-)
Thanks to everyone so far....
mhr
On Thu, Dec 11, 2008 at 2:47 PM, MHR mhullrich@gmail.com wrote: <snip>
unpacking a tar archive into the root directory.
Hm - well, _I_ never do that, and I rather doubt that yum >did, either,
Since you got it from rpmforge, I assume it was an rpm and not a tar file.
but I suppose that would depend on what's in libdvdcss. >I find it hard to believe that it wasn't something else more >subtle with k3b, but, again, who knows?
Or, more probably, with the libdvdcss <snip>
I did ask on the rpmforge list. Waiting to hear back from there, too.
I am just now beginning to really appreciate >virtualization.
A bunch of the gurus on this list use it. If I had a box with more RAM, I would try it.
Still, past experience told me this would not be a >problem. I guess that would best be described as naive....
I think since it works OK on your Desktop at home, that's not so naive. But, the HW is different and something may be awry on the HW on your Workstation at work. Or, there may have been a power glitch, while you were installing the SW. Is your Workstation on a UPS? Been using K3b for a long time here and never a disaster, like you experienced today.
On Thu, Dec 11, 2008 at 2:29 PM, Lanny Marcus lmmailinglists@gmail.com wrote:
On Thu, Dec 11, 2008 at 2:47 PM, MHR mhullrich@gmail.com wrote:
<snip> >>> unpacking a tar archive into the root directory. > Hm - well, _I_ never do that, and I rather doubt that yum >did, either,
Since you got it from rpmforge, I assume it was an rpm and not a tar file.
but I suppose that would depend on what's in libdvdcss. >I find it hard to believe that it wasn't something else more >subtle with k3b, but, again, who knows?
Or, more probably, with the libdvdcss
<snip> > 1) I did ask on the rpmforge list. Waiting to hear back from there, too. > > 2) I am just now beginning to really appreciate >virtualization.
A bunch of the gurus on this list use it. If I had a box with more RAM, I would try it.
Still, past experience told me this would not be a >problem. I guess that would best be described as naive....
I think since it works OK on your Desktop at home, that's not so naive. But, the HW is different and something may be awry on the HW on your Workstation at work. Or, there may have been a power glitch, while you were installing the SW. Is your Workstation on a UPS? Been using K3b for a long time here and never a disaster, like you experienced today. _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Virtualization is great and all, but not sure that watching a dvd on a virtual machine would work so well.
on 12-11-2008 1:01 PM Rob Townley spake the following:
On Thu, Dec 11, 2008 at 2:29 PM, Lanny Marcus lmmailinglists-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org wrote:
On Thu, Dec 11, 2008 at 2:47 PM, MHR mhullrich-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org wrote:
<snip> >>> unpacking a tar archive into the root directory. > Hm - well, _I_ never do that, and I rather doubt that yum >did, either, Since you got it from rpmforge, I assume it was an rpm and not a tar file.
but I suppose that would depend on what's in libdvdcss. >I find it hard to believe that it wasn't something else more >subtle with k3b, but, again, who knows?
Or, more probably, with the libdvdcss
<snip> > 1) I did ask on the rpmforge list. Waiting to hear back from there, too. > > 2) I am just now beginning to really appreciate >virtualization. A bunch of the gurus on this list use it. If I had a box with more RAM, I would try it.
Still, past experience told me this would not be a >problem. I guess that would best be described as naive....
I think since it works OK on your Desktop at home, that's not so naive. But, the HW is different and something may be awry on the HW on your Workstation at work. Or, there may have been a power glitch, while you were installing the SW. Is your Workstation on a UPS? Been using K3b for a long time here and never a disaster, like you experienced today. _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Virtualization is great and all, but not sure that watching a dvd on a virtual machine would work so well.
WOW... We have to actually "work" on our workstations! There is a DVD player in the breakroom if we need it. ;-P
On Thu, Dec 11, 2008 at 2:26 PM, Scott Silva ssilva@sgvwater.com wrote:
WOW... We have to actually "work" on our workstations! There is a DVD player in the breakroom if we need it. ;-P
It WAS work related. Honest! You do believe me, don't you?
)-;
mhr
On Thu, Dec 11, 2008 at 12:56 PM, MHR mhullrich@gmail.com wrote:
I am running CentOS 5/2 (latest updates) with the GNOME DE on a 32-bit machine (at work).
I have k3b installed, and I was trying to copy a DVD earlier this morning, but k3b said it couldn't read encrypted DVDs.
So, I installed libdvdcss from rpmforge and restarted k3b. It hung the system. I rebooted, and / had been damaged. After running e2fsck from the repair prompt, I rebooted and a whole slew of errors revolving around various /var directories that did not ecist occurred.
I have been trying to repair /var, and so far with a fair modicum of success, but I've hit an interesting wall - two, actually.
- The gdm refuses to come up. It claims that "Server Authorization
directory (daemon/ServAuthDir) is set to /var/gdm, but this does not exist...."
However:
# ll -d /var/gdm drwxrwx--T 2 root gdm 4096 Dec 11 10:31 /var/gdm # ll /var/gdm total 8 -rw-r----- 1 root root 45 Nov 26 10:47 :0.Xauth -rw-r--r-- 1 root root 63 Dec 11 09:14 :0.Xservers
This is identical to my backup system (which is not surprising - I set up the dir and copied the files from here - was that a bad idea?).
- The following daemons fail to start: auditd, NFS statd, avahi and
HAL. I've tried to pin down why the avahi daemon won't start because it keeps logging permissions errors trying to create the pid file /var/run/avahi-daemon//pid, but the setup of /var /var/run and /var/run/avahi-daemon are all identical to this (backup) machine. Any suggestions?
Or is there a better, more comprehensive repair facility available?
BTW, OT: Does anyone know why this might have happened? I have all this installed at home, no problems whatsoever (libdvdcss works seamlessly with all my DVD tools, including k3b).
Thanks!
mhr _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Was SELINUX in enforcing mode? Rebuilding directories and files that previously had mandatory labels seems like it would cause problems until labels were reapplied.
On Thu, Dec 11, 2008 at 12:35 PM, Rob Townley rob.townley@gmail.com wrote:
Was SELINUX in enforcing mode? Rebuilding directories and files that previously had mandatory labels seems like it would cause problems until labels were reapplied.
I have SELinux turned off.
mhr
-
Jeff -
Lanny Marcus -
MHR -
Rob Townley -
Robert Nichols -
Scott Silva -
Toby Bluhm -
William L. Maltby