Re: hostname curiosity [SOLVED] ...but curioser and curioser
by Steve Blackwell
---- "Mikkel L. Ellertson" <mikkel(a)infinity-ltd.com> wrote:
> Steve wrote:
> > ---- "Mikkel L. Ellertson" <mikkel(a)infinity-ltd.com> wrote:
> >> Steve wrote:
> >>> Cool! Adding
> >>>
> >>> DHCP_HOSTNAME=yes
> >>>
> >>> to /etc/sysconfig/network-scripts/ifcfg-eth0 seems to have done
> >>> the trick. Now there is a longer pause during booting when it gets
> >>> to the Setting up eth0...[OK] line so I'm guessing it is waiting a
> >>> little longer to get an answer from the dhcp server.
> >>> Btw, where are the docs for the ifcfg-eth0 file syntax? There is
> >>> no man page for it.
It looks like I spoke too soon. Yesterday when I booted up in the morning I got the hp<number> for the host name back, ie not what I was expecting. Rebooting got the correct name again.
> > The second thing is that there is no DHCP_HOSTNAME option listed
> > in the /usr/share/doc/initscripts-8.76.4/sysconfig.txt and yet it
> > DHCP_HOSTNAME="yes" seems to work evem though PERSISTENT_DHCLIENT is
> > still not set.
> >
> It should be set to the hostname you desire. I am surprised that it
> is not in the file. It is processed by the ifup-eth script.
I did some looking around in the ifup-eth scrupt and yes, you are correct in that whatever DHCP_HOSTNAME is set to gets sent to the dhcp server to say that this is the hostname I want. But that's the problem - I want the dhcp server to tell the client what hostname to use, not the other way around. From the dhclient man page:
-R <option>[,<option>...]
Specify the list of options the client is to request from the server. The option list must be a single string consisting of option names separated
by at least one command and optional space characters. The default option list is:
subnet-mask, broadcast-address, time-offset, routers,
domain-name, domain-name-servers, host-name, nis-domain,
nis-servers, ntp-servers
This tells me that the client >>requets<< host-name >>from<< the server by default so it should not be necessary to configure anything to get this. (Hmmm... maybe the dhcp protocol does not >require< that the server honour the request). Obviously, this is not working all the time so I modified ifup-eth and replace the -q switch to dhclient with a -v to try and get a better idea of what is going on. I saw some messages flash by during the boot but I can't find where they have been logged. Any ideas or do I need to log them manually by adding '> dhclient.log 2>&1 ' to the dhclient startup call?
> > The third thing is that something is creating a
> /etc/dhclient-eth0.conf file at boot time that contains this line:
> >
> > send host-name "yes"; # temporary RHL ifup addition
> >
> I am not sure what is creating it - I will have to look into it.
>
> > Obviously, my host name is not "yes" so this must be some
> > special, undocumented use of send host-name that make the
> > dhcp client get the hostname from the server.
> >
> I think it is causing the DHCP server to send the client a hostname.
> It is probably a "feature" of the server, rather then something with
> the client.
I figured this one out. The ifup-eth script writes the dhclient-eth0.conf file with the line
send host-name "<whatever DHCP_HOSTNAME in ifcfg-eth0 is set to>";
so when I wrote DHCP_HOSTNAME=yes in ifcfg-eth0 I got
send host-name = "yes";
in dhclient-eth0.conf.
Steve
15 years, 4 months
secure-login related log messages when using kdm/gnome
by mike cloaked
I use kdm for my login manager, and gnome as the desktop. Whenever I restart
the system I get the following in the "Connections (secure-login) section of
my logfiles:
**Unmatched Entries**
kdm: :0: gnome-keyring-daemon: couldn't lookup keyring component
setting: Failed to contact configuration server; some possible causes are
that you need to enable TCP/IP networking for ORBit, or you have stale NFS
locks due to a system crash. See http://www.gnome.org/projects/gconf/ for
information. (Details - 1: Not running within active
session)gnome-keyring-daemon: couldn't lookup ssh component setting: Failed
to contact configuration server; some possible causes are that you need to
enable TCP/IP networking for ORBit, or you have stale NFS locks due to a
system crash. See http://www.gnome.org/projects/gconf/ for information.
(Details - 1: Not running within active session)gnome-keyring-daemon:
couldn't lookup pkcs11 component setting: Failed to contact configuration
server; some possible causes are that you need to enable TCP/IP networking
for ORBit, or you have stale NFS locks due to a system crash. See
http://www.gnome.org/projects/gconf/ for information. (Details - 1: Not
running within active session): 1 Time(s)
It happens on all machines where I have kdm/gnome.
Does anyone know why these log messages arise, and whether there is
something that needs changing/fixed?
--
View this message in context: http://www.nabble.com/secure-login-related-log-messages-when-using-kdm-gn...
Sent from the Fedora List mailing list archive at Nabble.com.
15 years, 4 months
Re: Fedora fail for Blender :( Ubuntu works... windows also fail :)
by Hiisi
> Hi,
> this past weekend we organized Blender workshop and to my surprise
> Fedora 10 failed on few computers for running this workshop! :(
> This is the bug: https://bugzilla.redhat.com/show_bug.cgi?id=47915
>
> The good news is that we had few laptops with ATI graphic cards so
> they worked. To make Fedora not look like it sucks Windows came to the
> rescue :) Blender running on Windows laptops has even more issues :)
> Latest Ubuntu has no issues with running Blender on same hardware.
>
> Do you have some success/fail stories regarding Blender?
>
> ps. Blender is awesome!!!
>
> Valent.
>
I'm running Blender on Fedora 9 and it works well except selecting. Not
all edges of selected object are pink. For example, when I start blender
only two edges of default cube are pink. All other seems to be all
right.
In a school where I teach Blender, on Windows machines with all the
latest updates everything is fine.
--
Hiisi.
15 years, 4 months
Fedora 10 installation failing and dropping to shell
by Jason Kania
Hello,
I downloaded the Fedora 10 CD and attempted to install. The
installation starts and then drops to shell as follows when not in
quiet mode:
...
starting udevd
creating devices
Waiting for system to settle
mounting /dev/root
lrwxrwxrwx 1 0 0 3 Jan 28 01:10 /dev/root ->sr0
---------------------------------------------------------
WARNING: Requested /sbin/init binary does not exist on rootfs
---------------------------------------------------------
Dropping to a shell. Good luck!
I checked in the /sbin and init directory and it is not there. Is this
a detection issue of some sort or a known bug in the installation? I
looked for this problem quite a bit in google, but did not see an
explanation or work around that was comprehensible. Any suggestions?
Thanks,
Jason
15 years, 4 months
Re: Need second opinion, is my HD failing?
by Bryn M. Reeves
Paulo Cavalcanti wrote:
> But in Richard's case, 955 seems odd to me:
>
> 5 Reallocated_Sector_Ct 0x0033 100 100 036 Pre-fail Always
> - 955
>
> Probably, you are right, and the value is OK. But I have never seen a
> counting like
> this before. I had a defective disk once, which increased 20 or 30 bad
> sectors a day.
> Therefore, such a high score would not be a surprise for me (Seagate
> replaced the disk for me,
> even it being more than 2 years old).
It is quite a high number and many of my drives do have raw values
much closer to zero, although I have at least a few with raw values in
the 100s that still have 100 as the normalised value and are working fine.
The problem with trying to interpret the raw values is that they are
completely under the control of the vendor. The only thing S.M.A.R.T.
specifies is the size of the field. Some vendors have previously taken
a single field and used it to encode multiple values (e.g. breaking it
up into several sub-fields). For example, some IBM drives encode three
distinct temperature measurements in the raw value for the
Temperature_Celsius attribute.
Because of this, unless you know the scheme being used for a given
vendor/drive model it's impossible to make any accurate assumption
from the raw value alone - you just have to trust the firmware to
decrement the normalised value appropriately as the drive begins to
deteriorate.
Regards,
Bryn.
15 years, 4 months
compiz-fusion: how to set a different wallpaper on different desktop
by Dario Lesca
Hi, I use wallpapoz for change the wallpaper on different desktop.
When I active compiz and change desktop, wallpapoz not work and do not
change the wallpaper on new desktop.
How to I can do that with compiz active?
Many thanks
--
Dario Lesca <d.lesca(a)solinos.it>
15 years, 4 months
Login keyring weirdness
by Christoph Höger
Hi,
I have this problem since f10 beta: Whenever I start evolution or run a
ssh connection with key based authentication enabled I am asked to
unlock my login keyring as it was not unlocked on login.
I am not able to unlock it with my login password.
What am I supposed to enter here?
thanks
Christoph
15 years, 4 months
Dell OptiPlex 745 reboot problem
by Tony Molloy
Hi,
I've just installed Fedora-9 on a lab of Dell OptiPlex 745 (SFF) machines. (
only in 1 lab TG )
After running firstboot when I went to reboot the machines they just hang and
I had to do a hard reboot. I thought this was a minor glitch and ignored it.
Now however when the machines boot into Fedora-9 the "reboot" and "suspend"
buttons do not work. The windowing system just shuts down and I get a text
prompt and the machines just hang there.
As thes are dual boot machines this will cause a lot of problems starting
monday when the students return ;-(
Is ther some magic incantation to grub to sort this problem or any body got
any ideas.
Thank's
Tony
15 years, 4 months
A bug in syslogd?
by Peter Teoh
I did a simple thing - modified my syslog.conf:
cat /etc/syslog.conf
mail.none;authpriv.none;cron.none /var/log/messages
So virtually, there is nothing to go to /var/log/messages, although my
dmesg's output did output a lot of other stuff....as I instrumented
the kernel to do printk()....something like every file traversal will
generate several entries in dmesg output.
Nevertheless, since /var/log/messages to get, as I check, its content
is always zero (after I did an initial truncation) - why is syslogd
showing such a high performance:
First snapshot:
27790 root 20 0 268m 203m 14m S 4.3 20.2 2:38.36 opera
3459 root 20 0 315m 33m 8160 S 2.6 3.3 8:16.22 Xorg
532 root 15 -5 0 0 0 S 0.3 0.0 0:01.13 ata/0
2417 root 20 0 1888 648 548 S 0.3 0.1 0:01.73 syslogd
3770 root 20 0 73928 10m 4688 S 0.3 1.0 0:15.76
gnome-terminal
3772 root 20 0 5092 532 484 S 0.3 0.1 0:00.16
wpa_supplicant
28349 root 20 0 2344 1028 788 R 0.3 0.1 0:00.05 top
1 root 20 0 2224 604 564 S 0.0 0.1 0:01.11 init
Next snapshot:
3459 root 20 0 315m 33m 8160 S 11.2 3.3 8:18.51 Xorg
2420 root 20 0 1828 460 392 R 1.9 0.0 0:01.82 klogd
27790 root 20 0 268m 203m 14m S 1.9 20.2 2:39.87 opera
28350 root 20 0 2340 936 700 R 1.9 0.1 0:00.01 top
1 root 20 0 2224 604 564 S 0.0 0.1 0:01.11 init
2 root 15 -5 0 0 0 S 0.0 0.0 0:00.00 kthreadd
3 root RT -5 0 0 0 S 0.0 0.0 0:00.00
migration/0
4 root 15 -5 0 0 0 S 0.0 0.0 0:00.40
ksoftirqd/0
5 root RT -5 0 0 0 S 0.0 0.0 0:00.00
watchdog/0
Over a period of time, I observed that klogd and syslogd is toggling
to be among the top few candidate all the time - toggling, meaning
switching between one and another.
Can someone explained this behavior? Shouldn't the syslogd be
consuming almost zero cpu % since there is zero output to
/var/log/messages?
PS: I did restart syslogd after /etc/syslogd.conf modification.
--
Regards,
Peter Teoh
15 years, 4 months
Assistance with DHCP server setup.
by Simon Slater
Hi all. The last few days I've attempted to setup a DHCP server
on our small SOHO network, and went round in a big circle to where I
started.
What I would like to achieve is:
/-> eth0 192.168.1.? -> DSL router 192.168.1.254 -> internet.
Server \
(dell.local)\->eth1 192.168.1.1 -> LAN
To begin with the Linksys router was serving addresses so I
turned this function off. From the sample .conf, various howtos and
posts to this list I came up with a very simple dhcpd.conf which could
be added to later as extra functionality is needed, as follows:
[root@dell ~]# cat /etc/dhcpd.conf
#Sample /etc/dhcpd.conf
# (add your comments here)
default−lease−time 600;
max−lease−time 7200;
option subnet−mask 255.255.255.0;
option broadcast−address 192.168.1.255;
option routers 192.168.1.254;
#option domain−name−servers 192.168.1.1, 192.168.1.2;
#option domain−name "mydomain.org";
authoritative;
subnet 192.168.1.0 netmask 255.255.255.0 {
range 192.168.1.100 192.168.1.199;
host dell.local {
hardware ethernet 00:1F:1F:09:38:A2;
fixed-address 192.168.1.1;
}
}
[root@dell ~]#
The first problem is, should eth0 take an address via dhcp or be
set one? Should it be bound to a MAC address in the dhcpd.conf? Does
this file look okay?
The second situation is that:
[root@dell ~]# service dhcpd start
Starting dhcpd: [FAILED]
[root@dell ~]#
with a Selinux denial:
SELinux is preventing dhcpd (dhcpd_t) "read write" unconfined_t.
with SEtroubleshooter giving no suggestions for a fix. How should I
progress here?
I have set things back to the way they were after install ( I
think). The /var/lib/dhcpd.leases file exists. Ports 67 & 68 are
allowed through the firewall and all eth* devices are trusted (using
system-config-firewall). Any advice will be most thankfully welcomed.
--
Regards,
Simon Slater
Registered Linux User #463789. Be counted at: http://counter.li.org/
15 years, 4 months