Udev rules config and multiple NICs

So I have four NICs and I want them all set up on individual subnets with static IPs. eth0 is connected to my internet router. So, per the Arch wiki, I tried configuring /etc/udev/rules.d/10-network.rules like so:
SUBSYSTEM=="net", ATTR{address}=="08:00:27:4b:c0:38", NAME="net0"
SUBSYSTEM=="net", ATTR{address}=="08:00:27:ca:75:f6", NAME="net1"
SUBSYSTEM=="net", ATTR{address}=="08:00:27:24:26:4b", NAME="net2"
SUBSYSTEM=="net", ATTR{address}=="08:00:27:35:10:a7", NAME="net3"
and my /etc/rc.conf like this:
# Network
interface=net0
address=192.168.1.11
netmask=255.255.255.0
gateway=192.168.1.1
interface=net1
address=192.168.101.11
netmask=255.255.255.0
interface=net2
address=192.168.102.11
netmask=255.255.255.0
interface=net3
address=192.168.103.11
netmask=255.255.255.0
The problem is upon booting up, the only card that gets configured and brought up is net3. I'm not really sure where to go from here to start troubleshooting. Thanks if you can help!

You have to use something else (e.g. netcfg) than network to configure multiple interfaces. The rc.conf configuration of multiple interfaces is deprecated also.
http://www.archlinux.org/news/netcfg-28 … atibility/

Similar Messages

OATS and multiple NICs

on a controller system with multiple network interface, is there a way to force OATS to use a specific one????
our scenario: The Controller system has NIC A and NIC B. All OATS agents are connected only to NIC B. Yet all OATS services (at ports 8088,9001, etc.) are for some reasons bound to just NIC A. We can ping and connect to those agents on NIC B but can't talk to the services on NIC A.
Ideas? Thanks for any inputs
-Bao

The WebLogic management console is the most likely place to look. I recall that you can change the ports here, so the NIC management may be possible as well. The JMSServer may be something to investigate after you are logged into the Console. Also the Console-> Servers->AdminServer(admin) has a general configuration setting named "Listen Address".
http://localhost:8088/console
username=oats and the password is the password you previously chose during the install.
Here is another suggestion from the WebLogic guide.
http://docs.oracle.com/cd/E21764_01/doc.1111/e14770/weblogic_server_issues.htm#CIHFHEGF
Edited by: glenn.s on May 10, 2012 2:55 PM

Replace chmod 666 /dev/dri/card0 with udev rule [Solved]

Hello everyone ! , im configuring my xorg.conf to use 3d, i have a S3 Unichrome Pro (K8M800) chip , and i have it working right now , but in the manual that i read to do this theres a hack that i want to do the right way:
Link
a. First, the permission issue. This is a dirty hack, but I have not taken any time to learn the innards of udev. As root, type "chmod 666 /dev/dri/card0". This will enable a regular user to use dri. I know I should fix it via udev's config, and I intend on figuring that out in the near future and posting an update to this HOWTO. For now, I put this command after "modprobe via" in my /etc/rc.d/rc.local file.
So i was wondering if maybe someone can help me to do the udev trick, as i have in my rc.local the line but if this can be done via udev i want to do it that way.
I was reading the udev article at the wiki but with all that KERNEL %k %n and that stuff i have no idea how to do it, if u think that its better for me to learn this the hard way maybe some usefull link will be good
Thank you.

Starting with /dev/hdd, there is already a rule for this in the default ruleset:
BUS=="ide", KERNEL=="hd[a-z]", SYSFS{removable}=="1", SYSFS{media}=="cdrom*", NAME="%k", GROUP="optical"
This creates /dev/hdd as follows:
brw-rw---- 1 root optical 22, 64 2006-08-10 19:11 /dev/hdd
so all you need to do is add yourself to the optical group. I don't use gnomebaker myself, but the underlying setup is the same for all burning apps.
I was going to post a similar answer for your dri problem as well, because we do have this rule by default:
KERNEL=="card[0-9]*", NAME="dri/%k", GROUP="video"
which should create /dev/dri/cardN with root:video ownership, but I can't verify that - for some reason, my laptop has
crw-rw---- 1 root root 226, 0 2006-09-08 09:11 /dev/dri/card0
instead i.e. GROUP="video" is not applied. If yours shows up as root:video, however, just add yourself to the video group as well.

[SOLVED]system fails to boot since adding udev rules for automounting

Hello
I have recently been trying to use udev rules to automount, and putting together stuff from the wiki, forums and general googling around have produced the following set of rules:
# automounts usb hdd and pendrives as usbhd-sdx; no messing around with
# volume labels or other confusing stuff
# matches all sdx devices except the internal hdd, sda
KERNEL=="sd[b-z]", NAME="%k", SYMLINK+="usbhd-%k", GROUP="users", OPTIONS="last_rule"
# imports filesystem information
ACTION=="add", IMPORT{program}="/sbin/blkid -o udev -p %N"
# creates mount points and sets up symlinks
ACTION=="add", KERNEL=="sd[b-z][0-9]", SYMLINK+="usbhd-%k", GROUP="users", NAME="%k"
ACTION=="add", KERNEL=="sd[b-z][0-9]", RUN+="/bin/mkdir -p /media/usbhd-%k"
ACTION=="add", KERNEL=="sd[b-z][0-9]", RUN+="/bin/ln -s /media/usbhd-%k /mnt/usbhd-%k"
# global mount options
ACTION=="add", ENV{mount_options}="relatime"
# filesystem-specific mount options (777/666 dir/file perms for ntfs/vfat)
ACTION=="add", ENV{ID_FS_TYPE}=="vfat|ntfs", ENV{mount_options}="$env{mount_options},gid=100,dmask=000,fmask=111,utf8"
# automount ntfs filesystem with ntfs-3g driver
ACTION=="add", KERNEL=="sd[b-z][0-9]", ENV{ID_FS_TYPE}=="ntfs", RUN+="/bin/mount -t ntfs-3g -o %E{mount_options} /dev/%k /media/usbhd-%k", OPTIONS="last_r$
# automount all other file systems
ACTION=="add", KERNEL=="sd[b-z][0-9]", ENV{ID_FS_TYPE}!="ntfs", RUN+="/bin/mount -t auto -o %E{mount_options} /dev/%k /media/usbhd-%k", OPTIONS="last_rule"
# unmounts and removes the mount points
ACTION=="remove", KERNEL=="sd[b-z][0-9]", RUN+="/bin/rm -f /mnt/usbhd-%k"
ACTION=="remove", KERNEL=="sd[b-z][0-9]", RUN+="/bin/umount -l /media/usbhd-%k"
ACTION=="remove", KERNEL=="sd[b-z][0-9]", RUN+="/bin/rmdir /media/usbhd-%k", OPTIONS="last_rule"
This seemed to be working very well unitl I tried to boot this morning and the boot process stopped at "processing UDev events" with the following message:
iTCO_wdt: Unexpected close, not stopping watchdog!
It pauses at this point for 10-15 seconds and then reboots.
Having searched a bit, I found the following similar post on the forums: http://bbs.archlinux.org/viewtopic.php?pid=459375
Which suggests that the problem might lie with this line:
ACTION=="add", IMPORT{program}="/sbin/blkid -o udev -p %N"
I have renamed the file so that it no longer has the udev .rules extension and now the system boots fine. Does anyone have any suggestions as to why the above rules might be causing this behaviour and how I might go about fixing it?
Thanks
Last edited by useradded (2010-07-02 22:58:14)

Hey falconindy
That was the final kick up the logical a$$ that I needed to get some kind of grip on udev rules. I now have a fully functional rule that applies only to /dev/sdxy and not to everything else as well, so no more boot trauma, THANK YOU.
I will mark this thread as solved and post my new rule for the benefit of anyone who might read this.
New rule (no boot problems):
# automounts usb hdd and pendrives as label or as usbhd-sdxy if no label present
# ensures the following is _only_ run for sdxy devices excluding internal hdd, sda
KERNEL!="sd[b-z][0-9]", GOTO="personal_usb_automount_settings_end"
# imports filesystem information
# provides access to following variables:
# ID_FS_UUID; ID_FS_UUID_ENC; ID_FS_VERSION; ID_FS_TYPE; ID_FS_VERSION; ID_FS_LABEL
# accessible via ENV{variable}; $env{variable}|%E{variable}
IMPORT{program}="/sbin/blkid -o udev -p %N"
# Get a label if present, otherwise name usbhd-%k
ENV{ID_FS_LABEL}!="", ENV{dir_name}="%E{ID_FS_LABEL}"
ENV{ID_FS_LABEL}=="", ENV{dir_name}="usbhd-%k"
# creates mount points and sets up symlinks
ACTION=="add", SYMLINK+="%E{dir_name}", GROUP="users", NAME="%k"
ACTION=="add", RUN+="/bin/mkdir -p /media/%E{dir_name}"
ACTION=="add", RUN+="/bin/ln -s /media/%E{dir_name} /mnt/%E{dir_name}"
# global mount options
ACTION=="add", ENV{mount_options}="relatime"
# filesystem-specific mount options (777/666 dir/file perms for ntfs/vfat)
ACTION=="add", ENV{ID_FS_TYPE}=="vfat|ntfs", ENV{mount_options}="$env{mount_options},gid=100,dmask=000,fmask=111,utf8"
# automount ntfs filesystem with ntfs-3g driver
ACTION=="add", ENV{ID_FS_TYPE}=="ntfs", RUN+="/bin/mount -t ntfs-3g -o %E{mount_options} /dev/%k /media/%E{dir_name}", OPTIONS="last_rule"
# automount all other file systems
ACTION=="add",ENV{ID_FS_TYPE}!="ntfs", RUN+="/bin/mount -t auto -o %E{mount_options} /dev/%k /media/%E{dir_name}", OPTIONS="last_rule"
# unmounts and removes the mount points
ACTION=="remove", RUN+="/bin/rm -f /mnt/%E{dir_name}"
ACTION=="remove", RUN+="/bin/umount -l /media/%E{dir_name}"
ACTION=="remove", RUN+="/bin/rmdir /media/%E{dir_name}", OPTIONS="last_rule"
# exit
LABEL=="personal_usb_automount_settings_end"
Last edited by useradded (2010-07-02 22:59:20)

Writing udev rules [SOLVED]

Hi guys.
I'm on a mission to run win7 on QEMU, and I also want to be able to use my USB-ports. Now, archwiki tells me to do this:
$ qemu-system-i386 -usbdevice host:vendor_id:product_id disk_image
You can find vendor_id and product_id of your device with lsusb command.
Note: If you encounter permission errors when running QEMU, see Udev#Writing udev rules for information on how to set permissions of the device.
I then made a file called 10-adm.rules both in, /etc/udev/rules.d and /usr/lib/udev/rules.d
In it I wrote:
KERNEL=="sdc[0-9]*", GROUP="storage"
My user with wich i lauch QEMU is in the group storage, and my usb always turns out as sdc*.
But it still gives me the permission erros.
Last edited by kimbo (2014-11-28 14:06:24)

now I renamed the files to 99 instead of 10, and I even overkilled it with the GROUP:= instead of GROUP=
tried:
udevadm control --reload-rules
udevadm trigger
But it still doesnt work
lsusb gives me:
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 006 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 004: ID 18a5:0302 Verbatim, Ltd Flash Drive
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 001 Device 002: ID 04f2:b23b Chicony Electronics Co., Ltd
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 004 Device 002: ID 04ca:3002 Lite-On Technology Corp.
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
And my qemu-launch-commad it:
qemu-system-x86_64 -m 1G -enable-kvm -cpu host -usbdevice host:18a5:0302 win7
oh, I dont really undestand what you by ' the group that /dev/sdc* ends up having'
Last edited by kimbo (2014-11-13 23:04:47)

Udev rules question

I have a pl2303 converter which appears as /dev/ttyUSB0, owned by root,uucp. I do not want to run minicom as root nor do I want to add myself to the uucp group, as either of these approaches lend themselves to potential issues to the system. Correct me if I'm wrong, but I would need this udev rule:
KERNEL=="ttyUSB[0-9]", MODE="0666"
I have not written any custom rules before, and therefore my /etc/udev/rules.d dir only contains two 75-*.rules.optional files. Obviously I should create a new *.rules file with this rule, but I'm concerned that a custom rules file will inhibit the implicit default rules in /lib/udev/rules.d and I'm unsure which integer to precede the custom rules file (greater than the default?).

You generally want your rules file to be first, so most people tend to do it at 10-....
If you build your rule correctly, it should not affect the other rules - you should uniquely identify the device and make sure it doesn't accidentally pick up other devices. You should be able to find an attribute (eg vendor and/or model) that helps.
This is useful, though a little out of date (the command to get the device information is wrong now, and some fields have changed).
http://www.reactivated.net/writing_udev_rules.html
See also man udev and man udevadm to update the information in there.

UDEV rules for Xilinx FPGA devices

I recently purchased a Digilent Nexys3 FPGA board with a Xilinx Spartan 5 FPGA on it. I finally got around to finding some time to using it, and setting up the development environment under Arch linux.
After spending ages reading around about various problems and workarounds, I finally managed to get Xilinx ISE Webpack installed and working, as well as Digilent Adept software for downloading the designs to the development board. One issue still remains though. I am unable to download the designs to the FPGA as a normal user, but only as root.
I'm pretty sure this is a udev rules problem. Digilent supplies a rules file (attached below) which is supposed to grant permissions to all users to access the FPGA devices over USB. I placed this rule file under /etc/udev/rules.d, but I am still unable to access the devices as normal user due to the following error:
$ djtgcfg enum
libusb couldn't open USB device /dev/bus/usb/001/003: Permission Denied.
libusb requires write access to USB device nodes.
$
I believe that udev rules syntax changed since the last time I had a look at it, so maybe the supplied rules file still uses the old syntax? Any suggestions to fixing this rules file (attached below)?
# 52-digilent-usb.rules -- UDEV rules for Digilent USB Devices #
# Author: MTA #
# Copyright 2010 Digilent Inc. #
# File Description: #
# This file contains the rules used by UDEV when creating entries for #
# Digilent USB devices. In order for Digilent's shared libraries and #
# applications to access these devices without root privalages it is #
# necessary for UDEV to create entries for which all users have read #
# and write permission. #
# Usage: #
# Copy this file to "/etc/udev/rules.d/" and execute #
# "/sbin/udevcontrol reload_rules" as root. This only needs to be done #
# immediately after installation. Each time you reboot your system the #
# rules are automatically loaded by UDEV. #
# Revision History: #
# 04/15/2010(MTA): created #
# 02/28/2011(MTA): modified to support FTDI based devices #
# Create "/dev" entries for Digilent device's with read and write
# permission granted to all users.
SYSFS{idVendor}=="1443", MODE="666"
ACTION=="add", SYSFS{idVendor}=="0403", SYSFS{manufacturer}=="Digilent", MODE="666", RUN+="/usr/local/sbin/dftdrvdtch %s{busnum} %s{devnum}"
# The following rules (if present) cause UDEV to ignore all UEVENTS for
# which the subsystem is "usb_endpoint" and the action is "add" or
# "remove". These rules are necessary to work around what appears to be a
# bug in the Kernel used by Red Hat Enterprise Linux 5/CentOS 5. The Kernel
# sends UEVENTS to remove and then add entries for the endpoints of a USB
# device in "/dev" each time a process releases an interface. This occurs
# each time a data transaction occurs. When an FPGA is configured or flash
# device is written a large number of transactions take place. If the
# following lines are commented out then UDEV will be overloaded for a long
# period of time while it tries to process the massive number of UEVENTS it
# receives from the kernel. Please note that this work around only applies
# to systems running RHEL5 or CentOS 5 and as a result the rules will only
# be present on those systems.
Thanks!
-Igor
Last edited by UQ-igor (2011-12-24 03:56:10)

Try this
Use ATTR or ATTRS.
-SYSFS{idVendor}=="1443", MODE="666"
-ACTION=="add", SYSFS{idVendor}=="0403", SYSFS{manufacturer}=="Digilent", MODE="666", RUN+="/usr/sbin/dftdrvdtch %s{busnum} %s{devnum}"
+ATTRS{idVendor}=="1443", MODE="666"
+ACTION=="add", ATTRS{idVendor}=="0403", ATTRS{manufacturer}=="Digilent", MODE="666", RUN+="/usr/sbin/dftdrvdtch %s{busnum} %s{devnum}"
+ATTR{idVendor}=="1443", MODE="666"
+ACTION=="add", ATTR{idVendor}=="0403", ATTR{manufacturer}=="Digilent", MODE="666", RUN+="/usr/sbin/dftdrvdtch %s{busnum} %s{devnum}"

3 unable to open '/etc/udev/rules.d/':m error during booting

After a recent upgrade to the kernel26-2.6.28.5-1-i686, I got the above error during booting in one of the laptop.
My search did not give any positive solution for this till now. I read that now udev gets its rules from /lib/udev/rules.d.
But still rules files in /etc/udev/rules.d and /lib/udev/rules.d are different. I know that any symllink work around will not work in this case, I tried but as expected the error message appears again.
So, I roll back to kernel26-2.6.28.4-1-i686 and everything back to normal.
Any Idea to resolve this? (package is not broken as I use the same mirror for another PC and in that I am not getting any error)
Last edited by kgas (2009-02-19 16:10:23)

This issue got resolved on its own with the latest kernel (kernel26-2.6.28.6-1-i686) update. I will leave this topic for some more time to see others response...

Using disk label on udev rules?

Hello,
Until just a few hours ago, when I plugged my external hard drive in I had to run a script to mount it with sudo, as I couldn't get it automounted. Then I saw a workaround in some thread over here, using the udev rules proposed at the bottom of the Udev wiki article.
After a reboot, and after some hours I didn't remember I had created that rule, so I went to konsole and run my script, which returned "Disc not found". That's because it checks there is that device under /dev/disk/by-label/. As I checked afterwards, the disk was automounted on some directory under /media, but it there were no symlinks to it nor to my pendrive under /dev/disk. This is a problem as my script did something like "mount /dev/disk/by-label/$LABEL /mnt/$LABEL", and Amarok's collection is under there, so if I change the collection directory every time I reboot (or every time I plug the disk) it might be mounted on a different directory (current one is /media/usbhd-sdb1), depending on the order I plug the usb storage devices, or even random if more than one is plugged on boot (not sure about this).
I suppose this is because when I plug the disk (or pendrive) udev applies the first rule it finds, so I think a possible workaround for my problem is to change the mount directory on the udev rule. And here is where my question comes: I see NAME="%k" and then %k used as the device name under /dev (in this case sdb1). Could I use some variable to keep the label of the disk and then mount it on /mnt/$LABEL, for example? I think this way I couldn't automount any device without label, but every pendrive I've seen is labeled by default so that wouldn't be a problem.
Thank you for reading that bible, and thank you again if you answer
Bye!
Edit: maybe this thread should be under Kernel & Hardware. If so, please move it
Last edited by Surgat_ (2008-09-02 23:52:47)

I think what you were looking for may have been $env{ID_FS_LABEL}. I've got the following as my udev rules for getting automatically mounted drives with the mountpoint set to the partition label instead of just a number. Not particularly robust when there are two partitions with the same label, but I haven't gotten around to writing a fix for it yet. If it works for you, maybe we should put it on the wiki.
KERNEL=="sd[b-z]", NAME="%k", SYMLINK+="usb%m", GROUP="users", OPTIONS="last_rule"
ACTION=="add", KERNEL=="sd[b-z][0-9]", SYMLINK+="usb%n", GROUP="users", NAME="%k"
ACTION=="add", KERNEL=="sd[b-z][0-9]", RUN+="/bin/mkdir -p /mnt/usb%n"
ACTION=="add", KERNEL=="sd[b-z][0-9]", RUN+="/bin/ln -s /mnt/usb%n /media/$env{ID_FS_LABEL}"
ACTION=="add", KERNEL=="sd[b-z][0-9]", PROGRAM=="/lib/udev/vol_id -t %N", RESULT=="vfat", RUN+="/bin/mount -t vfat -o rw,noauto,flush,quiet,nodev,nosuid,noexec,noatime,dmask=000,fmask=111 /dev/%k /mnt/usb%n", OPTIONS="last_rule"
ACTION=="add", KERNEL=="sd[b-z][0-9]", RUN+="/bin/mount -t auto -o rw,noauto,sync,dirsync,noexec,nodev,noatime /dev/%k /mnt/usb%n", OPTIONS="last_rule"
ACTION=="remove", KERNEL=="sd[b-z][0-9]", RUN+="/bin/umount -l /mnt/usb%n"
ACTION=="remove", KERNEL=="sd[b-z][0-9]", RUN+="/bin/rm /media/$env{ID_FS_LABEL}"
ACTION=="remove", KERNEL=="sd[b-z][0-9]", RUN+="/bin/rmdir /mnt/usb%n", OPTIONS="last_rule"

Stranger udev.rules error

I am trying to use my pda (Palm Tungsten E, but that should be irrelevant) under KDE. Well, I do not get that far!
According to the standard udev.rules the PDA should be seen as a tty device with the rule
KERNEL=="ttyUSB[0-9]*", NAME="tts/USB%n"
Later there is a rule specifically for Palm Pilots
BUS=="usb", KERNEL=="ttyUSB[0-9]*", SYSFS{product}="Palm Handheld... ", SYMLINK+=pilot
(Should not the later rule be disregarded since a valid rule has already been encountered?)
However, I get nothing. The messages.log tells
usb 3-2: new full speed USB device using uhci_hcd and address 2
usb 3-2: configuration #1 chosen from 1 choice
That's all. No devices (not even /dev/tts/USB1), no module loaded, nothing (that I can detect).
Connections etc are ok - in Slackware the PDA is connected at /dev/tts/USB1 (later rule disregarded!) and can be used as root. So no positive clue there.
So, what is different in Arch? HAL? and most important, what should I do?

Ok. Add usbserial and visor modules to the MODULES in /etc/rc.conf (as it seems they are not loaded automatically). Restart machine or run /etc/udevstart and replug your palm. Can you see /dev/tts/USBx now at least? (x might be different number each time you plug in your palm).
Note that the second standard rule that should create /dev/pilot link will not work for you because SYSFS{product} is different than yours and will not match with the rule.
Now create /etc/udev/rules.d/010.udev.rules file and put these rules inside (there are two lines only! watch out for forum line wraps):
KERNEL=="ttyUSB[0-9]*", NAME:="tts/USB%n", GROUP:="users"
BUS=="usb", KERNEL=="ttyUSB[0-9]*", SYSFS{idVendor}=="0830", SYSFS{idProduct}=="0060", SYMLINK:="pilot"
Restart machine or start /etc/udevstart and replug your palm. It should create /dev/tts/USBx and /dev/pilot (which you should use in palm applications). The owner for /dev/tts/USBx should be root:users. If it needs special write permissions you can add
, MODE:="0660"
to the first rule in /etc/010.udev.rules file. You can of course change "users" to any group you like just remember to add your user to this group.
BTW If you have too many modules loaded you can customize /etc/mkinitrd.conf - check wiki for details.

Udev rules or buggy hardware driver?

hello out there,
here's my situation, after setting this rule for my graphics tablet...
KERNEL=="event*", SYSFS{manufacturer}="AIPTEK International Inc.",
NAME="input/aiptektablet", MODE="0644"
i revieve the following on next boot...
The superblock could not be read or does not describe a correct ext2
filesystem. If the device is valid and it really contains an ext2
filesystem (and not swap or ufs or something else), then the superblock
is corrupt and you might try running e2fsck with an alternate superblock
E2fsck -b 8193 <device>
after removing the rule and rebooting the drive check goes fine and everything runs smooth again.
note:The idea to remove the aiptek udev rule came when I noticed all udev events in /dev/disk/* pointed to the aiptek tablet instead of their usual sda sdb sdc sdd hda1 hda2 etc.
I would still like a consistent input point for my tablet someday, but this situation seems over my head.
any ideas?
thanks

Thanks for your answer Rexilion.
The author of the module, and the rule, suggest to put the rule in /etc/udev/rules.d under the file name 90-logitech-g710-plus.rules. So that is what I have tried up to now, along with putting the file in /usr/lib/udev/rules.d and changing the preceding number to 10 instead of 90. The only thing that my tinkering has done is that changing ACTION=="add" to ACTION=="add|change" has allowed udevadm trigger to apply the rule. Of course, I could use that to generate some real ugly workaround, but I would rather not. Otherwise, from what I have seen, you are probably right about a race condition, but I have been in the dark as to how to resolve it.
At this point, I have put the rule in /usr/lib/udev/rules.d and used "mkinitcpio -p linux" to rebuild the image, but it doesn't make a difference as to what default rules are in the image. In mkinitcpio.conf, I have the udev hook, but I figure that is kind of defautl, am I to modify that hook? What am I missing? I have read and reread the wiki about udev, and mkinitcpio, but I'm very unclear as to how to add the rule for the keyboard to the default set.
@jasonwryan: Perhaps I'm being stupid, but is loading a module and binding it to the keyboard considered a long running process? I would have thought it is an event which is very short and needs no further attention or resources once it is done.

I don't understand how udev rule work

I had a problem with xsane only running as root a while ago and so copied a udev rule into /etc/udev/rules.d
The scanner was then detected and added to the scanner group and I could use it. I've very recently had to do the same with a different scanner. But in both cases the same rule already existed in /lib/udev/rules.d/53-sane.rules
Why was this not used in the first place? Is there something wrong with my udev that it would not look in /lib/udev/rules.d and find the correct rule... or are people supposed to copy them over from /lib/udev/rules.d to /etc/udev/rules.d ?!?

http://reactivated.net/writing_udev_rules.html
Read that front to back and you'll know how it works .

MDNS/Bonjour switchs randomly between multiple NICs

Dear Community,
I have a problem with a MacServer and multiple NICs...lets name it MSERV
NIC1 goes into the WWW (194.x.x.x) (Protected with ASA firewall)
NIC2 goes into LAN (172.x.x.x) (no firewall)
I need the WWW NIC for ProfileManager and also the LAN NIC for filesharing and TMB.
Both IPs are reachable from our LAN. DNS is good (nslookup shows up as expected) but mDNS/Bonjour makes some problems:
If I do ping MSERV.local sometimes it resolves the 194 and sometimes the 172 address.
This is a big problem for our time machine server backups! When clients choose the 194-path when they discover their backup target on the server, we generate a HUGE load on our ASA firewall.
How can I force Bonjour only to use the 172er LAN NIC.
Or did I get something dramatically wrong?
Any suggestions?

Tearjerker wrote:
Thank you for your reply!
So you mean to disable Bonjour on the server and set up DNS name as TM target?
Problem is here: How can I do this? Only way I know to connect to my backup server on clients is
System Prefs. -> TM -> Choose Disk. And as far as I know is this a Bonjour connection, right?
When I disable Bonjour service on the server (already tried!) my clients are unable to connect to their target.
Any thoughts?
You should not need to disable Bonjour. Bonjour aka. mDNS uses the .local domain suffix if you use a hostname like MSERV.local it will use Bonjour to try and find the TCP/IP address. (We will ignore the possibility someone might use the .local domain in an Active Directory setup.)
If your using a full-blown DNS e.g. server.domain.com then this is nothing to do with Bonjour and where it will point to is up to your DNS server. In this case you would point it to the 172.x.x.x IP address of the server.
As you want the same server to be contactable via the WAN interface as well for the rest of the work to access what you will want to do is have a 'split-horizon' DNS setup. With this the internal machines use an internal DNS server which points to the internal IP address of the server. You will also have the same domain hosted externally and the same server name but pointing to the external WAN IP address. What you do then have to be careful about is to ensure that all the external host names in that domain are defined also on the internal DNS server even if they are servers hosted externally. For example if you have a www.domain.com server hosted externally so it only has an external (public) IP address then you will still need to define this on your internal DNS server so your internal machines know where to find it, since your internal machines will be using your internal DNS server. If you fail to do this then while the outside world using an external DNS server may be able to access your www server, your internal machines would not be able to because your internal DNS server would otherwise say it does not exist.

Domain Agents and E6900 with multiple NICs.

Hello,
I have some issues configuring Sun management Center 3.6 (and 3.6.1) on a network with this characteristics:
2 x E6900 (with 2 Domains each one)
1 x v210
The Server Component of Sun Management Center is installed on the v210. In this machine are installed the Platform Agents of the 2 System Controllers of the 2 E6900.
Each domain has installed de Sun Management Center Agent.
The network configuration of the Domains is this:
Each domain has 3 NIC on 3 different network interfaces, and each NIC has its own hostname. One of this NIC is destinated to the administration network to be used for the Sun Management Center. This nic, as I said before, has its own hostname but it its different of the nodename (the node name is the same as the public interface of the domain, not the administration interface).
The Domains does not use DNS, and the information included on the /etc/hosts is related to local NICs and the NICs of the other Domains, but anyone contains the NIC of the admin interface of other domain nor the NIC of the v210.
The v210 does not has DNS neither, and has only their own information on /etc/hosts.
All Domains has Solaris 9, and v210 has Solaris 10.
Basically the problem is:
"The Domain Agents cannot communicate with SunMC Server", when I manually add each agent. The discovery process doesn't work neither.
I checked the domain-config.x of one domain and it shows:
agent = {
agentServer = "venwas1"
logicalAddressMode = "ip"
snmpPort = "1161"
"venwas1" is the hostname of the public interface, nor the administration interface.
I changed the value for the hostname of the administration interface with no success; then changed for the ip address of the administration interface; and no success neither.
The default platform agent works, but does not appers with the corrent icon on the SunMC console. Its appears like a folder icon, and the documentation said that the Platform Agents is represented with a specific Icon.
Please, anyone can help me ?
Thanks in advance.

Hi Alejandro,
I checked the domain-config.x of one domain and it
shows:
agent = {
agentServer = "venwas1"
logicalAddressMode = "ip"
snmpPort = "1161"
nwas1" is the hostname of the public interface, nor
the administration interface.
I changed the value for the hostname of the
administration interface with no success; then
changed for the ip address of the administration
interface; and no success neither.If you just change the hostname or IP, but don't do anything else, it won't work as the Agent has already created security keys for that old entry and needs to be reseeded. Change the entry to the IP that's on the same network as your SunMC Server, then run this command on the Agent while it's turned off:
/opt/SUNWsymon/sbin/es-run base-usm-seed.sh -s <SEED> -c agent
..replacing <SEED> with the same seed/password you used on your SunMC Server. Then start the Agent back up, wait for it a couple minutes, and try to make an icon for it again.
The default platform agent works, but does not appers
with the corrent icon on the SunMC console. Its
appears like a folder icon, and the documentation
said that the Platform Agents is represented with a
specific Icon.If the icon is incorrect then your SunMC Server is missing some files (or you've found a bug). You'll want all these addons installed on your Server:
Dynamic Reconfiguration for Sun Fire High-End and Midrange Systems
Sun Fire Midrange Systems Platform Administration
Sun Fire Midrange Systems Domain Administration
If you think you may be missing some (look in /var/opt/SUNWsymon/install/install.[timestamp] to see how you answered questions during your last install) you can run /opt/SUNWsymon/sbin/es-inst again and point it to your distribution media and it will only ask about packages that may be missing. If you do end up adding in some of the Midrange addons then a simple Console restart should refresh your Platform icons.
Regards,
[email protected]
http://www.HalcyonInc.com

Mail Adapter - Multiple mail ID and multiple mail servers config.

Hi All
I am doing BPM synch scenario in which i get the response from SAP box and send the response via email adapter. I am using mail.xsd and doing mail config. in message mapping. However in the TO field i am able to give only one email ID. If i give multiple email ID's mail is not received. I tried comma and semi-colon as separator. Still not working? I have two questions in configuring TO option:
1) How to send to multiple id's? I am using Lotus Notes.
2)How to send to multiple mail servers? I have to send to Lotus Notes id's and outlook express id's also simultaneously.
Thanks for your help in advance
Warm Regards
Samuel

Hi,
Please find here with some observations about it,
1) How to send to multiple id's? I am using Lotus Notes.
If you have specified an IMAP server under URL, the message is saved in the specified folder but is not sent to the receiver specified under To.
Then even if Under To, you had specified the e-mail address that will receive the message would be separated with a semicolon. It will not work.
Please verify about it .
The below link will also help you to verify if there is anything missing
Mail Adapter (XI) - how to implement dynamic mail address
/people/michal.krawczyk2/blog/2005/03/07/mail-adapter-xi--how-to-implement-dynamic-mail-address
BPM:Single Sender and Multiple Receivers based on synchronous
exchange(switch) part-1
/people/prasadbabu.nemalikanti3/blog/2006/03/10/bpmsingle-sender-and-multiple-receivers-based-on-synchronous-exchangeswitch-part-1
Generic Message Interface in SAP Exchange Infrastructure Email Integration Scenarios
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/00d5a235-4803-2a10-f682-889d67c69975
(If your using Alert Framework then)
If you want to send it to multiple email addresses and all email addresses are user of XI then you can define "Role" and attach that role to everyuser and make this role as receipent of alert .
Thanks
Swarup
Thanks
Swarup

Udev rules config and multiple NICs

Similar Messages

Maybe you are looking for