There two RTL811/8168BB NIC cards installed both running the r6168 driver. I have changed /etc/network/interfaces file to reflect two nics (eth0 external DHCP and eth1 static). In the Advance/Network/settings , Linuxmce recognizes only eth1 and it creates a virtual NIC at eth1:1. My changes to etc/network/interface get re-wrtten on re-boot to eth1 and eth1:1 consistent with the settings in Advanced/Network/Settings. Both cards are functional as verfied by switching eth0 and eth1 in /etc/udev/rules'/rules.d/70-persistent-net.rules. What step am I missing?
Hi,
Is there an on-board NIC and an installed card? What's the hardware arrangement?
Thanks.
Its an ASUS AT5IONT-I motherboard with one integrated NIC and a second NIC installed on the PCI-e slot.
Something went wrong during setup, and the database holds the belief that you only have one nic. I will see if I can figure out how to fool it into believing the truth.
For now, make your interfaces file reflect the proper interface names (eth0 eth1), however make them reflect the opposite function (put your internet on internal and vise versa), go into webadmin, advanced/networking and at the bottom of the page choose "swap interfaces", and see if that does not write the change.
I made eth0 internal and eth1 external in the interfaces file. When I swapped the NIC cards using WebAdmin, it caused the system to reboot. The system then overwrote the interface file to eth1:1 to dhcp (external) and eth1 to static.
Can I please see the output of cat /etc/udev/rules.d/70-persistent-net.rules
Thanks for you asisstance: Here is the output:
# This file maintains persistent names for network interfaces.
# See udev(7) for syntax.
#
# Entries are automatically added by the 75-persistent-net-generator.rules
# file; however you are also free to add your own entries.
# PCI device 0x10ec:0x8168 (r8169)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="bc:ae:c5:9a:1a:7a", ATTR{type}=="1", KERNEL=="eth*", NAME="eth0"
# PCI device 0x10ec:0x8168 (r8169)
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="00:0a:cd:1e:64:6c", ATTR{type}=="1", KERNEL=="eth*", NAME="eth1"
Maybe you've hit a similar problem to one I had, if set-up didn't complete successfully with two nics, as l3mce said you may have incorrect info in the db.
See http://forum.linuxmce.org/index.php/topic,11576.0.html you may have two conflicting records on the network info expected in your install.
-Coley.
Just so I understand this right,...
Because I've been messing around with MythTV settings, playback configuration user setting, etc., before getting to the network configuration...
Q: Which nic should the Internet connection be on, and which should the internal network be on,... ? (onboard versus card)
Q: And can LinuxMCE be connected to a router which Does DHCP for any direct connectly devices?
(Specifically, I have a Micsaverde Vera v.1, and my Iconia A500 tablet does not like my other wireless router)
... and have LinuxMCE do DHCP for anything connected to IT for internet and internal cross talk?
I know my Linksys WRT54GS running Gargoyle can be set to allow for other DHCP connections,... and turn off DHCP.
I don't know how easily the Vera can be configured to do the same.
Until I get a chance to reconfigure the network, Vera connects first, then the other router to it, and currently then to all other machines including my LinuxMCE Core.
I briefly tried; Cable modem to LinuxMCE Core to Vera to Linksys, but everything lost its Internet connection but the Core (The Vera may have been running DHCP, but the Linksys had it turned off).
I intend to move my Gargoyle-powered router inside, with LinuxMCE sitting between the Vera and the Linksys. So, it would be; Cable Modem to Vera to LinuxMCE Core to Linksys/Gargoyle router (with DHCP turned off) to everything else.
Will the above work,... or do I have to get more aggressive with the setup of the Vera?
When I first loaded the system, I originally had an Intel Gigbit NIC card. After LinuxMCE installation and connecting a NAS to the internal NIC I realised that the NIC was'nt recoginized since the system failed to acknowledge the NAS. On giving up getting that card to work , I installed the Realtek NIC. I know this card is functional.
How do I correct the database?. Specifically how do I log into the database and what SQL statements do I need to execute? (The link did'nt provide enough information).
I ran it via phpMyAdmin, I ran the query via the web interface.
or you can do the following and see if you come back with more than one record.
from a terminal
mysql -u root -D pluto_main <enter>
next at the mysql> prompt
SELECT IK_DeviceData FROM Device_DeviceData WHERE FK_DeviceData=32; <enter>
does it come back with just entries for eth0 and eth1
-Coley.
The query returns:
eth1:1,dhcp|eth1,192.168.80.1,255.255.255.0
one row in set
ok as expected your db has the incorrect info, that first entry needs to point to eth0 not eth1:1.
Does the command "ip addr" return the expected results? two cards eth0 and eth1?
-Coley.
Here is the result of ip addr:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0a:cd:1e:64:6c brd ff:ff:ff:ff:ff:ff
inet 192.168.1.20/24 brd 192.168.1.255 scope global eth1
inet6 fe80::20a:cdff:fe1e:646c/64 scope link
valid_lft forever preferred_lft forever
3: eth0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast state DOWN qlen 1000
link/ether bc:ae:c5:9a:1a:7a brd ff:ff:ff:ff:ff:ff
inet 192.168.80.1/24 brd 192.168.80.255 scope global eth0
4: pan0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN
link/ether be:6c:94:6d:9a:5f brd ff:ff:ff:ff:ff:ff
/jph
you should be able to re-run the script to detect the actual network setup and have this info inserted into the db.
If you run
sudo /usr/pluto/bin/Network_Config.sh --ext-dhcp
this script should insert the appropriate info into the record and restart your network services with the new settings.
You can verify you now have two cards detected via the webadmin and they have the info as expected.
-Coley.
I ran the script which generated two error messages:
Stopping DHCP server - [FAIL]
Failed to send 'clear' to local ypserver RPC: Program not registeredUpdating passwd.byusid.
Followed by msg: Now you can run ypinit -s dcerouter on all slave services.
In WebAdmin under Advanced/Network only eth1:1 (dhcp) and eth1 (static) were recognized. No NIC were displayed in the device tree. The file /etc/network was re-written to eth1:1 (dhcp) and eth1 (stat). I changed this back to eth0 (dhcp) and eht1 (static) and then restated the network to re-establish my internet connection.
/jph
I found a very old post on a blog. I don't know if it still applies or if it will lead to stability issues:
To recognize the second NIC:
UPDATE Device_DeviceData SET IK_DeviceData="eth0,dhcp|eth1,192.168.80.1,255.255.255.0" WHERE FY_DeviceDeviceData=32;
/jph
Just got your update - was mid reply.
That SQL is what the script tries to do behind the scenes.
either run the update if you are confident to do it from the cli
Or if you have phpMyAdmin installed you can edit the record from there.
-Coley.
I executed the update query directly on the database and verfied that the row was updated. I re-loaded the orbitor and now eth0 (dhcp) and eth1(static) are both recognized in WebAdmin. In the Advanced/Network/Network settings I was able the refresh the the IP on eth0. So far so good.
I reconnected my NAS on eth1 where I have media from a previous installation. If the dhcp server is active on eth1, should'nt the system recognize the NAS instaneously or will this take sometime?
/jhp
I assume you meant reloaded the router, as in dcerouter, and not Orbiter.
Are there shares already exported on the NAS?
If the NAS doesn't get picked up make sure it isn't being ignored - Advanced->Configuration->Unknown Devices
If this issue continues consider starting a new Thread so it doesn't get lost under the wrong title.
-Coley.
Thanks for your help and your patience.
I will start a new thread, if I continue to experience problems with the NAS.
//jph
Quote from: JaseP on February 28, 2012, 04:27:22 PM
Just so I understand this right,...
Because I've been messing around with MythTV settings, playback configuration user setting, etc., before getting to the network configuration...
Q: Which nic should the Internet connection be on, and which should the internal network be on,... ? (onboard versus card)
Makes no difference to the system. Typically the onboard will be detected first, unless it is one of the blacklisted prior to driver support when kubu loads... then it will be backwards. LMCE doesn't care one way or the other. It figures out which one has internet coming in and makes that external, and spits dhcp on the other. However if one is gigabit, and one is not, I would recommend specifying the gigabit for your internal network.
Quote from: JaseP on February 28, 2012, 04:27:22 PMQ: And can LinuxMCE be connected to a router which Does DHCP for any direct connectly devices?
(Specifically, I have a Micsaverde Vera v.1, and my Iconia A500 tablet does not like my other wireless router)
... and have LinuxMCE do DHCP for anything connected to IT for internet and internal cross talk?
There is magic involved with this (meaning I cannot explain why), but if you have multiple things spitting DHCP attached to lmce, odd behavior ensues. It creeps through wan ports... it tunnels through things I do not understand it to be able to in an attempt to find life. It is very greedy and prolific.
Quote from: JaseP on February 28, 2012, 04:27:22 PMI know my Linksys WRT54GS running Gargoyle can be set to allow for other DHCP connections,... and turn off DHCP.
I don't know how easily the Vera can be configured to do the same.
Until I get a chance to reconfigure the network, Vera connects first, then the other router to it, and currently then to all other machines including my LinuxMCE Core.
I briefly tried; Cable modem to LinuxMCE Core to Vera to Linksys, but everything lost its Internet connection but the Core (The Vera may have been running DHCP, but the Linksys had it turned off).
I intend to move my Gargoyle-powered router inside, with LinuxMCE sitting between the Vera and the Linksys. So, it would be; Cable Modem to Vera to LinuxMCE Core to Linksys/Gargoyle router (with DHCP turned off) to everything else.
Will the above work,... or do I have to get more aggressive with the setup of the Vera?
Vera is pluto. Putting them both on the same wire, they are gonna bump heads. Tkmedia has the two playing nicely I believe. He will have to give you more details. It will have to be a very non-standard configuration.
I do not personally see the point in having both plutos, and hiding off things from DCERouter is counterintuitive to me... LMCE cannot take advantage of resources on the machines, and you cannot control LMCE through those interfaces. It is designed to just integrate everything seamlessly. You are a grown up, and can use your equipment however you see fit obviously... I am certainly not trying to get into anything by asking... but why do you want to compartmentalize something which exists to provide layered transparency?