Solaris Routing


 
Thread Tools Search this Thread
Operating Systems Solaris Solaris Routing
# 1  
Old 11-08-2012
Solaris Routing

Hello All,

Problem is that redundancy doesn't work, probably a routing issue I need to get sorted out When 172.29.11.x net is down, neither node3 nor node4 is trying 172.29.12.x network,The problem lies in network 172.29.13.x

Code:
 if 172.29.11.x network is down, node3/node4 should route to network 172.29.13.x over 172.29.12.x network
  if 172.29.12.x network is down,node3/node4 should route to network 172.29.13.x over 172.29.11.x network

As we can see in the routing tables we do have two routes(through 12.x and 11.x) available for 13.x network .

Can you please help me understand is there way i am missing or need to add any other static routes to fix this issue


Code:
NODE3: ACTIVE
  outing Table: IPv4
    Destination           Gateway           Flags  Ref     Use     Interface
  -------------------- -------------------- ----- ----- ---------- ---------
  default              172.29.1.1           UG        1      90060
  172.29.1.0           172.29.1.51          U         1      38606 nge0
  172.29.1.0           172.29.1.51          UG        1          0
  172.29.11.0          172.29.11.8          U         1      82940 e1000g0
  172.29.11.0          172.29.11.15         U         1          0 e1000g0:1
  172.29.12.0          172.29.12.8          U         1      52849 e1000g1
  172.29.12.0          172.29.12.15         U         1          0 e1000g1:1
  172.29.13.0          172.29.11.1          UG        1      53927
  172.29.13.0          172.29.12.1          UG        1          0
  224.0.0.0            172.29.1.51          U         1          0 nge0
  127.0.0.1            127.0.0.1            UH       31    8824669 lo0
  
  og 
  
  NODE4: standby
  
  Routing Table: IPv4
    Destination           Gateway           Flags  Ref     Use     Interface
  -------------------- -------------------- ----- ----- ---------- ---------
  default              172.29.1.1           UG        1      85888
  172.29.1.0           172.29.1.52          U         1      38120 nge0
  172.29.11.0          172.29.11.9          U         1      60344 e1000g0
  172.29.12.0          172.29.12.9          U         1      52478 e1000g1
  172.29.13.0          172.29.11.1          UG        1        178
  172.29.13.0          172.29.12.1          UG        1          0
  224.0.0.0            172.29.1.52          U         1          0 nge0
  127.0.0.1            127.0.0.1            UH       25    6088816 lo0


Best Regards
Nadeem

Last edited by Scott; 11-08-2012 at 06:29 PM.. Reason: Please use code tags (and less formatting).
# 2  
Old 11-08-2012
As a quick stab at this (and I've only had a brief look at AIX HA) I'm a little confused as to what you have. Code tags would make your output easier to read too.

Can you confirm:-
  • The primary/boot IP addresses on each card on each server, and that they can see each other.
  • Any permanent IP addresses added to the servers so you can guarantee which you are using.
  • The service IP addresses you have.


I think that you have public connections to a 172.29.1.xxx address, and that is the one that needs to be protected. If subnets 11, 12 or 13 drop off, you want to know about it and perhaps take some action. The subnet 172.29.1.xxx should be fault tolerant.

You do seem have a route defined from subnets 11/12 to 13, but I wonder why. Perhaps i am misreading what you intend.

Can you paste the output from ifconfig -a or whatever it is on Solaris these days. I've been over 10 years since I've had one to work on, lovely as they are. We would need to see the IP addresses, aliases (aliai? My Latin is not good) netmasks, gateways (contents of /etc/defaultrouter?)

A quick brief of which is which would be good too.


Thanks, in advance,
Robin
Liverpool/Blackburn
UK
# 3  
Old 11-08-2012
Here is the ifconfig -a

Code:
#Node3= Active 
#ifconfig -a
lo0: flags=2001000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu 8232 index 1
        inet 127.0.0.1 netmask ff000000
e1000g0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
        inet 172.29.11.8 netmask ffffff00 broadcast 172.29.11.255
        ether 0:21:28:5e:5b:a0
e1000g0:1: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
        inet 172.29.11.15 netmask ffffff00 broadcast 172.29.11.255
e1000g1: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 3
        inet 172.29.12.8 netmask ffffff00 broadcast 172.29.12.255
        ether 0:21:28:5e:5b:a1
e1000g1:1: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 3
        inet 172.29.12.15 netmask ffffff00 broadcast 172.29.12.255
nge0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 4
        inet 172.29.1.51 netmask ffffff00 broadcast 172.29.1.255
        ether 4:4b:80:80:80:3
#
 
Node4 (standby node):
# ifconfig -a
lo0: flags=2001000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu 8232 index 1
        inet 127.0.0.1 netmask ff000000
e1000g0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
        inet 172.29.11.9 netmask ffffff00 broadcast 172.29.11.255
        ether 0:21:28:5e:5b:14
e1000g0:1: flags=1000842<BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
        inet 172.29.11.15 netmask ffffff00 broadcast 172.29.11.255
e1000g1: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 3
        inet 172.29.12.9 netmask ffffff00 broadcast 172.29.12.255
        ether 0:21:28:5e:5b:15
e1000g1:1: flags=1000842<BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 3
        inet 172.29.12.15 netmask ffffff00 broadcast 172.29.12.255
nge0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 4
        inet 172.29.1.52 netmask ffffff00 broadcast 172.29.1.255
        ether 0:21:28:5e:5b:12


Last edited by Scott; 11-08-2012 at 06:29 PM.. Reason: Code tags...
# 4  
Old 11-13-2012
Can anyone look into this..?
# 5  
Old 11-14-2012
I see a conflict with both hosts wanting to be 172.29.12.15.

Can you answer the other questions so we have the full picture please.



Robin
Liverpool/Blackburn
UK
# 6  
Old 11-20-2012
Thanks Robin, those are VIP used for fail-over for active and standby.

Just one more thing to re-phrase and a query.

We have two static entries in the route table for 13.x network via 11.x and 12.x. since 11.x is the first entry it generally prefer this path and even its try to route the packets its goes down(via 11.x).

wanted to know when it start to routes packets to 13.x to via 12.x and when will it assume that 11.x network is down and it has seek another best route to route the packets via 12.x .(below is the snippet of route table again)

Code:
Routing Table: IPv4
Destination Gateway Flags Ref Use Interface
-------------------- -------------------- ----- ----- ---------- ---------
default 172.29.1.1 UG 1 90060
172.29.1.0 172.29.1.51 U 1 38606 nge0
172.29.1.0 172.29.1.51 UG 1 0
172.29.11.0 172.29.11.8 U 1 82940 e1000g0
172.29.11.0 172.29.11.15 U 1 0 e1000g0:1
172.29.12.0 172.29.12.8 U 1 52849 e1000g1
172.29.12.0 172.29.12.15 U 1 0 e1000g1:1
172.29.13.0 172.29.11.1 UG 1 53927
172.29.13.0 172.29.12.1 UG 1 0
224.0.0.0 172.29.1.51 U 1 0 nge0
127.0.0.1 127.0.0.1 UH 31 8824669 lo0


Thanks
Nadeem

Last edited by DukeNuke2; 11-20-2012 at 09:50 AM..
# 7  
Old 11-21-2012
Sorry, I didn't notice that only one is listed as UP. On AIX high-availability, the IP address is assigned only when failover is detected, hence my confusion.

I'm not sure that your 172.29.11.x addresses will re-route necessarily. If the service address you are offering is 172.29.12.15, then this should be the address for the default route out. You currently may have clients that target 172.29.12.15, but the return seems to be via 172.29.1.1 using card nge0 on each.

If you, from your own desk open a simple telnet or ssh connection to 172.29.12.15, what does your session show up as with the following:-
Code:
netstat -na | grep <your desk ip address>
netstat -nr | grep <your desk ip address>

I think that your routing may be all going out via nge0 except what is on the local subnet of each of the other cards.

Hopefully, I'm worrying unecessarily, but then I'm a bit stumped where you are getting stuck.


Robin
Liverpool/Blackburn
Login or Register to Ask a Question

Previous Thread | Next Thread

9 More Discussions You Might Find Interesting

1. UNIX for Advanced & Expert Users

Solaris 10 routing table issues

Hello Hope someone can help with this problem. We are running Solaris 10 with a current kernel patch of 142900-09. We appear to be getting a serious issue with the routing table as shown below: Output from netstat -rnv Destination ....Mask ............Gateway ........Device... (2 Replies)
Discussion started by: gregsih
2 Replies

2. Solaris

solaris routing between two networks

ce0: flags=1100843<UP,BROADCAST,RUNNING,MULTICAST,ROUTER,IPv4> mtu 1500 index 2 inet 10.162.212.132 netmask ffffff00 broadcast 10.162.212.255 ether 0:14:4f:55:82:9 ce1: flags=1100843<UP,BROADCAST,RUNNING,MULTICAST,ROUTER,IPv4> mtu 1500 index 3 inet 10.231.11.232 netmask... (3 Replies)
Discussion started by: busyboy
3 Replies

3. IP Networking

solaris routing between two networks

ce0: flags=1100843<UP,BROADCAST,RUNNING,MULTICAST,ROUTER,IPv4> mtu 1500 index 2 inet 10.162.212.132 netmask ffffff00 broadcast 10.162.212.255 ether 0:14:4f:55:82:9 ce1: flags=1100843<UP,BROADCAST,RUNNING,MULTICAST,ROUTER,IPv4> mtu 1500 index 3 inet 10.231.11.232 netmask... (1 Reply)
Discussion started by: busyboy
1 Replies

4. Solaris

ip routing in solaris

I have a SUN ultra 5 machine running Solaris 6. it has two ethernet interfaces qe0 192.168.0.111 and qe1 192.168.1.111 the two subnets are 192.168.0.111 255.255.255.0 192.168.1.111 255.255.255.0 the specified routes are add route -net 192.168.0.0 255.255.255.0 192.168.0.111 add route... (3 Replies)
Discussion started by: adol65
3 Replies

5. Solaris

Routing problem of Solaris

Hi, I got problem of Sun Solaris 8 routing. IP of SunA is 10.16.64.164, and have routing table: 10.16.64.0 10.16.64.164 U 1 19236 bge1 10.6.0.0 10.6.0.3 U 1 16534 bge0 224.0.0.0 10.6.0.3 U 1 ... (4 Replies)
Discussion started by: milo
4 Replies

6. IP Networking

solaris 8 networking routing

hi all, how do i make sure that the new routing should take effect on the os memory right away? what should i run after i add all the new "route add ....". I have an example below: bash-2.03# netstat -rn Routing Table: IPv4 Destination Gateway Flags Ref Use Interface... (1 Reply)
Discussion started by: itik
1 Replies

7. Solaris

Solaris zone routing table

Does anyone know how to delete a an ip off the routing table ? i tried route delete 10.0.0.0 and gateway but it still comes back. i have 2 zones sharing one interface. help... (4 Replies)
Discussion started by: 23patrick
4 Replies

8. Solaris

Basic Solaris static routing

Hi All, :(:confused: I have scoured the web and can't find this anywhere. What I am trying to do is set up a static route through an available gateway on a server, and then test connectivity through TELNET. I have done this on AIX (I already had this procedure). On AIX I would use SMIT. ... (7 Replies)
Discussion started by: jeffpas
7 Replies

9. IP Networking

routing problem in solaris

hi, below is the situation: my server is in 128.10.200.xxx network. i have a solaris computer running in a subnet. this computer got 2 nic card. 1 is 172.18.198.xxx , another 1 is 10.100.xxx.11 there is a machine stick with this computer, ip is 10.100.xxx.12 so, 10.100.xxx.11 and... (3 Replies)
Discussion started by: kyhah
3 Replies
Login or Register to Ask a Question