Unable to Bring true 64 to network after changeing the IP address

[sunil.ranganath]

Hi All,
I changed the IP address of V5 based server and no matter what i do i am not able to get the server on network.
It is very critical.
There might be some step that might have missed.
Please help.....

[Ivan Ferreira]

Maybe you have a network connectivity problem, maybe because of the autonegotiation value. This will cause the niffd to remove the ip address from the network interface.

What is the output of the ifconfig -a.
What is the output of:
netstat -ni
sleep 20
netstat -ni
hwmgr get attr -cat network

How did you changed the ip address, using sysman?

[Rob Leadbeater]

Hi,

You might also want to check that your default gateway/routing is correct.

You can check this with:

# netstat -rn

Also if you've changed networks, is there now a firewall that might be preventing access ?

Hope this helps,

Regards,

Rob

[Hein van den Heuvel]

>> It is very critical.

Then call support!

>> There might be some step that might have missed.

Yeah, you missed the step where you explained us exactly what you changed, which tools you used.

sysman? vi /etc/rc.config ? ifconfig ? lan_config ?


Maybe you just need an: rcinet restart?
Maybe /etc/host is now out of sync


You also missed the step where you show us what makes you think the network no longer works. What errors to you get?
- ipconfig -a
- ping ...


Hope this helps some,
Hein van den Heuvel (at gmail dot com)
HvdH Performance Consulting

[sunil.ranganath]

I moved the server altogether into a new subnet.

Steps followed:
1. updated the /etc/hosts file.
2. using ifconfig ee0 176.6.63.135 netmask 255.255.255.0 up i chnaged the IP.

3. used route add default 176.6.63.1 1 to update the gateway.

4.Updated the /etc/routes file.

5. Then i powered off the system and moved it to the new location and rebooted it.

I checked the hosts file. routes file and cross checked through Sysman as well.

The output of ifconfig -a also says the interface is up.
The interface pings itself and the loopback.
But packets drop 100% when the gateway is pinged.
The network ports are verified to be working fine and are configured for the same subnet.

The netstat -I ee0 -s output gives the following errors:
block check error
Frame check error
packet receive failure.

I compared the various outputs of netstat with the other true64 server which is working fine on the network.

[Vladimir Fabecic]

So you did not update /etc/rc.config file?
Why you did not use sysman to change IP address and gateway?
OK, let's see what else can be a problem.
Did you check everithing is OK with network switch port?
Did you check your NIC parameter in console:
>>> show ei*_mode (or whatever is name of you NIC).
And please post output of:
# ifconfig -a
and
# netstat -nr

[SJO EGGER · Accepted solution]

change the subnet/ip-adress in /etc/rc.config for ee0/tu0 and in the /etc/hosts file. then reboot the server or make a rcinet restart.

regards

[sunil.ranganath]

As soon as i ran these commands and checked the /etc/rc.config file, i saw that it was updated with latest values.

Output of ifconfig -a:

zblrdv24:root >ifconfig -a
ee0: flags=c63
inet 176.6.63.135 netmask ffffff00 broadcast 176.6.63.255 ipmtu 1500

lo0: flags=100c89
inet 127.0.0.1 netmask ff000000 ipmtu 4096

sl0: flags=10

tun0: flags=80

Output of netstat -nr:
zblrdv23:root >netstat -nr
Routing tables
Destination Gateway Flags Refs Use Interface

Route Tree for Protocol Family 26:

Route Tree for Protocol Family 2:
default 176.6.63.1 UGS 3 13140 ee0
127.0.0.1 127.0.0.1 UHL 7 43492 lo0
176.6.63/24 176.6.63.137 U 4 18639 ee0
176.6.63.137 176.6.63.137 UHL 1 17028 ee0

The only difference i see between the ifconfig -a of working server and this is the added parameter "PROMISC" AS SEEN ABOVE.

[Brendan Murphy_5]

Hi Sunil,
Why not run the command
# sys_check -all > hostname.html
and post the resulting output html file. This will provide all the relevant information for your system configuration. Then we can check all the network configyration files as well as netstat output & last 500 lines of binary.errlog.
Personally I suspect a switch port/NIC incompatability eg : swicth port set to 10Mg while NIC set to 100Mg FastFD, or incorrectly set VLAN.

Regards,

Brendan

[Vladimir Fabecic]

Talk with network guys and ask them about switch port settings and status and VLAN settings.

[sunil.ranganath]

Network ports are fine and are in the same LAN. i checked by connecting a laptop with same ip.

It`s working fine

[Hein van den Heuvel]

>> 5. Then i powered off the system and moved it to the new location and rebooted it.

Up untilly that point I thought (I could be stupid/dense) that you did not physically touch the box. That we did not have to worry about cables and connections.

But with this info the whole picture changes. EVERYTHING is suspect. The cable, the switch, the way you plugged it in.

When you plug it in, do you get a console message (/ver/adm/messages) that the network is up, at a certain (100mb) speed.
Do you get a green light and some activity at the interface?

When you tried that PC (good test btw!), was it using the same cable removed from Tru64 box, plugged into the PC for a while?

Is the switched zoned? autosense?



Good luck!
Hein.

[Brendan Murphy_5]

You should still post the sys_check.

Laptop's tend to autonegotiate NIC speed setting. AlphaServers are frequently set at SRM firmware level >>> prompt. For example, If switch is set to 100Mb Full Duplex & AlphaServer SRM is set to 10Mb Full Duplex, then you won't get anything. You need to halt system, & depending on NIC type set ewa0_mode or eia0_mode to auto. Then boot & see what happens. (you need to be sure that auto is NOT set at both the swicth & SRM) You're best tying down at the switch level.

Regards,

Brendan

[Vladimir Fabecic]

Give some more informations about hardware.
And please post output of:
>>> show ei*_mode
And also check everything that Hein suggested.

[sunil.ranganath]

Hi All,
The problem is solved!!
i deleted all configuration and started of from scratch.
assigning IP,gateway, host file, link speed, auto-negotiation etc.,
even the network part was started from scratch and the culprit was the network cable....
i could not beleive it, coz i removed it from the package myself,it was a damn new cable..

Just want to say one thing, i am basically a solaris guy, this was the first time i entered a HP based forum.
Beleive me this is the most active forum i`ve ever come across. i just posted the problem yesterday at 16:00 hrs and within 30 hrs, i have more than 5 people and over 8 responses.
i really appreciate it.
Thanks a lot to all of you.
will continue to be an active member of this group.

My contact:
Sunil Kumar.K.R
Analyst - UNIX & Clearcase Administrator, IMS
Societe Generale Global Solution Centre,
6th floor, Navigator Building,
ITPL,Bangalore, India.

Office: +91 28095 859
Mob: +91 9886016265
Mail to: sunil.ranganath@sgcib.com

[Hein van den Heuvel]

Welcome to the HP ITRC Forums!

5 people, 4 continents. :-).

Thanks for the kind words.

You may want to 'close' your topic and optionally assign 'points' to help future readers focus on the replies that helped you most and to acknoledge the efforts by the contributors.

http://forums1.itrc.hp.com/service/forums/helptips.do?#28

Cheers,
Hein
[0 points for this!]