Showing results for 
Search instead for 
Did you mean: 

11.23 networking setup

Go to solution
John Talaga

11.23 networking setup

Hello all,

I recently put a fresh installation of 11.23 on our server network, and it was working fine (aside from a random SSH disconnect issue). After apply patches, no network traffic is leaving the local network on this server. There are other servers on this network that can communicate with separate networks without issue and this server can communicate with other servers on the same network without issue, but no traffic is reaching the gateway.

# netstat -rn lists the appropriate default route to the gateway.


All of these config files have accurate or appropriate data in them, similar to the configuration that is working on other servers on the same network.

One side affect I have noticed is that swinstall takes an extraordinarily long time to process anything because of the lack of network communication. When starting up swinstall, I see it hangs on the "Contacting the target..." message.

Any thoughts on what might be causing the issues with networking?

Any help is greatly appreciated.



Esteemed Contributor

Re: 11.23 networking setup


Does "ping " work ?

"When you look into an abyss, the abyss also looks into you"
John Talaga

Re: 11.23 networking setup

No, the gateway does not respond to pings.

This is true for other servers on the same network, and they work without issue.

Respected Contributor

Re: 11.23 networking setup

Hello John ,
It may be duplex or negotiation error of some sort possibly at your lan switch end .
Can you post the output of
landmin -x n ,
lanadmin -g mibstats n and
landmin -g mibstats_ext n

where n is the ppa number indicated by lanscan .

To get around the the slow SD patch install problem make sure your host name is correct in the /etc/hosts file and change nsswitch.conf to use files first for DNS. SD does DNS name look up every time it calls a new function.
Help is out there always!!!!!
Respected Contributor

Re: 11.23 networking setup

Hello John,
Sorry , typing trouble. The commands should of course be :

lanadmin -x n
lanadmin -g mibstats n
lanadmin -g mibstats_ext n

Help is out there always!!!!!
John Talaga

Re: 11.23 networking setup


Here is the output:

# lanadmin -x 0
Current Config = 100 Full-Duplex AUTONEG

# lanadmin -g mibstats 0

Tue , May 11,2010 12:58:57

PPA Number = 0
Description = lan0 HP PCI 10/100Base-TX Core [100BASE-TX,FD,AUTO,TT=1500]
Type (value) = ethernet-csmacd(6)
MTU Size = 1500
Speed = 100000000
Administration Status (value) = up(1)
Operation Status (value) = up(1)
Last Change = 510345
Inbound Octets = 1516066569
Inbound Unicast Packets = 1130797
Inbound Non-Unicast Packets = 19470
Inbound Discards = 0
Inbound Errors = 0
Inbound Unknown Protocols = 54
Outbound Octets = 29432963
Outbound Unicast Packets = 275762
Outbound Non-Unicast Packets = 31
Outbound Discards = 0
Outbound Errors = 0
Outbound Queue Length = 0
Specific = 655367

Ethernet-like Statistics Group

Index = 1
Alignment Errors = 0
FCS Errors = 0
Single Collision Frames = 0
Multiple Collision Frames = 0
Deferred Transmissions = 0
Late Collisions = 0
Excessive Collisions = 0
Internal MAC Transmit Errors = 0
Carrier Sense Errors = 0
Frames Too Long = 0
Internal MAC Receive Errors = 0

As for the SD problems, the thing that struck me as odd is that I do have files before dns in nsswitch.conf and hosts file is correct, but still the delay when using swinstall. I think that even tho nsswitch.conf had files before dns, it was still trying to use dns first, and had to timeout when trying to reach the dns server.

Thanks much for your help,

John Talaga

Re: 11.23 networking setup

Some additional information.

Running this command:

ndd -get /dev/ip ip_ire_status shows this in the last line regarding the default gateway.


Our Gateway does not respond to ping requests but see the output from the following command:

# ndd -get /dev/ip ip_ire_gw_probe

As I said before, netstat -rn shows the default route set correctly:

Routing tables
Destination Gateway Flags Refs Interface Pmtu UH 0 lo0 4136
2XX.1XX.6X.X 2XX.1XX.6X.X UH 0 lan0 4136
2XX.1XX.6X.0 2XX.1XX.6X.X U 2 lan0 1500 U 0 lo0 4136
default 2XX.1XX.6X.254 UG 0 lan0 1500

# ndd -set /dev/ip ip_ire_gw_probe 0
# route add default 2XX.1XX.6X.254
add net default: gateway 2XX.1XX.6X.254: Network is unreachable

What could be wrong that the system cannot reach the default gateway? Other HP-UX installations on this network have no issue.

Thanks again,
S. Ney
Trusted Contributor

Re: 11.23 networking setup


Not sure if this is your situation but when I had a long swinstall time it was due to the fact that my servers were an ignite clone. Check /var/adm/sw/target_hosts You should see your server hostname and loopback in the file. Also another possible reason for the delay is your hostname not matching the primary DNS entry name. We had to get our DNS team to make a change and switch a DNS alias to primary on their side. (This was on an rx6600 11.23)

Lastly when we had network problems I had to work with my local network team. We traced the cables to the patch panel and then to the cisco switch. I was able to ping the switch IP from my server. It ended up being a misconfiguration on the switch itself and the network team had to apply a patch to the cisco switch.
Hans Ove
Occasional Advisor

Re: 11.23 networking setup


When upgrading from hpux10.20 to 11.(11/23/31) I always put the three following lines into the bottom of this file: /etc/rc.config.d/nddconf



Now the hpux 11.x will act like previous versions of hpux, no need to ping the gateway...
Respected Contributor

Re: 11.23 networking setup

Hello John,
There are no errors on the card, so the flapping duplex problem is not likely to be the cause.
For the record my ip_ire_gw_probe is set to 1 but that is on a gigabit card.
It may be a dumb question but is the subnet mask set correctly, and is the port in the switch configured for this subnet and in the correct vlan if you are using them ?

What version of the cumulative ARPA Transport patch do you have installed ?
Help is out there always!!!!!
Jim Keeble
Trusted Contributor

Re: 11.23 networking setup

The ip_ire_gw_probe setting is your problem. With that set to the default value of 1, the system pings the gateway periodically to see if it has died, and if it doesn't respond it stops sending traffic to it.

Set it to 0 as in the example given by Hans to avoid the problem.

You will have to either reboot or manually remove and re-add the default gateway after changing the setting. You can change it on the fly with :

ndd -set /dev/ip ip_ire_gw_probe 0

... but you need the nddconf setting to make it survive a reboot.

I think the latest Arpa Transport patches may change the behaviour such that when enabled, it reports the dead gateway on the system console but continues to send the traffic. Don't remember the exact details.
John Talaga

Re: 11.23 networking setup

Hello all,

I resolved this issue by installing all missing patches. I'm not sure which patch/patches specifically were responsible for fixing this issue, but network traffic can leave the local network now.

I checked prior to applying these patches and I did have installed the most recent cumulative ARPA transport patch, and all dependencies, so I am not sure it was related to that.

Thanks all for your help and your suggestions.

Steven E. Protter
Exalted Contributor

Re: 11.23 networking setup

Shalom John,

I recommend that at least annually, you update the system with a QPK major patch set. This will help keep your system running healthy.

QPK's are issued bi-annually.

Steven E Protter
Owner of ISN Corporation
John Talaga

Re: 11.23 networking setup

Sound advice. Something I will make sure to do in the future, as I see now that missing patches can cause strange behavior.

Thanks again all,