HP-UX

OK HPUX masters I need help. I have a HPUX 11.23 server that I am using as a Ignite server that services two HPUX services for backups only. The other day I noticed that our Make Net Recoveries were failing. Upon attempting to login to the server SSH and could not I then tried to ping and could not. I then connected to the MP and switched to the CO which was successful, I check the disk space which was fine, I checked the syslog.log which had no odd messages, but when I opened SAM and entered network config it basically locked up and crawl over 2 hours before opening the net config SAM window. I have to shut down the box to get back into MP > CO. Any ideas would be greatly appreciated.

I am in no way a "HP-UX master", but this sounds suspiciously like a resolution problem. Most times when the system is fine but network response is infinitely bad it is for one of two reasons:

1) The DNS server you are using responds very slowly (or not at all). This might come from a severed link between the two machines (for instance a firewall blocking port 53) or whatever. Have a look at /etc/nsswitch, "files" should go before "bind", meaning name resolution first tries a local /etc/hosts file and only then the DNS server. You might want to put all the "important" systems into /etc/hosts so that you do not have to rely on DNS being available.

2) Another possible scenario would be a "physically" severed link to the network. Possible causes would be a switch port set to half-duplex while the system port is set to full-duplex (or vice versa), different link speeds, Auto-Negotiation gone awry, or similar. Half-duplex vs full-duplex is tricky because "ping" will work (most of the time, anyways) but the line speed will be reduced to near zilch.

I hope this helps.

bakunin

Thanks bakunin for the reply, a good rely as well. I check those things as it did seem to be a resolution issue. I just resolved the issue moments ago, it was Lan0 issue, I shut down that interface and configure Lan1 and all is good. Thanks for the reply.

If the HP-UX box is not pingable, and its NIC access is slow then I suspect the NIC hardware or the LAN cable is broken.
Does a

Code:

/usr/sbin/lanscan

work?
If yes, on all lan devices that are in UP state does an

Code:

/usr/sbin/ifconfig lanX

work?

---------- Post updated at 09:58 AM ---------- Previous update was at 09:56 AM ----------

Just saw your answer. Seems to be a hardware problem indeed.

how do I close or resolve a thread. I dont have time to scour the site now but would later and do not want the thread to remain open??

Take a look at APA (auto port aggregation), so operating system does the failover on another interface without downtime.

Since lan1 is working fine, i would guess that cable needs to be replaced, since it is rarely the entire card.

Regards
Peasant.

I have marked the thread 'solved' for you to discourage new replies. The thread will close itself in a reasonable amount of time. Meanwhile you can still come back if the problem decides it isn't solved after all.