telnet error: Could not open a connection to host on port 23

[Gene Laoyan]

We have a problem telnetting into our test box. We just installed HP-UX 11.23. Telnet was not on initially so we ran the following commands...

insf -ed psuedo
insf -e

Now we can connect but then a few seconds later we get this error: "Connection to host lost."

W try and telnet again and we get this error:
Could not open a connection to host on port 23 : Connect failed

What could be wrong?
The machine has a rock solid ping.

Thanks

[Court Campbell]

I don't quite see how insf turned on telnet. Last time I checked it was controlled by inetd. There should be a line in your /etc/inetd.conf file that looks like so:

telnet stream tcp nowait root /usr/lbin/telnetd telnetd

if it exists and is uncommented it should hopefully be working. if it is commented you could uncomment it and then run "inetd -c". But I would actually dump telnet in favor of secure shell.

[Gene Laoyan]

I had a problem with telnet previously on a virtual HP-UV machine and it was sugeted in this forum to run those commands and then bounce the VM. It worked.
I'll check to see if the line you identified is commented out or not......

[Steven Schweda]

It Telnet were not configured, I'd expect a
"connection refused" complaint instead of
this stuff. I'd guess that there's a more
generic network problem of some kind.

[Gene Laoyan]

Ok, here's what was in the .conf file...
telnet stream tcp6 nowait root /usr/lbin/telnetd telnetd

I changed it to the following...
telnet stream tcp nowait root /usr/lbin/telnetd telnetd

I removed the 6 from the value tcp6 and bounced the VM and it worked. Why did it default to ipv6?

[Gene Laoyan]

Well....
I was premature on my previous reply and it's still doing the same thing. I can't see anything wrong with the network because it's on one switch and there's ony 3 machines on the switch and the pings are consistant at time=<10ms.

I'm stumped.

[Court Campbell]

what is your syslog output?

[Court Campbell]

> The machine has a rock solid ping.

That's what she said.

I couldn't resist.

[Steven Schweda]

> I'm stumped.

And I don't know much, but there could be any
number of general system resource problems
which might disable a Telnet process. Is
there anything interesting in any of the
general-purpose system logs? Is there any
memory available? Does anything work well
on the console, where the network is not
involved?

[Steven Schweda]

> [...] pings are consistant at time=<10ms.

I don't claim that it's significant, but on
my home junk collection, I normally see ping
times under 1ms:

alp $ ping urtx
PING urtx.antinode.org (10.0.0.8): 56 data bytes
64 bytes from 10.0.0.8: icmp_seq=0 ttl=64 time=1 ms
64 bytes from 10.0.0.8: icmp_seq=1 ttl=64 time=0 ms
64 bytes from 10.0.0.8: icmp_seq=2 ttl=64 time=0 ms
64 bytes from 10.0.0.8: icmp_seq=3 ttl=64 time=0 ms


----urtx.antinode.org PING Statistics----
4 packets transmitted, 4 packets received, 0% packet loss
round-trip (ms) min/avg/max = 0/0/1 ms

That's just 100MHz, full-duplex, with a
dirt-cheap stupid switch in between.

[Gene Laoyan]

The ping times are from a Windoze box.
I'm going to reinstall the OS on the Virtual Machine and try it again.

[Gene Laoyan]

Steven Schweda,

Where can I find the logs for any of the General purpose systems? There is (or at least should be) enough memory on the box, there's 8GB of RAM and the only thing installed is the MCOE. When we connect to the console everything appears to be fine. At least we dont notice anything that is not working other than we can't telnet to it. What general purpose system apps would conflict with telnetd?

[Duncan Edmonstone · Accepted solution]

Gene,

1. Can you confirm this is a physical host and not a VM?

2. Have you checked the network stats using lanadmin? Are you seeing any errors being logged?

3. Is the networking subsystem logging any errors?

netfmt -t 50 -f /var/adm/nettl.LOG00

should show the last 50 errors.

HTH

Duncan

[Gene Laoyan]

I'm sorry, i'm doing three things at once. It looks like I said it was a VM but it is a physical rx2620. Sorry bout that.
After watching the server boot I noticed an error:
"Start Native Adapter Agent..........FAIL"

Any ideas on what the "Native Adpter" is?
Maybe the NIC? I'll switch to the other NIC and try that.

Thanks

[Gene Laoyan]

OK, figured it out with the help of Duncan.
My co-worker stole the IP address of our Integrity server. !@#$@@#$
Looking at nettl.LOG000 gave us the hints we needed.
Now all is good. Thanks everyone!