Regarding Telnet is responsing slow

[Jai Gopal]

Hi Admins

Hi Admins

I have rp2470 server running with 11i which is working as test webserver.Now the problem is that system is response very slow to telnet. If I try to telnet localhost i.e telnet 0.It take 30 min to give me login prompt.After that telnet response will become normal.This problem happen if I take reboot on box.After every reboot first time telnet will take 30 min after it will become normal .The same will happen to user also who is accessing the box. If I see the netstat output it will show me the connection is established of every connection which get invoked.There is no error log in syslog file.Entry in inetd.conf and services seem to be ok.There is no inetd.sec & resolve.conf file in system.Even i tried to change the Speed of Lan card from Auto Neg. to fixed 100FD but still be problem presist.I am using the core I/O lan card of box.Can you guide me what else can i look to resolve this problem.

Thanx in Advance.

Regards,

Gupta

[Shaikh Imran]

Hi,
Have you tried disabling the autonegotiation from the server and manually setting the Duplex and 100 MB/s speed.
Also please try to look at the switch side
settings if possible try to change the port from the switch.
If its in Vlan try to shift it to different vlan and check.

Regards

[Jai Gopal]

Hi,

I have changed the speed of Switch also to 100 FD from Autoneg.but the problem still persist.i do not have vlan on switch.

Regards

Gupta

[Naveej.K.A]

HI GUPTA,

HOW ABOUT CHECKING YOUR /ETC/HOSTS FILE...

127.0.0.1 localhost loopback
x.x.x.x your_host_name

with best wishes
Naveej

[Hoefnix]

You are not using a resolve.conf?
Configure your /etc/nsswitch.conf to use only local files.

HTH,

Peter

[Michael Tully]

I've seen this occur where there is a token ring connection also to the box. The token-ring will attempt to take control of the route.

Have you got your routing tables set correctly?
What happens during a traceroute. Does it find the source quickly across your network?
Does this happen, telnet from many systems or just one LAN?

[Jai Gopal]

Peter/Tully,

The nsswitch.conf file use local file only.

The System is directly connected to Switch only.Issue as i mentioned above in addition to that is the every time when i reboot my box,the system will respond to ping request immdly but none of the service envoked by inetd will not work like telnet,ftp etc.As i mentioned that telnet 0 itself will take almost 30min first time.After that telnet or ftp will working normal until i will reboot the box.

The server is acessed in Lan only and no token ring is there.

Rgds

Gupta

[Bharat Katkar]

Gupta,
Now that looking at the previous discussions it time to check not the network load but your server. It looks like heavily loaded.

Try analying the output of
# w
And observer the load average figures.

And then act accordingly.
Hope that helps.
Regards,

[SAHA]

I have similar problem long back and solved by replacing faulty ethernet card.
Will this box multihomed ?

thanks,

[Abdul Rahiman]

Also do a tracerroute bothways(from the server and remote machine) to make sure that the network connectivity is established via the best route as well as check the routing table. Is this a multihomed server and connected to internet? Then you should check the routing to the machine.

[Abdul Rahiman]

Has this issue resolved?
It sounds like the inetd daemon itself has some problem during startup. Have you checked teh timestamp on `ps -ef | grep inetd` to see if inetd daemon was started up as soon as the machine was rebooted or is it also takig 30 minutes to get started ?
If yes, you might be having some misconfiguration related to DNS entries. Try doing a nslookup to the clientname.

[Mohanasundaram_1]

Hi Gupta,

Looks like you have problem in name resolution. Have you configured NIS? check name resolution through nslookup. If this comes quickly then you focus in other area. If it takes time, then resolve that problem.

Probably moving the nsswitch.conf out of the way and checking is a good Idea. If name resolution is ok, check the /var/adm/nettl.LOG* files. You can format them through
#netfmt -Nlf /var/adm/nettl.LOG00 > /tmp/file

will give you the text output in /tmp/file. check here for any network problems.

Another good Idea is to ensure LAN cumulative patches are up-to-date.

Cheers,
Mohan.

[Brian Hackley · Accepted solution]

Gupta:

I wonder if its possible you have the corrupt umtp problem? Get telnetd patch PHNE_24829 or better.
Special Installation Instructions:
PHNE_24829 contains a fix for the telnetd code defect
described in SR: 8606220839 (JAGad89975) - telnetd writes
to the wrong entry in /etc/utmpx on logout.

Although the SR: 8606220839 (JAGad89975) fix will prevent
any further corruption of /etc/utmpx(4), installing
PHNE_24829 will not correct any existing corruption in the
/etc/utmp(4) or /etc/utmpx(4) files.

Therefore if you are installing PHNE_24829 to fix the SR:
8606220839 (JAGad89975) defect, to completely resolve the
problem you must also ensure that the /etc/utmp and
/etc/utmpx files are cleared of any previous corruption
caused by this defect.

The /etc/utmp and /etc/utmpx files may be cleared using the
following procedure:

Before installing PHNE_24829 insert two lines into the
/etc/inittab(4) file as follows, then save /etc/inittab and
continue the PHNE_24829 patch installation.

init:3:initdefault:
utm1::sysinit:> /etc/utmp # clear current logon \
accounting files
utm2::sysinit:> /etc/utmpx # clear current login \
accounting files

After PHNE_24829 is installed and the system rebooted, you
may delete the above two entries from /etc/inittab or retain
them. In the latter case, /etc/utmp and /etc/utmpx will be
cleared every time the system is rebooted.


Also: do you get same problem with rlogin/remsh/ftp/ssh ?

Last resort:
Turn on inetd connection logging if not already doing with inetd -l
tusc -flvp -T "" -ccc -o tusc-output pid-of-inetd

Telnet to box, look in syslog.log for the PID of telnetd that came in
Examine tusc output for clues.

Hope that helps,
-> Brian Hackley