DNS

[Airmid]

Hi,

I'm running Tru64 5.1B and every two weeks it seems the system looses all of its DNS information. This what I get when I issue the dig and host commands:

# dig anubis.xxx.xxx
dig: isc_socket_bind: address in use
# host anubis.xxx.xxx
host: isc_socket_bind: address in use
#

Additionally, netstat -r hangs, but I can issue netstat -rn, arp -a is very, very slow in response, but it does eventually respond.

The temp workaround is a reboot. Any ideas you guys can send my way??

nslookup cannot resolve the name servers, but I can ping them.

# nslookup gemini.alz.org
*** Can't find server name for address 10.1.0.221: Timed out
*** Can't find server name for address 10.1.0.220: Timed out
*** Default servers are not available
# ping 10.1.0.221
PING 10.1.0.221 (10.1.0.221): 56 data bytes
64 bytes from 10.1.0.221: icmp_seq=0 ttl=128 time=2 ms
64 bytes from 10.1.0.221: icmp_seq=1 ttl=128 time=0 ms
64 bytes from 10.1.0.221: icmp_seq=2 ttl=128 time=1 ms

----10.1.0.221 PING Statistics----
3 packets transmitted, 3 packets received, 0% packet loss
round-trip (ms) min/avg/max = 0/1/2 ms

# ping 10.1.0.220
PING 10.1.0.220 (10.1.0.220): 56 data bytes
64 bytes from 10.1.0.220: icmp_seq=0 ttl=128 time=1 ms
64 bytes from 10.1.0.220: icmp_seq=1 ttl=128 time=0 ms

----10.1.0.220 PING Statistics----
2 packets transmitted, 2 packets received, 0% packet loss
round-trip (ms) min/avg/max = 0/0/1 ms

Thanks!
Airmid

[Kapil Jha]

Hello Airmid,
What is the patch level of this system.
Your whole system is running slow or just these two commands.
Can you please check lsof|wc to see how many open files are there.
Because what I feel is ur connection are open
which are accumulation and creating problem.
You may try to restart processes which has open connections.
hope this help,
BR,
Kapil

[Airmid]

Actually the performance of the system is fine, it just when we try telnet and ftp connections it takes about 90 seconds. These symptoms are very consistent and like I stated once we reboot everything is fine for two weeks and then it acts up again.

But I will look into your suggestions.

Thanks!
A.

[Airmid]

Hi,

Well, I did the lsof and found a ton of files owned by the gshmmod daemon:

gshmmod 851 root 4093u IPv4 0x522c68c0 0t0 UDP *:*

I get to this point and I never get back my prompt just kind of hangs. I know this is the GS series view and discovery used with insight manager. Could this be the issue. I'm going to search to see if there is a bug or something.

Thanks!! Any suggestions would be great.

Oh, yeah whenever this happens sendmail hoses up and starting and stopping doesn't work.

A.

[Airmid]

Hi,

I believe I found a resolution to this issue. When looking for issue with gshmmod, I found this:

QAR 95761- Fixed the GS Platform Discovery HMMO (gshmmod) consuming many UDP sockets resulting in the failure of BIND client and nslookup.

This is exactly what I'm experiencing!

Thanks for pointing me in the right direction, must appreciation!

Airmid

[Kapil Jha]

Yes u got ur problem.
Glad it helped. ;)
BR,
Kapil

[Airmid]

An issue with the gshmmod daemon, fix in version 1.5.