Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Bizarre Connectivity Issue

Status
Not open for further replies.

mcintoshs

MIS
Oct 27, 2010
2
US
I have a bizarre network connectivity issue.

There is a server on our network with IP address 10.0.1.23 which is in the default VLAN 10.0.0.0/16 on HP 5406zl core switch. All servers on that VLAN can ping that system successfully and reliably with no dropped packets on a lengthy persistent ping.

Systems on another VLAN 10.6.0.0/23 cannot reliably ping that server with IP of 10.0.1.23. If one were to do a persistent ping, you would find ping dropouts lasting 15-30 minutes. However, they can successfully ping every other system in 10.0.0.0/16 reliably without dropping any packets at all.

A tracert during periods of packet loss shows that pings from 10.6.0.0/23 are reaching the VLAN gateway on the core switch but then getting dropped. What could be causing this?
 
Welcome to tek-tips!

That is indeed a strange problem. There are four things that I can think of off hand that could cause that type of problem: 1 - the machine is disconnected or turned off during the periods of down time (assuming that it isn't ping-able from another host at that same time), 2 - deliberate blocking of the pinging host via a firewall, 3 - a networking problem such as an intermittent failure in a switch or a bad cable or connection, 4 - software (or hardware) development activities that have networking or ICMP messaging turned off temporarily.

I would first see if it is simultaneously visible to other machines and if it is, try swapping some of the switch gear and cables to rule out a hardware problem.

Seeing that it is a server, I would also look for any sort of mal-ware that could be intentionally blocking responses as well as check it for a throughput overload condition.


 
Interestingly, I have started having a very similar issue since 24 hours ago on a VRRP pair of Procurve 5400s.
I have K_14_63 on mine, and you?
 
Thanks for the responses. I'm running K_14_60 on the 5406zl's.
Norway2: none of the 4 scenarios explain this because the server is never powered down and can be pinged persistently without drops by any server in its own subnet.
 
I recently updated my firmware from K_14_47 - did you update yours recently?

With mine, there are a number of servers on a special subnet and I can ping some of them from the workstation subnet but not others, even though they are all up.
From the 5400 itself (which does the routing), I can ping them all.
I would normally be inclined to blame the servers, myself, so I might insist they show me the servers' IP config.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top