Comware

Hello,
We have an interesting problem with 4108gl, probably just something I overlooked in setup.
We can't seem to ping to our remote Wireless WAN networks (Connected via Extreme switch doing ARP PROXY)(10.10.113.x/10.10.116.x) from the 4108GL and beyond in the network. When we remove the 4108gl we can ping from say the 4000m to the remote WAN. However, when the 4108GL is re-introduced, we cannot ping the remote network. I was able to get it to work momentairly when I did a tracert from my desktop, it would ping for a few minutes and then drop off.
Ideas?

The 4108 is a routing capable switch. Is IP routing turned on? Could someone else have its IP address? Is it perhaps responding to (bad) ICMP redirects? Does it have multiple routes to the destination? Can you ping its management interface when this happens? If you are using its routing capabilities, are you pinging from an HPUX 11.0 or newer or something else which has dead gateway detection (3 minute timeout if the gateway does not respond to ping in three minutes it stops using it.) Could you have the mask set wrong?
Is STP turned on? Could you have tagging turned on on the port to the Extreme or the port in a different vlan?

Ron

Just thought of another off the wall possibility that actually happened on out network. We had a device with 2 NICs that was sending a lot of data to another host on the other side of a router. The host also had two NICs. Traffic would flow from device A Nic 1 to device B Nic 1 but return traffic was sent from device B Nic 2 to device A Nic 2. device A would ARP for the MAC of the router when it first started but since there were no packets coming in from the router after a while the ARP would timeout and it would forget where to send them. When we changed the routing tables back to using the same path for both directions the problem went away so check your arp table or fdb to make sure the arp is not getting lost.

Ron

The 4108 is a routing capable switch. Is IP routing turned on? Could someone else have its IP address? Is it perhaps responding to (bad) ICMP redirects? Does it have multiple routes to the destination? Can you ping its management interface when this happens? If you are using its routing capabilities, are you pinging from an HPUX 11.0 or newer or something else which has dead gateway detection (3 minute timeout if the gateway does not respond to ping in three minutes it stops using it.) Could you have the mask set wrong?
Is STP turned on? Could you have tagging turned on on the port to the Extreme or the port in a different vlan?

Ron

Hi Ron,
I will post my config...let me know if you see anything. Also, I believe we have a different VLAN on the extreme switch on certain ports (for the wireless WAN port 24) A vendor setup the Extreme switch, and I need to get the config from that.

Basically here is our config

CAMPUS(A)<---FIBER------>Campus (B on 4108GL)<------------>(Campus (C&D)Wireless on Extreme switch (proxy arp for all campuses)

hostname "Main, 4108GL"
snmp-server contact "Joe Schmo"
snmp-server location "Main-RM7"
time timezone 5
time daylight-time-rule Continental-US-and-Canada
cdp run
module 1 type J4864A
module 2 type J4862B
module 3 type J4862B
module 4 type J4862B
module 5 type J4862B
module 6 type J4862B
module 7 type J4862B
module 8 type J4862B
ip default-gateway 10.10.104.1
ip routing
ip arp-age 100
no timesync
snmp-server community "public" Unrestricted
snmp-server host 10.10.104.248 "public" All
vlan 1
name "DEFAULT_VLAN"
untagged A1-A3,B1-B24,C1-C24,D1-D24,E1-E24,F1-F24,G1-G24,H1-H24
ip address 10.10.104.60 255.255.0.0
ip igmp
exit
fault-finder bad-driver sensitivity high
fault-finder bad-transceiver sensitivity high
fault-finder bad-cable sensitivity high
fault-finder too-long-cable sensitivity high
fault-finder over-bandwidth sensitivity high
fault-finder broadcast-storm sensitivity high
fault-finder loss-of-link sensitivity high
stack commander "XXXX"
password

Two ideas:

You have:
fault-finder bad-driver sensitivity high
fault-finder bad-transceiver sensitivity high
fault-finder bad-cable sensitivity high
fault-finder too-long-cable sensitivity high
fault-finder over-bandwidth sensitivity high
fault-finder broadcast-storm sensitivity high
fault-finder loss-of-link sensitivity high


I would expect these to leave a note in the log if something triggered them but you never know. I'm also not sure what they do if they find a fault. Shutdown the port? You might try looking in the log to see if you see anything. Also try changing them to LOW or NONE and see if that makes a difference.

Look at port status (sh interface) and see if you see a problem - could it be as simple as having the wrong cable? Maybe a crossover cable instead of a straight or vice versa? I know the 2600s do auto mxi/mxix but I'm not sure of the 4108 and even if it does if you turn off Auto it usually stops doing it. Do you see link lights and is the port status good on all links?

You might also try turning off IP Routing if you are not using it.

no ip routing

I see from the config file that you have everything in the same default vlan and untagged so that should not be a problem. I don't see anything about STP so I can't tell if it's on or not. Can you ping the IP address of the 4108 from each campus?

Ron

Ron,
Thanks for the suggestions. I am currently at a remote site and cannot ping the 4108GL or anything that is connected or beyond it. I do not have STP on because someone had mentioned it maybe an issue. I will turn off IP Routing after hours and also the other suggestions you had. Will let you know later today.

FYI- This switch replaced a 4000M and I was hoping it was almost plug and play..guess I was off a bit.

Thanks for help,
Matt

Okay. Update on this. Winter Break at School so I have had time to address this. Still having the connection issue. However, more info.
I updated the 4108gl to the latest firmware (7.63) and then reset to factory defaults. One thing that I noticed From the CLI or any desktop on switch: If I traceroute to the remote network Gateway(remote=10.10.113.1 via 10.10.104.254 Extreme switch running proxy arp) it gets through. I can then Ping the gateway no problem. If I try to ping any other system on the 10.10.113.x network, no joy. The same thing happens on our other remote network (10.10.116.x). Is this some sort of ARP issue?

C:\>ping 10.10.113.1

Pinging 10.10.113.1 with 32 bytes of data:

Request timed out.
Request timed out.
Request timed out.
Request timed out.

Ping statistics for 10.10.113.1:
Packets: Sent = 4, Received = 0, Lost = 4 (100% loss

C:\>tracert 10.10.113.1

Tracing route to 10.10.113.1 over a maximum of 30 hops

1 28 ms 1 ms 1 ms 10.10.104.254
2 11 ms 8 ms 24 ms 10.10.113.1

Trace complete.

C:\>ping 10.10.113.1

Pinging 10.10.113.1 with 32 bytes of data:

Reply from 10.10.113.1: bytes=32 time=10ms TTL=127
Reply from 10.10.113.1: bytes=32 time=8ms TTL=127
Reply from 10.10.113.1: bytes=32 time=8ms TTL=127
Reply from 10.10.113.1: bytes=32 time=7ms TTL=127

Ping statistics for 10.10.113.1:
Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 7ms, Maximum = 10ms, Average = 8ms

C:\>