HP server lost connection to a 6509 switch issue

[baselzind]

a customer had a hp server lose network connection to a 6509 switch , he tried to remove and return the network cable with no avail until he restarted the server that when the network connection came back , the problem the customer opened a ticket for the issue but the server support without checking the server are saying the switch is to blame , i believe it is very unlikely that the switch is the problem but i need to provide a response . im thinking should i provide the "show interface" to check  the interface and "show log" to prove nothing is wrong with switch side?

correction: the server was restarted to fix the problem not the switch

[Andrew Khalil]

Dear @baselzind

At least you need to know what was going on at this time of network failure, 

you can use the command #show logging | include xxxxxxxxxx

while xxxxxxxxx is the time when the problem occurred.

You can also ask if what about the other devices connected to the same switch, were they running normally or not? 

You can also, check the CPU, memory, fan status etc. to prove how healthy is the switch! 

Please don't forget to rate any helpful responses and to mark the solution if your issue will be solved! 

Bst Rgds,

Andrew Khalil 

[Bobby Stojceski]

There's a number of unknown variables here to be able to offer a lot of help. Getting the logs and interface stats are good starts, but maybe find out what the server does.

• Is it a virtualization server like ESX, HyperV, OVM etc?
• Does the server have NIC teaming setup?
• Is the switch port in an error disabled state?  (sh int status)
• What's the config of the port? Is it appropriate for the server?
• When you say 'lost network connection', do you mean the link light completely goes out or does the light stay on but server has no network connectivity?
• I assume a shut/no shut on the switch port does not resolve the issue when it occurs? Note that this can have a different effect to removing a network cable and reinstalling it. Such as spanning tree having shutdown the port for example.
• Has it happened only the once or several times?
• What type of port is it?  (SFP, TenGig, GigEth etc)

Without any of the above info it will be difficult to determine what may be happening. HP support won't help you with switch config review and if the only thing they have heard is that rebooting the switch fixes it, they will always blame the switch.

[baselzind]

i mixed up , the server restart not the switch that fixed the issue

[Leo Laohoo]

Well, if the switch has been rebooted then there's nothing one can do about it because all the "evidences" have been erased.
But judging from the description where a reboot fixed the link, I'd say the link went into error-disable.

[balaji.bandi]

@Leo Laohoo agree with leo here, all evidence is gone, we can not re-produce. until switch have some SYSLog server configured, then go back and verify the logs what happend and post the logs here,so can can suggest something here.

Most of Server environment have Dual Ethernet connection to Switch, also worth look arp time out, some time Server team always impatient and force you to reboot the switch to resolve.

this could have been resolved by rebooting the server also. but in this switch was rebooted.

this always challenge 2 deparment handle different environment and demarcations.

[baselzind]

sorry the server was rebooted to fix the issue not the switch

[Leo Laohoo]

Do you know what port the server was plugged into? If you do, post the complete output to the command "sh interface <PORT> controll".
So the server was reloaded? Why? Because the server wasn't reachable by remote access? How about ping?

[baselzind]

i got some of the config as you said:-

interface GigabitEthernetX/X
switchport
switchport access vlan X
switchport mode access
switchport nonegotiate

GigabitEthernetX/XX is up, line protocol is up (connected)
Hardware is C6k 1000Mb 802.3, address is 0024.9709.28e9 (bia 0024.9709.28e9)
MTU 1500 bytes, BW 1000000 Kbit, DLY 10 usec,
reliability 255/255, txload 1/255, rxload 1/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
Full-duplex, 1000Mb/s, media type is 10/100/1000BaseT
input flow-control is off, output flow-control is on
Clock mode is auto
ARP type: ARPA, ARP Timeout 04:00:00
Last input never, output 00:00:21, output hang never
Last clearing of "show interface" counters never
Input queue: 0/2000/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/40 (size/max)
5 minute input rate 2000 bits/sec, 2 packets/sec
5 minute output rate 3000 bits/sec, 3 packets/sec
47157217 packets input, 5761254091 bytes, 0 no buffer
Received 5169 broadcasts (0 multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 0 multicast, 0 pause input
0 input packets with dribble condition detected
53269290 packets output, 6531507863 bytes, 0 underruns
0 output errors, 0 collisions, 3 interface resets
0 babbles, 0 late collision, 0 deferred
0 lost carrier, 0 no carrier, 0 PAUSE output
0 output buffer failures, 0 output buffers swapped out

[paul driver]

Hello

Can you post the output of the following:
Show interface GigabitEthernetX/X status

also
conf t
interface GigabitEthernetX/X
spanning-tree portfast
exit

[Leo Laohoo]

---

@baselzind wrote:

Last input never, output 00:00:21, output hang never

---

It's not a network issue.  The server stopped talking.

[baselzind]

thank you alot for the input , but does that mean the server is still not sending any data into the switch ? im confused?

[Leo Laohoo]

---

@baselzind wrote:
but does that mean the server is still not sending any data into the switch

---

Correct.  

[Andrew Khalil]

@baselzind

@Leo Laohoo

Maybe check the "netstat -a" cli command for server! 

[baselzind]

but this doesn't make any sense? because that mean the server isnt working which the IT team would know because it is a oracle server