11-28-2021 09:50 PM - edited 11-28-2021 09:52 PM
Hi
We installed new Cisco C9300 stack (v16.12.4)
and connected ports Te1/1/1-4 and Te2/1/1-4 to 4 HP DL360 servers (each HP server has two NICs)
So Te1/1/1-4 is connected to all 4 HP servers
And the other Te2/1/1-4 is also connected to the same 4 HP servers but the other NICs
The problem is that only Te2/1/1-4 is going UP. the other Te1/1/1-4 is still Down.
All transceivers vendor is Cisco. So they're compatible.
we crossed between the cables
we tried shut/no shut
we rebooted the servers ...
nothing suspicious appeared in the "show logg"
we consulted with HP and they blame cisco (please don't ask me why)
Not sure what we should do in order to bring the other 4 interfaces UP.
Has anyone came across this problem?
11-28-2021 10:38 PM
Please stop using 16.12.X.
11-28-2021 10:42 PM
This is what we get from the box
Which version you recommend for this switch?
11-28-2021 10:47 PM
@AbedB wrote:
Which version you recommend for this switch?
Whatever version but 16.12.X.
11-28-2021 11:51 PM - edited 11-28-2021 11:51 PM
Done
We upgraded to 17.06.02
But still we're facing the same problem
CORE_SW#sh ip int brie | i Te Te1/1/1 unassigned YES unset down down Te1/1/2 unassigned YES unset down down Te1/1/3 unassigned YES unset down down Te1/1/4 unassigned YES unset down down Te2/1/1 unassigned YES unset up up Te2/1/2 unassigned YES unset up up Te2/1/3 unassigned YES unset up up Te2/1/4 unassigned YES unset up up
Switch Ports Model SW Version SW Image Mode ------ ----- ----- ---------- ---------- ---- * 1 41 C9300-24T 17.06.02 CAT9K_IOSXE BUNDLE 2 41 C9300-24T 17.06.02 CAT9K_IOSXE BUNDLE
11-29-2021 12:01 AM
Post the output to the following commands:
11-29-2021 12:09 AM
in the show logg you will see nothing but the UP ports:
*Nov 29 07:46:54.453: %LINK-3-UPDOWN: Interface TenGigabitEthernet2/1/1, changed state to up *Nov 29 07:46:54.461: %LINK-3-UPDOWN: Interface TenGigabitEthernet2/1/2, changed state to up *Nov 29 07:46:54.468: %LINK-3-UPDOWN: Interface TenGigabitEthernet2/1/3, changed state to up *Nov 29 07:46:54.476: %LINK-3-UPDOWN: Interface TenGigabitEthernet2/1/4, changed state to up *Nov 29 07:46:56.455: %LINEPROTO-5-UPDOWN: Line protocol on Interface TenGigabitEthernet2/1/1, changed state to up *Nov 29 07:46:56.463: %LINEPROTO-5-UPDOWN: Line protocol on Interface TenGigabitEthernet2/1/2, changed state to up *Nov 29 07:46:56.470: %LINEPROTO-5-UPDOWN: Line protocol on Interface TenGigabitEthernet2/1/3, changed state to up *Nov 29 07:46:56.477: %LINEPROTO-5-UPDOWN: Line protocol on Interface TenGigabitEthernet2/1/4, changed state to up
Nothing there regarding the Down ports. Lokks like the Cisco does not recognize even the SFP transceivers.
In show inventory port 1-4 is not recognized:
#show inventory NAME: "c93xx Stack", DESCR: "c93xx Stack" PID: C9300-24T , VID: V04 , SN: FOC25012345 NAME: "Switch 1", DESCR: "C9300-24T" PID: C9300-24T , VID: V04 , SN: FOC25012345 NAME: "StackPort1/1", DESCR: "StackPort1/1" PID: STACK-T1-50CM , VID: V01 , SN: MOC25012345 NAME: "StackPort1/2", DESCR: "StackPort1/2" PID: STACK-T1-50CM , VID: V01 , SN: MOC25012345 NAME: "Switch 1 - Power Supply A", DESCR: "Switch 1 - Power Supply A" PID: PWR-C1-350WAC-P , VID: V02 , SN: DCC24512345 NAME: "Switch 1 - Power Supply B", DESCR: "Switch 1 - Power Supply B" PID: PWR-C1-350WAC-P , VID: V02 , SN: DCC24512345 NAME: "Switch 1 FRU Uplink Module 1", DESCR: "8x10G Uplink Module" PID: C9300-NM-8X , VID: V03 , SN: FJZ25012345 NAME: "Te1/1/7", DESCR: "SFP-10GBase-LR" PID: SFP-10G-LR-S , VID: V01 , SN: OPM22412345 NAME: "Te1/1/8", DESCR: "1000BaseLX SFP" PID: , VID: , SN: AGC14012345
Only 7+8 which they're SM transceivers working fine.
11-29-2021 12:18 AM
>All transceivers vendor is Cisco. So they're compatible.... > Cisco does not recognize even the SFP transceivers.
- The SFP's need to be compatible with the specific-9300 model too ,verify with :
M.
11-29-2021 12:22 AM
Hi
They are
All 8 twinax SFPs are same vender (Cisco) , all same: SFP-H10GB-CU3M
Only 4 go up and other 4 go down
Anyway, we escalated the issue again to HP TAC, We suspect problem in there NICs
We're waiting for them to respond
I'll update also here for sure
Thanks
11-29-2021 12:33 AM
Swap those DAC cables around. Move the DAC from switch 2 to switch 1. Move DAC from switch 2 to switch 1.
Observe if the problem "stays" with switch 1 or moves to switch 2.
Next, when doing this, pay attention to the logs. During the removal or insertion, there should be log entries. We want to see that.
11-29-2021 12:41 AM - edited 11-29-2021 12:44 AM
OK
So its funny yeah ...
We went to the DataCenter to physically see the problem
And ... Seems like the system engineers and the vendor engineer is not that much in engineering when they can't tell in HP server the difference between fiber-channel port and regular SFP port (attached photo)
And believe me, I just summarized two essays feelings in two lines ...
Problem solved
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide