[j-nsp] QFX5100 3rd party optic/DAC

[Darren O'Connor]

Handy commands...

I've got Juniper ones and a non-Juniper one here:

    Xcvr 51      REV 01   740-038623       QSFP+-40G-CU1M
    Xcvr 52      REV 01   740-038623     QSFP+-40G-CU1M
    Xcvr 53               NON-JNPR              UNKNOWN


And I'm seeing some interesting stuff. On the Juniper cable:

TFXPC0(vty)# show qsfp 1 identifier
  Identifier:             QSFP+
  Connector type:         No Separable Connector
  Encoding:               Unspecified
  Vendor name:            Molex Inc.
  Vendor OUI:             00093A
  Vendor PN:              1110409055
  Vendor Rev:             B

...redacted...

  Transceiver code:
    10G/40G Ethernet:     40GBase-CR4




On the Non working one:

FXPC0(vty)# show qsfp 3 iden
  Identifier:             QSFP
  Connector type:         Copper Pigtail
  Encoding:               Unspecified
  Vendor name:            Molex Inc.
  Vendor OUI:             00093A
  Vendor PN:              112-00178
  Vendor Rev:             A0
  
...redacted...

  Transceiver code:
    Gigabit Ethernet:     1000Base-CX


So maybe this cable is only 1000Base?  What makes this even more weird is that the 4300 sees it as the same, but it comes up as 40Gb:

(vty)# show qsfp 1 identifier
  Identifier:             QSFP
  Connector type:         Copper Pigtail
  Encoding:               Unspecified
  Vendor name:            Molex Inc.
  Vendor OUI:             00093A
  Vendor PN:              112-00178
  Vendor Rev:             A0
  
...redacted...

  Transceiver code:
    Gigabit Ethernet:     1000Base-CX

root at 4300-STACK> show virtual-chassis vc-port
fpc0:
--------------------------------------------------------------------------
Interface   Type              Trunk  Status       Speed        Neighbor
or                             ID                 (mbps)       ID  Interface
PIC / Port
1/1         Configured         -1    Up           40000        1   vcp-255/1/0




I would expect this to come up at 1Gb if that was the case. The QFX5100s also refuse to bring up those DACs even if they are the only cables available. 



> Date: Mon, 29 Sep 2014 10:34:27 -0700
> Subject: Re: [j-nsp] QFX5100 3rd party optic/DAC
> From: jackson.tim at gmail.com
> To: darrenoc at outlook.com
> CC: aaron.dewell at gmail.com; juniper-nsp at puck.nether.net
> 
> Some good debugging:
> 
> start shell
> vty fpc0
> show qsfp list
> show qsfp <id>
> 
> 
> 
> 
> On Mon, Sep 29, 2014 at 10:16 AM, Darren O'Connor <darrenoc at outlook.com> wrote:
> > I've got the very latest code on as it's in a lab D25
> >
> > I've got two 3rd party DACs and it's going from QFX5100 to QFX5100. Both have the same issue. Same two cables in QSFP ports on the back of the 4300s works perfectly fine.
> >
> > Date: Mon, 29 Sep 2014 08:25:38 -0600
> > Subject: Re: [j-nsp] QFX5100 3rd party optic/DAC
> > From: aaron.dewell at gmail.com
> > To: darrenoc at outlook.com
> > CC: juniper-nsp at puck.nether.net
> >
> > What version of code? D10 (frs) had some issues with some cables which is resolved in more current versions.
> > Also if this is 5100 to 4300 make sure you have auto negotiation turned off on the 4300 (but that would probably fail with a juniper branded dac as well so unlikely to be the issue).
> > On Sep 29, 2014 6:43 AM, "Darren O'Connor" <darrenoc at outlook.com> wrote:
> > Anyone having any luck with this? I've got a few QSFP DACs that work perfectly fine on a 4300 stack, but the QFX5100 refuses to work with them. Work fine with a Juniper branded DAC.
> >
> >
> >
> >
> >
> >
> >
> > _______________________________________________
> >
> > juniper-nsp mailing list juniper-nsp at puck.nether.net
> >
> > https://puck.nether.net/mailman/listinfo/juniper-nsp
> >
> >
> > _______________________________________________
> > juniper-nsp mailing list juniper-nsp at puck.nether.net
> > https://puck.nether.net/mailman/listinfo/juniper-nsp