[c-nsp] X6708-10G-3C compatible with 7604 chassis?

[Chris Welti]

When you insert a card that needs more than 75cfm, the fan will be 
switched from restricted-power to full-power by the IOS software.
I'm starting to believe there are different FAN-MOD-4HS out there, and 
some of them only support 75cfm per slot
(300cfm in total) vs. those that also support 94cfm per slot in full 
power mode (which you need for the X6708-10G-3CXL).
Might be worth trying a newer hardware revision of the FAN-MOD-4HS.

Chris

Am 21.03.14 11:31, schrieb Lobo:
> Wow I never knew about that switch. Ours had the variable and full 
> options and we moved it over to full. The fan is definitely running at 
> the higher speeds but I think there's a problem with the reporting. 
> The router still can't get the proper information from the fan module 
> so it still thinks it's operating at 75 cfm. Reboots don't help it at 
> all and so the 6708 still thinks it doesn't have enough cooling 
> capacity to start it up.
>
> My colleague made an interesting observation and that is that prior to 
> inserting the X6708-10G-3C card the fan was operating correctly and 
> reporting everything like this:
>
> fan-tray 1:
> fan-tray 1 type: FAN-MOD-4HS
> fan-tray 1 mode: Restricted-power
> fan-tray 1 fan-fail: OK
> chassis per slot cooling capacity: 75 cfm
>
> Once the 6708 card is inserted, the fan module almost seems like it 
> loses it's ROM information or something because it then reports like 
> this:
>
> fan-tray 1:
> fan-tray 1 type:
> fan-tray 1 fan-fail: OK
> chassis per slot cooling capacity: 70 cfm
>
> This has happened twice now (once in the field and once in our lab). 
> I'm wondering if the insertion is somehow affecting the fan module?
>
> Jose
>
> On 3/21/2014 10:40 AM, Tim Durack wrote:
>> This might help you:
>>
>> In the Cisco 6504 E, on the FAN-MOD-4HS fan-tray there is a switch at 
>> the rear. This switch should be in the "A" position. See Figure 2.1.1.
>> This allows the fan to run in high-power mode. In this mode it 
>> generates an airflow of 100 ft3/min per slot.
>>
>> When the switch is in the "M" position it operates in the 
>> restricted-power mode and generates only 75 ft3/min per slot. The 
>> WS-X6716-10GE line card has a cooling requirement of 84 ft3/min. In 
>> this case it could prevent the line card from powering up.
>>
>> Note: on another FAN-MOD-4HS fan-tray the positions of the switch are 
>> labeled "VARIABLE" and "FULL". The switch should be put in the 
>> "VARIABLE" position.
>>
>>
>> On Fri, Mar 21, 2014 at 9:49 AM, Lobo <lobotiger at gmail.com 
>> <mailto:lobotiger at gmail.com>> wrote:
>>
>> Agreed. But even after a reboot, the chassis is still not
>> allowing this 6708 card to come up because it complains about
>> insufficient cooling.
>>
>> This is pretty annoying as their documentation shows that the 7604
>> is compatible with this card but now we're starting to wonder if
>> that's truly the case.
>>
>> Jose
>>
>>
>> On 3/21/2014 9:42 AM, Phil Mayers wrote:
>>
>> On 21/03/14 12:15, Lobo wrote:
>>
>> Even though the cards are not completely compatible for
>> totals # of
>> routes (3C vs 3CXL) the system should default to the
>> lowest card but I
>> would think that it would still power up the card and let
>> it function
>> normally.
>>
>>
>> No; the PFC/DFCs must all match. The other PFC/DFC need to be
>> *downgraded* to non-XL mode, and that need a chassis reboot.
>> _______________________________________________
>> cisco-nsp mailing list cisco-nsp at puck.nether.net
>> <mailto:cisco-nsp at puck.nether.net>
>> https://puck.nether.net/mailman/listinfo/cisco-nsp
>> archive at http://puck.nether.net/pipermail/cisco-nsp/
>>
>>
>> _______________________________________________
>> cisco-nsp mailing list cisco-nsp at puck.nether.net
>> <mailto:cisco-nsp at puck.nether.net>
>> https://puck.nether.net/mailman/listinfo/cisco-nsp
>> archive at http://puck.nether.net/pipermail/cisco-nsp/
>>
>>
>>
>>
>> -- 
>> Tim:>
>
> _______________________________________________
> cisco-nsp mailing list cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/