[j-nsp] Load balancing errors on 15.1R4

[Dragan Jovicic]

That does look bad.
Please keep us updated on you case progress if possible.

Thank you.

Best,
Dragan

On Fri, Oct 21, 2016 at 7:32 PM, Luis Balbinot <luis at luisbalbinot.com>
wrote:

> Case is open, but nothing yet. Could be related to PR1164101, but that's
> not Juniper's official position.
>
> Next week we are upgrading a lab MX480 to 16.2 and see if it persists.
>
> Luis
>
> On Oct 21, 2016 14:54, "Dragan Jovicic" <draganj84 at gmail.com> wrote:
>
>> Hi,
>>
>> Do you have any update on this? Have you opened a case for this maybe?
>>
>> Best
>> Dragan
>>
>> On Tue, Oct 18, 2016 at 4:14 PM, Luis Balbinot <luis at luisbalbinot.com>
>> wrote:
>>
>>> Hey.
>>>
>>> Is anyone else having issues with load-balancing on 15.1R4? I'm
>>> getting these FPC errors in multiple boxes:
>>>
>>> fpc0 LUCHIP(3) RMC 2  Uninitialized EDMEM[0x3ce333] Read
>>> (0x6db6db6d6db6db6d)
>>> fpc0 LUCHIP(3) PPE_2 Errors sync xtxn error
>>> fpc0 LUCHIP(3) PPE_15 Errors sync xtxn error
>>> fpc0 PPE Sync XTXN Err Trap:  Count 23064982, PC 376,     0x0376:
>>> call_table_launch_nh
>>>
>>> They are all MX960s with MPC 16x10G cards. This impacts on traffic to
>>> random destinations (route is OK in FIB and RIB but traffic is
>>> blackholed). If I disable per-packet lb everything works perfectly.
>>>
>>> This issue appears when we have flaps that force route
>>> recalculation/installation.
>>>
>>> I have a similar box running 12.3R3.4 that is just fine.
>>>
>>> Luis
>>> _______________________________________________
>>> juniper-nsp mailing list juniper-nsp at puck.nether.net
>>> https://puck.nether.net/mailman/listinfo/juniper-nsp
>>>
>>
>>