[j-nsp] ex4200 log message question

[Cord MacLeod]

On May 21, 2009, at 11:36 AM, Kevin Oberman wrote:

> It makes the thread very hard to follow.
>> Why not?
>>> Please don't top-post!
>
>> From: Cord MacLeod <cordmacleod at gmail.com>
>> Date: Thu, 21 May 2009 11:21:46 -0700
>>
>> Umm, being that this post was about the ex4200... the platform was  
>> the
>> ex4200.
>>
>> After I submitted a few PRs on 9.3R2 for the EX platform, the message
>> was these would be fixed in 9.3R3 and would be exceptionally stable.
>>
>> On May 21, 2009, at 11:19 AM, Kevin Oberman wrote:
>>
>>>> From: Cord MacLeod <cordmacleod at gmail.com>
>>>> Date: Thu, 21 May 2009 10:37:30 -0700
>>>> Sender: juniper-nsp-bounces at puck.nether.net
>>>>
>>>> Can you elaborate on 'the whole routing seems to fail'?  I was
>>>> actually told by JTAC to upgrade to 9.3R3 as soon as it came out as
>>>> it
>>>> would be exceptionally stable code.
>>>>
>>>>
>>> Umm. On which platform?
>>>
>>> On non-EX platforms, 9.3 looks pretty good, although there were
>>> show-stoppers for us in 93r2, r3 seems to have fixed them.
>>>
>>> This message was about the EX software and it is a VERY different
>>> beast.
>
> We are running 9.3r2 on EX systems. The most serious issue (to us)  
> won't
> be fixed in r3. :-( It forces us to add static ARP table entries in  
> the
> up-stream routers. Without those, the ARP responses stop after a while
> and things just stop. At least the work-around is not too bad.

Indeed that would be a major issue.  Strange I've never encountered  
it.  I'm running an ex4200-48T virtual chassis for my top of rack  
switches, ex4200-24-T virtual chassis for the agg layer and M7's up  
top.  They've been running stable for around 5 months now.  Although  
I'm only using the ex4200-24T in L3 mode.