[c-nsp] CEM unusable on ME-3600X-24CX with 15.3.3.S4

[Vitkovský Adam]

It would be sooo great if CISCO would reveal the documentation related to the HW show commands so that we could check whether the HW is programed correctly ourselves.  
That way we could open up a TAC case pinpointing the issue so it could be sent to developers right away and we would not need to go through the obligatory pain of initial introductions to TAC engineers which consumes from one to three hours in a webex session where one can't do any other work. Then we need to wait for the developers to send us the HW show commands. 

The best would be to get full documentation to sdcli as well so we would be able to confirm the bug by fixing it temporarily on our own but of course that would allow us the full control over the chases (e.g. to carve up the TCAM as we which) so that's not going to happen. 
But at least the documentation to the show commands would be greatly appreciated. 
 
adam

> -----Original Message-----
> From: Mark Tinka [mailto:mark.tinka at seacom.mu]
> Sent: Saturday, October 11, 2014 11:19 AM
> To: cisco-nsp at puck.nether.net
> Cc: Vitkovský Adam
> Subject: Re: [c-nsp] CEM unusable on ME-3600X-24CX with 15.3.3.S4
> 
> On Thursday, October 09, 2014 02:29:57 PM Vitkovský Adam
> wrote:
> 
> > It's great platform for IP services with current
> > 15.3.3.S4 code is almost flawless. The only major bug seems to be
> > occasional Rosen mGRE mVPN mcast HW programing related issues.
> 
> I've also been hitting IPv6 ACL programming issues on the ME3600X lately -
> running 15.4(2)S.
> 
> In odd cases, IPv6 ACL's that do not block IPv6 traffic end up doing so, or
> leading to packet loss.
> 
> Interestingly, a reboot does not fix, but rather, removing and re-applying the
> ACL (a number of times) does.
> 
> Mark.