[f-nsp] XMR/MLX-firmware: 5.3 and 5.4 broken for "cam-mode ipv4 dynamic" error-msg: ("Warning: IPv4 Route ADD: CAM entry creation FAILED") -- only 5200h works, why don't they fix this bug?

[Aaron Wendel]

Have you escalated this issue?  What has TAC been telling you?

On Apr 27, 2013, at 6:36 PM, "Gunther Stammwitz" <gstammw at gmx.net> wrote:

> Dear all,
>  
> there was no useful feedback out of this community and yes, of course we have rebooted after applying the setting..
> There is a firmware-bug with cam-mode ipv4 dynamic where the following warning appears in ones logs: “Warning: IPv4 Route ADD: CAM entry creation FAILED” always occurs in ipv4 dynamic cam mode.”
>  
> I found out that the XMR-MLX 5.3-tree is completely broken and have tested 5300c, 5300d and the latest 5300e.
> The 5.4tree seems to be broken too, at least 5400b is broken while I could not yet test 5400c.
> 
> à The only stable version seems to be 5200h out of the 5.2-tree, earlier versions are also broken.
>  
> I have seen in the change log of 5200h that several cam-related bugs have been closed with “DEFECT000427885” but I cannot find these fixes in either 5300e or 5400c… which makes sense, because these versions show the problem mentioned above.
>  
> Does anyone have an idea why foundry is not able to fix this bug?!?
> 
> Gunther
> 
> 
> 
>  
>  
> Von: foundry-nsp-bounces at puck.nether.net [mailto:foundry-nsp-bounces at puck.nether.net] Im Auftrag von Gunther Stammwitz
> Gesendet: Samstag, 15. Dezember 2012 01:22
> An: foundry-nsp at puck.nether.net
> Betreff: [f-nsp] Problem with dynamic cam-mode on netiron mlx 5.3.00d: "Warning: IPv4 Route ADD: CAM entry creation FAILED"
>  
> Dear colleagues,
>  
> has someone else seen problems with dynamic cam mode on the netiron mlx?
> I have upgraded the system to 5.3.00d and switched from static to dynamic cam mode at the same time.
>  
> From time to time I can see the following WARNING in the system log: “Warning: IPv4 Route ADD: CAM entry creation FAILED”.
> Does anyone know why this is happening and if the router tries to program the problematic entry into the cam again or is it getting lost?
>  
> Has anyone else seen these problems before – if so: which version are you running?
> Is anyone running dynamic cam-mode without problems? If so: which version is running fine?
> 
> This router is running bgp4 with full tables and two feeds and pushes about 3 gigabit per second on a NI-MLX-1Gx20-SFP-linecard in slot1.
>  
>  
> >show cam-partition brief
> CAM partitioning profile: ipv4-ipv6-2
> Slot 1 XPP20SP 0:
> # of CAM device                = 3
> Total CAM Size                 = 524288 entries (36Mbits)
> IP: Raw Size 491520, User Size 491520(0 reserved)
>   Subpartition  0: Raw Size 2048, User Size 2048, (0 reserved)
>   Subpartition  1: Raw Size 448283, User Size 448283, (0 reserved)
>   Subpartition  2: Raw Size 35001, User Size 35001, (0 reserved)
>   Subpartition  3: Raw Size 4819, User Size 4819, (0 reserved)
>   Subpartition  4: Raw Size 729, User Size 729, (0 reserved)
> […]
>  
>  
> >show cpu lp 1
> SLOT  #:               LP CPU UTILIZATION in  %:
>              in 1 second:  in 5 seconds:  in 60 seconds: in 300 seconds:
>      1:        8             8              8               8
>  
>  
> Kind regards
> Gunther
>  
> _______________________________________________
> foundry-nsp mailing list
> foundry-nsp at puck.nether.net
> http://puck.nether.net/mailman/listinfo/foundry-nsp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/foundry-nsp/attachments/20130427/90445684/attachment.html>