Complete.Org: Mailing Lists: Archives: freeciv-dev: June 2005:
[Freeciv-Dev] Re: (PR#13245) Catalyst 6500 crashes after several users j
Home

[Freeciv-Dev] Re: (PR#13245) Catalyst 6500 crashes after several users j

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
Subject: [Freeciv-Dev] Re: (PR#13245) Catalyst 6500 crashes after several users join to play in network mode
From: "Tamara, Enrique" <enrique.tamara@xxxxxx>
Date: Mon, 13 Jun 2005 12:05:37 -0700
Reply-to: bugs@xxxxxxxxxxx

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=13245 >

Hi, Per,

Thanks for your reply. Cisco has told me that excessive multicast traffic on a 
reserved multicast address is directly forwarded to the switch's CPU, and can 
eventually overwhelm it and make the whole swicth crash.

I suppose I'm not the first person in the world experiencing this problem. 
However, I've been researching on this issue since I came across the first 
crashes and I haven't found much information. Have you ever been told of a 
problem like this?

Regarding Freeciv,
1- Is there any way to configure it in unicast mode? Is it always in multicast 
mode by default?
2- Have you ever heard of a problem like this?


Thanks again for your interest.

Kind regards,

Enrique



-----Original Message-----
From: Per I. Mathisen [mailto:per@xxxxxxxxxxx] 
Sent: sábado, 11 de junio de 2005 20:39
To: Tamara, Enrique
Subject: Re: [Freeciv-Dev] (PR#13245) Catalyst 6500 crashes after several users 
join to play in network mode 


<URL: http://bugs.freeciv.org/Ticket/Display.html?id=13245 >

On Thu, 9 Jun 2005, Tamara, Enrique wrote:
> I'd like to report a problem we have detected when playing Freeciv in 
> network mode (several clients playing with a server). The switch which 
> connects all of the users is a Cisco Catalyst 6500 and it crashes 
> everytime we play the game. We also reported this problem to Cisco 
> TAC, but they state this is a normal problem due to the fact that the 
> game uses a reserved multicast address (0100.5e00.0001) and traffic is 
> directly forwarded to the Catalyst CPU. They say that excessive 
> multicast traffic to reserved multicast addresses can easily make the 
> switch crash.

I found some info here:
http://www.cisco.com/warp/public/105/mcastguide0.html

It seems the use of 225.0.0.1 as default multicast group is bad, and we should 
instead of 225.1.1.1.

  - Per







[Prev in Thread] Current Thread [Next in Thread]