VM TCP/IP Routing Question

2006-08-02 Thread Dennis Schaffer
Hi, Our network folks noticed that one of our zLinux systems is driving a lot of packet retransmissions, caused mainly because packets are being received out-of-sequence, outside the target workstation's response window (I hope I'm getting the terminology correct here). Going one step further,

Re: VM TCP/IP Routing Question

2006-08-02 Thread McKown, John
-Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Dennis Schaffer Sent: Wednesday, August 02, 2006 12:27 PM To: IBMVM@LISTSERV.UARK.EDU Subject: VM TCP/IP Routing Question Hi, Our network folks noticed that one of our zLinux

Re: VM TCP/IP Routing Question

2006-08-02 Thread Miguel Delapaz
Dennis, When more than one path to a destination exists, and they have the same cost (as defined by the routing protocol), z/VM TCP/IP will round robin packets to that destination through the available paths. For static routing, this behavior is controlled by the EQUALCOSTMULTIPATH parameter on

Re: VM TCP/IP Routing Question

2006-08-02 Thread Alan Altmark
On Wednesday, 08/02/2006 at 12:26 EST, Dennis Schaffer [EMAIL PROTECTED] wrote: Our network folks noticed that one of our zLinux systems is driving a lot of packet retransmissions, caused mainly because packets are being received out-of-sequence, outside the target workstation's response

Re: VM TCP/IP Routing Question

2006-08-02 Thread Dennis Schaffer
Alan Altmark wrote on 08/02/2006 01:02:34 PM: You need to get the Linux guests directly on the VSWITCH and get out of the virtual router business. The VSWITCH will be responsible for managing OSA redundancy at a physical level rather than an IP routing level. Alan, Thanks for your response.

Re: VM TCP/IP Routing Question

2006-08-02 Thread Dennis Schaffer
John/Miguel, Thanks for your responses. EqualCostMultiPath seems to be exactly what's causing this behavior. In theory, it seems EqualCostMultiPath would be ideal behavior to load-balance network traffic between the adapters, which seems like a good thing. However, I hadn't considered the

Re: VM TCP/IP Routing Question

2006-08-02 Thread Alan Altmark
On Wednesday, 08/02/2006 at 02:11 EST, Dennis Schaffer [EMAIL PROTECTED] wrote: Please understand that each of my OSA adapters are connected to different IP subnets (because, my network folks say, that provides optimum redundancy: completely different network hardware, from the OSA to the

Re: VM TCP/IP Routing Question

2006-08-02 Thread Dennis Schaffer
Miguel Delapaz wrote on 08/02/2006 01:00:40 PM: Dennis, When more than one path to a destination exists, and they have the same cost (as defined by the routing protocol), z/VM TCP/IP will round robin packets to that destination through the available paths. For static routing, this behavior

Re: VM TCP/IP Routing Question

2006-08-02 Thread David Boyes
Would a direct VSWITCH connection to the zLinux server (running RHEL AS 4) be more flexible in this area? In other words, is the Linux TCP/IP stack (w/ Zebra/OSPF) be more flexible? Miguel, I don't really expect you to answer this part of the question because you're not supposed to be a

Re: VM TCP/IP Routing Question

2006-08-02 Thread Marcy Cortes
this message. Thank you for your cooperation. -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Wednesday, August 02, 2006 13:46 To: IBMVM@LISTSERV.UARK.EDU Subject: Re: [IBMVM] VM TCP/IP Routing Question On Wednesday, 08/02