We access our 'test' Linux/390 via TCP/IP on the VSE/ESA system (from Connectivity Systems), their software provides a 'discover' command that really helps with this (not only on Linux); output looks like:
103 DISCOVER 152.159.87.113 08 Z1 0101 TCP910I Client manager connected. Generated on 10/28/01 at 23.57 Z1 0101 TCP915I DISCOVER 08 Z1 0101 TCP910I DISCOVER Ready: 08 Z1 0101 TCP915I SET HOST= 152.159.87.113 08 Z1 0101 TCP910I 152.159.087.113 08 Z1 0101 TCP910I DISCOVER Ready: 08 Z1 0101 TCP915I DISCOVER 08 Z1 0101 TCP910I Test for MTU 576 succeeded 08 Z1 0101 TCP910I Test for MTU 1501 failed 08 Z1 0101 TCP910I Test for MTU 1038 succeeded 08 Z1 0101 TCP910I Test for MTU 1269 succeeded 08 Z1 0101 TCP910I Test for MTU 1385 succeeded 08 Z1 0101 TCP910I Test for MTU 1443 succeeded 08 Z1 0101 TCP910I Test for MTU 1472 succeeded 08 Z1 0101 TCP910I Test for MTU 1486 succeeded 08 Z1 0101 TCP910I Test for MTU 1493 failed 08 Z1 0101 TCP910I Test for MTU 1489 succeeded 08 Z1 0101 TCP910I Test for MTU 1491 succeeded 08 Z1 0101 TCP910I Test for MTU 1492 succeeded 08 Z1 0101 TCP910I The best MTU discovered: 01492. I wonder if the OS/390 version has the same or similar command?. Hugo > -----Original Message----- > From: Nick Gimbrone [mailto:[EMAIL PROTECTED]] > Sent: Monday, December 10, 2001 11:02 PM > To: [EMAIL PROTECTED] > Subject: Re: Problem with Virtual CTCs between VM Linux guests and > OS/390 2.10 IP Stack. > > > > The MTU sizes on both sides of the CTC connection must > match. Make both > > sides 32768 or something. (VTAM CTC connections have the > same issues! If > > the buffer sizes don't match you get truncated PIUs...) > This is the nature > > of the CTC hardware (virtual or real). > No, it is in the nature of the drivers that they are not > constructed such that > they automatically "discover" the other side's MTU. Such > software could be > written, and it should be written... and then this > configuration pitfall would > be long gone (as it should be). > Peace. -njg >
