No, but it is possible that your VTAM thinks it already "owns" 0C13? What do you get when you do a "D U,,ALLOC,0C13,1" command? Your TCPIP STC should show up as the owner.
Mark Post -----Original Message----- From: Thomas Kunz [mailto:[EMAIL PROTECTED]] Sent: Friday, May 17, 2002 3:16 PM To: [EMAIL PROTECTED] Subject: Re: ESCON CTC between Linux LPAR and MVS LPAR Well now I can see that the kernel parm line file is being read and it has accepted my chandev=escon0 stuff, but I still get the following messages on the system log on the MVS LPAR: IEF196I IOS071I 0C13,34,NETB, MISSING DEVICE END IOS071I 0C13,34,NETB, MISSING DEVICE END 354 EZZ4310I ERROR: CODE=8010002B REPORTED ON DEVICE CTC0DEV. DIAGNOSTIC CODE: 04 EZZ4309I ATTEMPTING TO RECOVER DEVICE CTC0DEV here are my MVS tcpip profile definitions: DEVICE CTC0DEV CTC C12 LINK CTC0LINK CTC 1 CTC0DEV NETB in the above message is the started task name for our VTAM. Why would that show up in the message? Is there some VTAM configuration that must be done for this CTC connection? Thanks and have a great day! :-) Tom Kunz OIT/PTS Network & AIX Systems Support University of Minnesota 1300 S. 2nd St. Mpls., MN. 55454-1083 Suite 660 Phone: 612-624-8086 Fax: 612-626-1332 Email: [EMAIL PROTECTED] >>> [EMAIL PROTECTED] 05/17/02 10:02AM >>> Tom, You don't really need to worry about the differences with/without chandev right now. You're dealing with chandev, so we need to work with that scenario. There is no "chandev" command, per se. Your parmfile can be in EBCDIC. That's the result when all the VM folks use xedit to create theirs. You can tell if the kernel is seeing your parmfile because it echos the contents very near the beginning: Linux version 2.4.9-ac10 ([EMAIL PROTECTED]) (gcc version 2.95.3 20010315 (release)) #1 SMP Tue Nov 13 18:16:23 EST 2001 We are running under VM This machine has no IEEE fpu On node 0 totalpages: 32768 zone(0): 32768 pages. zone(1): 0 pages. zone(2): 0 pages. Kernel command line: root=/dev/dasda1 dasd=301,300,302-30f chandev=ctc0,0x0f00,0x0f01 It may be that when using devfs (which you're not), the channel number is concatenated to the device name, but I've never seen that happen myself. So, for now, I would assume that your device name is going to be ctc0, even though you specify "escon0,0x..." in your parameters. I believe a protocol of 3 is needed in your case. As Paul pointed out, if you want the Red Hat installation script to pick up your networking parameters, you need to specify the keyword in all upper case: CHANDEV=escon0,0xc30,0xc31,0,3 You should also specify it a second time in lower case (this is what gets passed to the kernel). chandev=escon0,0xc30,0xc31,0,3 Try specifying the HOST parameter for the installation scripts. This may override the value that gets used for the ifconfig command. HOST=fully.qualified.domain.name:ctc0:ip.add.re.ss:gateway.ip.add.ress Mark Post -----Original Message----- From: Thomas Kunz [mailto:[EMAIL PROTECTED]] Sent: Thursday, May 16, 2002 5:40 PM To: [EMAIL PROTECTED] Subject: Re: ESCON CTC between Linux LPAR and MVS LPAR Thanks for the suggestions - I did make sure the my read and write channels cross connected. I was reading the Device Drivers and Installation Commands for Linux Kernel 2.4 for S/390 and some things that I don't understand are: What is the difference between CTC/ESCON with the channel device layer Page 63 and CTC/ESCON without channel device layer. It seems like on the former one the syntax is escon0,0xc30,0xc31 and the latter one is ctc=escon0:0xc31:0xc31. When do you use one verses the other? Where does the chandev commande come in? Does my parm file command have to be created in ascii text or can I create it on MVS before I copy it to tape? When I boot it does not seem to acknowledge my parmfile. Some samples of my parm file that I have tried: dasd=730-733 ctc=noauto root=/dev/ram0 ro dasd=730-733 CHANDEV=escon0,0xc30,0xc31,0,3 root=/dev/ram0 ro dasd=730-733 CHANDEV=escon34,0xc30,0xc31,0,3 root=/dev/ram0 ro dasd=730-733 CHANDEV=escon0,0xc30,0xc31,0,1 root=/dev/ram0 ro dasd=730-733 CHANDEV=escon34,0xc30,0xc31,0,1 root=/dev/ram0 ro dasd=730-733 chandev=noauto,0x0,0xc29;noauto,0xc32,0xffff root=/dev/ram0 ro I used 34 sometimes because on page 64 of the Device Driver manual it says that the devicename ctc or escon is concatenated with the channel number. dasd=730-733 ctc=escon0:0xc30:0xc31:3 root=/dev/ram0 ro dasd=730-733 ctc=escon34:0xc30:0xc31:3 root=/dev/ram0 ro dasd=730-733 ctc=escon0:0xc30:0xc31:3 root=/dev/ram0 ro ctc=noauto dasd=730-733 ctc=escon34:0xc30:0xc31:3 ctc=noauto root=/dev/ram0 ro dasd=730-733 ctc=escon34:0xc30:0xc31:1 root=/dev/ram0 ro Was not sure what the protocol id should be - seemed like 3 was the best choice. When I get the prompt that asks "please enter network device" ctc0 escon0 iucv0 eth0 tr0 What should I enter? then it asks "please enter parameters to pass" why does it ask this if it is in the parm file? here I have tried many combinations like escon0,0xc30,0xc31 .... etc. and it so far has not liked anytime I enter escon and when I enter ctc0... is when I get all the error messages. Thanks and have a great day! :-) Tom Kunz OIT/PTS Network & AIX Systems Support University of Minnesota 1300 S. 2nd St. Mpls., MN. 55454-1083 Suite 660 Phone: 612-624-8086 Fax: 612-626-1332 Email: [EMAIL PROTECTED] >>> [EMAIL PROTECTED] 05/16/02 02:33PM >>> Tom, As you've discovered, there are some oddities about the way ESCON channel definitions work with Linux 2.4. One of those is that you should pass a value of "escon0" to the chandev layer, but the actual device name that gets created is ctc0. I view this as a bug, but haven't pursued it, yet. Make sure you have your read and write channels cross-connected. This can be done either in the MVS TCP/IP parameters, or via specifying your Linux/390 devices in the opposite order, i.e., if 0C12 connects to 0C30, and 0C13 connects to 0C31, then by specifying ctc0,0x0C31,0x0C30, you will have things correct. Normally if this is done wrong, the error messages that come out are fairly specific, but this looks like a good possibility. One area that you looked at could very well be a problem, but not necessarily of a nature to produce the errors you're seeing. If your MVS system is at 128.101.100.3, and your Linux/390 system is at 128.101.100.7, the CLAW won't know to forward traffic to .7 via .3 unless OROUTED is broadcasting that route. Even then, it's probably not good practice, and the Linux/390 system really should be in a separate subnet. As far as the kernel you are using, just about any of them should work with your current hardware. Mark Post -----Original Message----- From: Thomas Kunz [mailto:[EMAIL PROTECTED]] Sent: Thursday, May 16, 2002 9:56 AM To: [EMAIL PROTECTED] Subject: ESCON CTC between Linux LPAR and MVS LPAR Good Morning - I am trying to install redhat 7.2 linux on a s390 lpar. I downloaded the tapeinrd.img and the tapekrnl.img from ftp.redhat.com in /pub/redhat/linux/7.2/en/s390/images and created a parmfile. The system Boots fine and I get the prompt for a FQDN, etc. and a device type which if I select escon0 and fill in the rest of the IP stuff about virtual IP address and partner IP address I then get a response back that says: No such device escon0 SIOCSIFADDR No such device escon0 unknown interface. I have tried many different combinations when it asks to supply the parameters to pass - i.e. escon,0xc30,0xc3f,0,3 escon0,0xc30,0xc31 escon0,0xc30,0xc31,0,3 We do have escon ctc's defined. On the MVS 2.9 side it looks like this after I start TCPIP : UNIT TYPE STATUS 0C12 SCTC A-BSY 0C13 SCTC A-BSY In the above example if I use ctc0 instead of escon0 it accepts it but I never get a connection - netstat dev on mvs tcpip just shows connecting and not active and I get error messages: IEF196I IOS071I 0C13,34,NETB, MISSING DEVICE END IOS071I 0C13,34,NETB, MISSING DEVICE END 979 EZZ4310I ERROR: CODE=8010002B REPORTED ON DEVICE CTC0DEV. DIAGNOSTIC CODE: 04 EZZ4309I ATTEMPTING TO RECOVER DEVICE CTC0DEV which talk about XID's and VTAM or a channel error. My question(s) - do I have the right tape kernel images downloaded for escon support? Does anyone have a sample of what the parm file should look like regarding the ctc/escon parms? On the MVS side we run OROUTED - what should the bsdrouting parms look like? Do the IP addresses for the linux lpar and the mvs ctc have to be in different networks? I was using 128.101.100.7 for the linux IP address and 128.101.100.6 for the MVS CTC IP address. The MVS side has a CLAW interface to the network and is IP address 128.101.100.3. Is that a conflict? Anyone have a good resource of where to go to understand how this should be setup? Thanks and have a great day! :-) Tom Kunz OIT/PTS Network & AIX Systems Support University of Minnesota 1300 S. 2nd St. Mpls., MN. 55454-1083 Suite 660 Phone: 612-624-8086 Fax: 612-626-1332 Email: [EMAIL PROTECTED] ************************************************************************** This e-mail and its attachments have been scanned for viruses. NDIS/ADCS University of Minnesota **************************************************************************
