HIPERSOCKETS Not Working
Hello list. We are attempting to route some Linux traffic over HiperSockets that we have defined. The traffic flows in some directions but not all. Here is part of an email from the mainframe networking folks the describes the problem... We have hipersockets installed in all z/OS LPARS and in the z/VM LPAR. We have also installed it in two z/Linux guests. (utlz0002 and utlz0003). The z/OS LAPRS and the z/VM LPAR can now communicate to each other over the hipersocket links without errors. The z/OS LAPRS and the z/VM LPAR cannot communicate with the z/Linux guests. . The z/Linux guests can communicate with each other over the hipersocket link. The z/Linux guests can also communicate with the z/OS LAPRS and the z/VM LPAR over the hipersocket link. Symptom: A tcpdump from the z/Linux guest shows that a ping from a z/OS LPAR does arrive from the hipersocket link however no response to the ping was sent. Any help would be appreciated. Regards, Brent Litster Brent Litster Zions Management Services Company 2185 South 3270 West West Valley City 84119 (801) 844-5545 blits...@zionsbank.com === THIS ELECTRONIC MESSAGE, INCLUDING ANY ACCOMPANYING DOCUMENTS, IS CONFIDENTIAL and may contain information that is privileged and exempt from disclosure under applicable law. If you are neither the intended recipient nor responsible for delivering the message to the intended recipient, please note that any dissemination, distribution, copying or the taking of any action in reliance upon the message is strictly prohibited. If you have received this communication in error, please notify the sender immediately. Thank you.
Re: HIPERSOCKETS Not Working
On Monday, 01/31/2011 at 03:56 EST, Brent Litster brent.lits...@zionsbancorp.com wrote: A tcpdump from the z/Linux guest shows that a ping from a z/OS LPAR does arrive from the hipersocket link however no response to the ping was sent. The HiperSocket is working just fine, as demonstrated by your TCPDUMP. It's a routing problem. It's ALWAYS a routing problem. Look at the Linux guest's routing table to see where that packet is going. Ensure that z/OS and the Linux guest are using the same subnet mask and MTU size on their HiperSocket connection. When in doubt, draw your networks on a piece of paper and label them with IP addresses and subnet masks. It will quickly become apparent if you've got your virtual wires crossed. Alan Altmark z/VM and Linux on System z Consultant IBM System Lab Services and Training ibm.com/systems/services/labservices office: 607.429.3323 alan_altm...@us.ibm.com IBM Endicott
HiperSockets Definitions
I want to use hipersockets on a z/vm 5.3 host and I have the IBM HIPERSOCKETS IMPLEMENTATION GUIDE in section 4.3.1 it talks about the PROFILE TCPIP and the SYSTEM DTCPARMS files my question is how to get to these two files so I can update them. Thanks for your help Z/VM ROOKIE
Re: HiperSockets Definitions
Louis, They live on TCPMAINT 198. Good luck! Mark Wheeler UnitedHealth Group -- Excellence. Always. If Not Excellence, What? If Not Excellence Now, When? Tom Peters, author of The Little BIG Things Date: Thu, 18 Nov 2010 14:54:17 -0600 From: louis.gai...@its.ms.gov Subject: HiperSockets Definitions To: IBMVM@LISTSERV.UARK.EDU I want to use hipersockets on a z/vm 5.3 host and I have the IBM HIPERSOCKETS IMPLEMENTATION GUIDE in section 4.3.1 it talks about the PROFILE TCPIP and the SYSTEM DTCPARMS files my question is how to get to these two files so I can update them. Thanks for your help Z/VM ROOKIE
Re: HiperSockets Definitions
No fears.
1) Logon to the userid TCPMAINT.
2) ACCESS 198 P
{Just think of filemode 'P' as Production - it could be any free
filemode}
3) COPYFILE PROFILE TCPIP P = -1TCPIP P (OLDDATE
{This makes a backup copy of the current PROFILE TCPIP P, saving the
same file date/time,
but named PROFILE -1TCPIP P, think of it a little like a GDG on
z/OS}
4) COPYFILE SYSTEM DTCPARMS P = -1DTCPARMS P (OLDDATE
{Same backup story as step 3}
5) Use XEDIT PROFILE TCPIP P to make your changes, then to the same with
SYSTEM DTCPARMS P
Note: Be VERY CAREFUL when changing TCPIP files. You can take break your
network.
Mike Walter
Aon Corporation
The opinions expressed herein are mine alone, not my employer's.
louis.gai...@its.ms.gov
Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU
11/18/2010 02:54 PM
Please respond to
The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU
To
IBMVM@LISTSERV.UARK.EDU
cc
Subject
HiperSockets Definitions
I want to use hipersockets on a z/vm 5.3 host and I have the IBM
HIPERSOCKETS IMPLEMENTATION GUIDE
in section 4.3.1 it talks about the PROFILE TCPIP and the SYSTEM DTCPARMS
files my question is
how to get to these two files so I can update them.
Thanks for your help
Z/VM ROOKIE
The information contained in this e-mail and any accompanying documents may
contain information that is confidential or otherwise protected from
disclosure. If you are not the intended recipient of this message, or if this
message has been addressed to you in error, please immediately alert the sender
by reply e-mail and then delete this message, including any attachments. Any
dissemination, distribution or other use of the contents of this message by
anyone other than the intended recipient is strictly prohibited. All messages
sent to and from this e-mail address may be monitored as permitted by
applicable law and regulations to ensure compliance with our internal policies
and to protect our business. E-mails are not secure and cannot be guaranteed to
be error free as they can be intercepted, amended, lost or destroyed, or
contain viruses. You are deemed to have accepted these risks if you communicate
with us by e-mail.
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
On Friday, 04/30/2010 at 12:29 EDT, Martin, Terry R. (CMS/CTR) (CTR) terry.mar...@cms.hhs.gov wrote: Thanks a bunch Rich for the information! I think the Firewall guys think if we use a Hipersocket than you do not need the Firewall rules. But I told them that even if Hipersockets are used you would still need Firewall rules if required. Yes, and that's why I said you can't use HiperSockets to connect LPARs in different zones: No firewall. Any time data transits to a different zone, it goes thru a firewall. It is the firewalls that create the zones. Note that firewall and packet filter are not the same. A host may have a resident IPS/IDS solution, but it isn't a firewall. Firewalls are stand-alone devices. Alan Altmark z/VM Development IBM Endicott
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Hi Alan, Thanks for the information. It is much appreciated! Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Alan Altmark Sent: Friday, April 30, 2010 2:58 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS On Friday, 04/30/2010 at 12:29 EDT, Martin, Terry R. (CMS/CTR) (CTR) terry.mar...@cms.hhs.gov wrote: Thanks a bunch Rich for the information! I think the Firewall guys think if we use a Hipersocket than you do not need the Firewall rules. But I told them that even if Hipersockets are used you would still need Firewall rules if required. Yes, and that's why I said you can't use HiperSockets to connect LPARs in different zones: No firewall. Any time data transits to a different zone, it goes thru a firewall. It is the firewalls that create the zones. Note that firewall and packet filter are not the same. A host may have a resident IPS/IDS solution, but it isn't a firewall. Firewalls are stand-alone devices. Alan Altmark z/VM Development IBM Endicott
Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Hi Currently we use Hipersockets between on z/OS LPARS and our z/Linux guests. This works great. The question I have is can Hipersockets be used to go from one z/Linux running on one z/VM LPAR to another? Currently we us VSWITCHES with OSA to communicate between z/Linux guests. Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
On 04/29/2010 01:57 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Hi Currently we use Hipersockets between on z/OS LPARS and our z/Linux guests. This works great. The question I have is can Hipersockets be used to go from one z/Linux running on one z/VM LPAR to another? Currently we us VSWITCHES with OSA to communicate between z/Linux guests. /Thank You,/ / / /Terry Martin/ /Lockheed Martin - Citic/ /z/OS and z/VM Performance Tuning and Operating Systems Support/ /Office - 443 348-2102/ /Cell - 443 632-4191/ / / Yes, just attach a hipersocket triplet to each Linux guest in each of the LPARs. Configure the hsi interfaces appropriately and you're off to the races. -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
On Thursday, 04/29/2010 at 03:20 EDT, Rich Smrcina r...@velocitysoftware.com wrote: Yes, just attach a hipersocket triplet to each Linux guest in each of the LPARs. Configure the hsi interfaces appropriately and you're off to the races. But before you do that, you need to check with the network security folks. It is unlikely that they will permit an app server and a db server to be in the same security zone (i.e. connected without a firewall). But, either way, it's their call. Alan Altmark z/VM Development IBM Endicott
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Thanks for the information. For some reason I was under the impression that you could not use Hipersockets to talk between z/Linux guests, this is why we set up Vswitches and talk between guests using OSA. Am I confusing something here? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Rich Smrcina Sent: Thursday, April 29, 2010 3:20 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS On 04/29/2010 01:57 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Hi Currently we use Hipersockets between on z/OS LPARS and our z/Linux guests. This works great. The question I have is can Hipersockets be used to go from one z/Linux running on one z/VM LPAR to another? Currently we us VSWITCHES with OSA to communicate between z/Linux guests. /Thank You,/ / / /Terry Martin/ /Lockheed Martin - Citic/ /z/OS and z/VM Performance Tuning and Operating Systems Support/ /Office - 443 348-2102/ /Cell - 443 632-4191/ / / Yes, just attach a hipersocket triplet to each Linux guest in each of the LPARs. Configure the hsi interfaces appropriately and you're off to the races. -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
You *can* use hipersockets to talk between Linux guests. If the guests are running in the same z/VM system it makes more sense to use the VSwitch, than to use hipersockets. Of course, if the Linux guests are in different LPARs then you have to use hipersockets. I defer the security aspect of the discussion to the gentleman that had his *other* hat on earlier On 04/29/2010 09:26 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Thanks for the information. For some reason I was under the impression that you could not use Hipersockets to talk between z/Linux guests, this is why we set up Vswitches and talk between guests using OSA. Am I confusing something here? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Rich thanks for the information. One last thing you mention that If the guests are running in the same z/VM system it makes more sense to use the VSwitch, than to use hipersockets. Why is this? Is that the benefit of the Hipersockets is lost? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Rich Smrcina Sent: Thursday, April 29, 2010 10:38 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS You *can* use hipersockets to talk between Linux guests. If the guests are running in the same z/VM system it makes more sense to use the VSwitch, than to use hipersockets. Of course, if the Linux guests are in different LPARs then you have to use hipersockets. I defer the security aspect of the discussion to the gentleman that had his *other* hat on earlier On 04/29/2010 09:26 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Thanks for the information. For some reason I was under the impression that you could not use Hipersockets to talk between z/Linux guests, this is why we set up Vswitches and talk between guests using OSA. Am I confusing something here? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Why go to the hw when vm can do it? And the use of ucb is a problem at some shops. Yes, 65k isn't enough. Marcy - Original Message - From: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To: IBMVM@LISTSERV.UARK.EDU IBMVM@LISTSERV.UARK.EDU Sent: Thu Apr 29 22:06:45 2010 Subject: Re: [IBMVM] Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS Rich thanks for the information. One last thing you mention that If the guests are running in the same z/VM system it makes more sense to use the VSwitch, than to use hipersockets. Why is this? Is that the benefit of the Hipersockets is lost? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Rich Smrcina Sent: Thursday, April 29, 2010 10:38 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS You *can* use hipersockets to talk between Linux guests. If the guests are running in the same z/VM system it makes more sense to use the VSwitch, than to use hipersockets. Of course, if the Linux guests are in different LPARs then you have to use hipersockets. I defer the security aspect of the discussion to the gentleman that had his *other* hat on earlier On 04/29/2010 09:26 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Thanks for the information. For some reason I was under the impression that you could not use Hipersockets to talk between z/Linux guests, this is why we set up Vswitches and talk between guests using OSA. Am I confusing something here? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Vswitch is internal to CP, hipersockets makes most sense to move data between LPARs. Using hipersockets requires a trip out of the VM system (even if it's between virtual machines), whereas using the VSwitch stays within CP. On 04/29/2010 10:06 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Rich thanks for the information. One last thing you mention that If the guests are running in the same z/VM system it makes more sense to use the VSwitch, than to use hipersockets. Why is this? Is that the benefit of the Hipersockets is lost? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Thanks Rich. So in my case if I want my APP Layer z/Linux guest on the APP Layer LPAR to talk to my Data Layer z/Linux guest on the Data Layer LPAR I would use Hipersockets. Once I am in the Data Layer LPAR and want to talk between other Data Layer guests within the LPAR the use of a Vswitch is the way to go. Am I understanding this now? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Rich Smrcina Sent: Thursday, April 29, 2010 11:14 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS Vswitch is internal to CP, hipersockets makes most sense to move data between LPARs. Using hipersockets requires a trip out of the VM system (even if it's between virtual machines), whereas using the VSwitch stays within CP. On 04/29/2010 10:06 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Rich thanks for the information. One last thing you mention that If the guests are running in the same z/VM system it makes more sense to use the VSwitch, than to use hipersockets. Why is this? Is that the benefit of the Hipersockets is lost? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
And one last thing Would I have both a Ethernet connection and Hipersockets(His) interface set up on all of the guests to handle both methods? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -Original Message- From: Martin, Terry R. (CMS/CTR) (CTR) Sent: Thursday, April 29, 2010 11:26 PM To: IBMVM@LISTSERV.UARK.EDU Subject: RE: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS Thanks Rich. So in my case if I want my APP Layer z/Linux guest on the APP Layer LPAR to talk to my Data Layer z/Linux guest on the Data Layer LPAR I would use Hipersockets. Once I am in the Data Layer LPAR and want to talk between other Data Layer guests within the LPAR the use of a Vswitch is the way to go. Am I understanding this now? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Rich Smrcina Sent: Thursday, April 29, 2010 11:14 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS Vswitch is internal to CP, hipersockets makes most sense to move data between LPARs. Using hipersockets requires a trip out of the VM system (even if it's between virtual machines), whereas using the VSwitch stays within CP. On 04/29/2010 10:06 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Rich thanks for the information. One last thing you mention that If the guests are running in the same z/VM system it makes more sense to use the VSwitch, than to use hipersockets. Why is this? Is that the benefit of the Hipersockets is lost? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Correct. As long as all of those LPARs are in the same CEC, everything is good. That guy with the security hat on it looking pretty squirrely, careful now On 04/29/2010 10:26 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: Thanks Rich. So in my case if I want my APP Layer z/Linux guest on the APP Layer LPAR to talk to my Data Layer z/Linux guest on the Data Layer LPAR I would use Hipersockets. Once I am in the Data Layer LPAR and want to talk between other Data Layer guests within the LPAR the use of a Vswitch is the way to go. Am I understanding this now? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Yes. On 04/29/2010 10:27 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: And one last thing Would I have both a Ethernet connection and Hipersockets(His) interface set up on all of the guests to handle both methods? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS
Thanks a bunch Rich for the information! I think the Firewall guys think if we use a Hipersocket than you do not need the Firewall rules. But I told them that even if Hipersockets are used you would still need Firewall rules if required. Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Rich Smrcina Sent: Thursday, April 29, 2010 11:41 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Hipersockets between z/VM LPARS and z/Linux guests running on the LPARS Yes. On 04/29/2010 10:27 PM, Martin, Terry R. (CMS/CTR) (CTR) wrote: And one last thing Would I have both a Ethernet connection and Hipersockets(His) interface set up on all of the guests to handle both methods? Thank You, Terry Martin Lockheed Martin - Citic z/OS and z/VM Performance Tuning and Operating Systems Support Office - 443 348-2102 Cell - 443 632-4191 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2011
VSWITCH vs. HiperSockets
It seems that the performance using HiperSockets is superior to that of VSWITCH (according to z/VSE Using DB2 on Linux for System z, Chapter 2. Planning DB2, pp 30-37). Comments? Frank M. Ramaekers Jr. Systems Programmer MCP, MCP+I, MCSE RHCE American Income Life Insurance Co. Phone: (254)761-6649 1200 Wooded Acres Dr. Fax: (254)741-5777 Waco, Texas 76701 _ This message contains information which is privileged and confidential and is solely for the use of the intended recipient. If you are not the intended recipient, be aware that any review, disclosure, copying, distribution, or use of the contents of this message is strictly prohibited. If you have received this in error, please destroy it immediately and notify us at privacy...@ailife.com.
Re: VSWITCH vs. HiperSockets
That appears to be dependent upon the traffic. If the choice is between VSwitch and Hipersockets, then the traffic must be moving between virtual machines. Why leave the VM system to move packets? On 04/12/2010 01:21 PM, Frank M. Ramaekers wrote: It seems that the performance using HiperSockets is superior to that of VSWITCH (according to “z/VSE Using DB2 on Linux for System z”, Chapter 2. Planning DB2, pp 30-37). Comments? Frank M. Ramaekers Jr. Systems Programmer MCP, MCP+I, MCSE RHCE American Income Life Insurance Co. Phone: (254)761-6649 1200 Wooded Acres Dr. Fax: (254)741-5777 Waco, Texas 76701 _ This message contains information which is privileged and confidential and is solely for the use of the intended recipient. If you are not the intended recipient, be aware that any review, disclosure, copying, distribution, or use of the contents of this message is strictly prohibited. If you have received this in error, please destroy it immediately and notify us at privacy...@ailife.com. -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2010 - Apr 9-13, 2010 Covington, KY
Re: hipersockets on a VM layer 2 vswitch
Yes but the layer 2 vswitch is running on z/VM ver. 5.4. Does this mean we will need to set up a layer 2 hipersocket on the z/OS side as well? Thank you Mark, yes we are running z/OS ver. 1.10 on the LPAR running the TSMTCS app. Russell -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Mark Jacobs Sent: Wednesday, February 03, 2010 4:32 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: hipersockets on a VM layer 2 vswitch That lpar is running zOS 1.10. Mark Jacobs -Original Message- From: The IBM z/VM Operating System on behalf of David Boyes Sent: Wed 2/3/2010 4:23 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: hipersockets on a VM layer 2 vswitch You need a very modern z/OS for layer 2 connections to work. You don't say what level of z.OS, but I think it has to be minimum 1.10 if not 1.11. On 2/3/10 2:54 PM, russell.gendr...@custserv.com russell.gendr...@custserv.com wrote: Greetings, We currently have multiple Linux guest running on z/VM on both a layer 3 tcp/ip vswitch as well as a layer 2 vswitch. On one particular z/VM we have a hipersocket link to a z/OS system for using the TSMTCS backup application. Currently only the layer 3 guests use the hipersockets. We tried to define the hipersocket link on the z/OS system To a layer 2 Linux guest but there is no connection for some reason. Has anyone else tried this and if so were you successful? Thanks Russell Gendreau Time Customer Service, Inc. 813-554-2064
hipersockets on a VM layer 2 vswitch
Greetings, We currently have multiple Linux guest running on z/VM on both a layer 3 tcp/ip vswitch as well as a layer 2 vswitch. On one particular z/VM we have a hipersocket link to a z/OS system for using the TSMTCS backup application. Currently only the layer 3 guests use the hipersockets. We tried to define the hipersocket link on the z/OS system To a layer 2 Linux guest but there is no connection for some reason. Has anyone else tried this and if so were you successful? Thanks Russell Gendreau Time Customer Service, Inc. 813-554-2064
Re: hipersockets on a VM layer 2 vswitch
You need a very modern z/OS for layer 2 connections to work. You don’t say what level of z.OS, but I think it has to be minimum 1.10 if not 1.11. On 2/3/10 2:54 PM, russell.gendr...@custserv.com russell.gendr...@custserv.com wrote: Greetings, We currently have multiple Linux guest running on z/VM on both a layer 3 tcp/ip vswitch as well as a layer 2 vswitch. On one particular z/VM we have a hipersocket link to a z/OS system for using the TSMTCS backup application. Currently only the layer 3 guests use the hipersockets. We tried to define the hipersocket link on the z/OS system To a layer 2 Linux guest but there is no connection for some reason. Has anyone else tried this and if so were you successful? Thanks Russell Gendreau Time Customer Service, Inc. 813-554-2064
Re: hipersockets on a VM layer 2 vswitch
That lpar is running zOS 1.10. Mark Jacobs -Original Message- From: The IBM z/VM Operating System on behalf of David Boyes Sent: Wed 2/3/2010 4:23 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: hipersockets on a VM layer 2 vswitch You need a very modern z/OS for layer 2 connections to work. You don't say what level of z.OS, but I think it has to be minimum 1.10 if not 1.11. On 2/3/10 2:54 PM, russell.gendr...@custserv.com russell.gendr...@custserv.com wrote: Greetings, We currently have multiple Linux guest running on z/VM on both a layer 3 tcp/ip vswitch as well as a layer 2 vswitch. On one particular z/VM we have a hipersocket link to a z/OS system for using the TSMTCS backup application. Currently only the layer 3 guests use the hipersockets. We tried to define the hipersocket link on the z/OS system To a layer 2 Linux guest but there is no connection for some reason. Has anyone else tried this and if so were you successful? Thanks Russell Gendreau Time Customer Service, Inc. 813-554-2064
VLAN, VSWITCH, HiperSockets
I'd like to move all of our internal communications internally. Basically this consists of TCP/IP for z/VM and several z/VSEs. This includes PNET between z/VSEs. What would you recommend? I'm leaning toward HiperSockets. Pros Cons? Frank M. Ramaekers Jr. Systems Programmer MCP, MCP+I, MCSE RHCE American Income Life Insurance Co. Phone: (254)761-6649 1200 Wooded Acres Dr.Fax: (254)741-5777 Waco, Texas 76710 _ This message contains information which is privileged and confidential and is solely for the use of the intended recipient. If you are not the intended recipient, be aware that any review, disclosure, copying, distribution, or use of the contents of this message is strictly prohibited. If you have received this in error, please destroy it immediately and notify us at privacy...@ailife.com.
Re: VLAN, VSWITCH, HiperSockets
If the VSE's are guests on zVM - use VSWITCH. If they are LPARs -- then either hipersockets or maybe a shared OSA port (which can be uncabled).. Scott On Thu, Oct 29, 2009 at 9:31 AM, Frank M. Ramaekers framaek...@ailife.comwrote: I'd like to move all of our internal communications internally. Basically this consists of TCP/IP for z/VM and several z/VSEs. This includes PNET between z/VSEs. What would you recommend? I'm leaning toward HiperSockets. Pros Cons? Frank M. Ramaekers Jr. Systems Programmer MCP, MCP+I, MCSE RHCE American Income Life Insurance Co. Phone: (254)761-6649 1200 Wooded Acres Dr.Fax: (254)741-5777 Waco, Texas 76710 _ This message contains information which is privileged and confidential and is solely for the use of the intended recipient. If you are not the intended recipient, be aware that any review, disclosure, copying, distribution, or use of the contents of this message is strictly prohibited. If you have received this in error, please destroy it immediately and notify us at privacy...@ailife.com.
Re: VLAN, VSWITCH, HiperSockets
If the VSE's are running in the same z/VM system then use the VSWITCH. If they are not, but in the same CEC, use hipersockets. Frank M. Ramaekers wrote: I'd like to move all of our internal communications internally. Basically this consists of TCP/IP for z/VM and several z/VSEs. This includes PNET between z/VSEs. What would you recommend? I'm leaning toward HiperSockets. Pros Cons? Frank M. Ramaekers Jr. Systems Programmer MCP, MCP+I, MCSE RHCE American Income Life Insurance Co. Phone: (254)761-6649 1200 Wooded Acres Dr.Fax: (254)741-5777 Waco, Texas 76710 -- Rich Smrcina
Re: VLAN, VSWITCH, HiperSockets
ETHERNET or IP? Advantages/disadvantages? Frank M. Ramaekers Jr. Systems Programmer MCP, MCP+I, MCSE RHCE American Income Life Insurance Co. Phone: (254)761-6649 1200 Wooded Acres Dr.Fax: (254)741-5777 Waco, Texas 76710 -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Rich Smrcina Sent: Thursday, October 29, 2009 10:41 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: VLAN, VSWITCH, HiperSockets If the VSE's are running in the same z/VM system then use the VSWITCH. If they are not, but in the same CEC, use hipersockets. Frank M. Ramaekers wrote: I'd like to move all of our internal communications internally. Basically this consists of TCP/IP for z/VM and several z/VSEs. This includes PNET between z/VSEs. What would you recommend? I'm leaning toward HiperSockets. Pros Cons? Frank M. Ramaekers Jr. Systems Programmer MCP, MCP+I, MCSE RHCE American Income Life Insurance Co. Phone: (254)761-6649 1200 Wooded Acres Dr.Fax: (254)741-5777 Waco, Texas 76710 -- Rich Smrcina _ This message contains information which is privileged and confidential and is solely for the use of the intended recipient. If you are not the intended recipient, be aware that any review, disclosure, copying, distribution, or use of the contents of this message is strictly prohibited. If you have received this in error, please destroy it immediately and notify us at privacy...@ailife.com.
Re: VLAN, VSWITCH, HiperSockets
If they're VM guests, use VSWITCHes. Only use hipersockets if you have to cross LPARs. Hipersockets are a giant PITA to define and maintain. On 10/29/09 4:31 PM, Frank M. Ramaekers framaek...@ailife.com wrote: I'd like to move all of our internal communications internally. Basically this consists of TCP/IP for z/VM and several z/VSEs. This includes PNET between z/VSEs. What would you recommend? I'm leaning toward HiperSockets.
Re: VLAN, VSWITCH, HiperSockets
On 10/29/09 7:45 PM, Frank M. Ramaekers framaek...@ailife.com wrote: ETHERNET or IP? Advantages/disadvantages? For Linux, I tend to use ETHERNET, since they're likely to actually do more than IPv4. For others, or for backlevel systems that can't do layer 2, use IP. The ethernet ones tend to use a little more CPU since the ARP processing has to be done in each virtual machine vs having it suppressed in the IP type VSWITCH. If you do (or plan to do) IPv6, then you're going to want the layer 2 stuff.
Re: VLAN, VSWITCH, HiperSockets
I can find no indication that z/VSE supports ETHERNET (layer 2) mode yet. Frank M. Ramaekers wrote: ETHERNET or IP? Advantages/disadvantages? Frank M. Ramaekers Jr. Systems Programmer MCP, MCP+I, MCSE RHCE American Income Life Insurance Co. Phone: (254)761-6649 1200 Wooded Acres Dr.Fax: (254)741-5777 Waco, Texas 76710 -- Rich Smrcina Phone: 414-491-6001 http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2010 - Apr 9-13, 2010 Covington, KY
Re: FTP over HiperSockets network
Sort of. I have a z/OS that connects to a zLinux server through Hipersockets. On Fri, Jan 23, 2009 at 12:19 AM, Martin, Terry R. (CMS/CTR) (CTR) terry.mar...@cms.hhs.gov wrote: Hi Is anyone running FTP on z/VM connecting to FTP server on z/OS over a HiperSockets network? If so can you share with me the steps to take to get this working? Thanks in advance for the help!!! Terry -- Mark Pace Mainline Information Systems 1700 Summit Lake Drive Tallahassee, FL. 32317
Re: FTP over HiperSockets network
On Friday, 01/23/2009 at 12:20 EST, Martin, Terry R. (CMS/CTR) (CTR) terry.mar...@cms.hhs.gov wrote: Is anyone running FTP on z/VM connecting to FTP server on z/OS over a HiperSockets network? If so can you share with me the steps to take to get this working? The general sequence is: 1. Define a HiperSocket chpid and subchannels. Give the z/VM partition access to three of them (at least) and bring them online. 2. Add DEDICATE statements for the three subchannels to TCPIP's directory entry. 3. Add DEVICE/LINK/HOME/START statements to PROFILE TCPIP. The IFCONFIG command can assist you if you're uncertain about the statements you need. But don't take my word for it! Take a gander at the often-ignored and always-overlooked z/VM Connectivity book in the z/VM library. ;-) Please note that there is an error on p.82 IODEVICE ...,UNIT=IDQ USERPRM=(parmname,OS_num) is incorrect. It should read IODEVICE ...,UNIT=IDQ,CHPARM=(code) where (code) has the OS values in the table. Alan Altmark z/VM Development IBM Endicott
Re: FTP over HiperSockets network
Alan, we put ours in the individual user definition. It works. Should we have put it under TCPIP as you stated below, and what's the dif? Thanks. USER LNX052 PGDN52 768M 768M INCLUDE LINDFLT DEDICATE ED09 ED09 DEDICATE ED0A ED0A DEDICATE ED0B ED0B David Dean Information Systems *bcbstauthorized* -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Alan Altmark Sent: Friday, January 23, 2009 11:02 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: FTP over HiperSockets network On Friday, 01/23/2009 at 12:20 EST, Martin, Terry R. (CMS/CTR) (CTR) terry.mar...@cms.hhs.gov wrote: Is anyone running FTP on z/VM connecting to FTP server on z/OS over a HiperSockets network? If so can you share with me the steps to take to get this working? The general sequence is: 1. Define a HiperSocket chpid and subchannels. Give the z/VM partition access to three of them (at least) and bring them online. 2. Add DEDICATE statements for the three subchannels to TCPIP's directory entry. 3. Add DEVICE/LINK/HOME/START statements to PROFILE TCPIP. The IFCONFIG command can assist you if you're uncertain about the statements you need. But don't take my word for it! Take a gander at the often-ignored and always-overlooked z/VM Connectivity book in the z/VM library. ;-) Please note that there is an error on p.82 IODEVICE ...,UNIT=IDQ USERPRM=(parmname,OS_num) is incorrect. It should read IODEVICE ...,UNIT=IDQ,CHPARM=(code) where (code) has the OS values in the table. Alan Altmark z/VM Development IBM Endicott - Please see the following link for the BlueCross BlueShield of Tennessee E-mail disclaimer: http://www.bcbst.com/email_disclaimer.shtm
Re: FTP over HiperSockets network
On 1/23/2009 at 11:12 AM, Dean, David (I/S) david_d...@bcbst.com wrote: Alan, we put ours in the individual user definition. It works. Should we have put it under TCPIP as you stated below, and what's the dif? Terry wants to use it from z/VM, not Linux. So, the devices need to be given to TCP/IP. Mark Post
Re: FTP over HiperSockets network
Got it, thanks. David Dean Information Systems *bcbstauthorized* -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Mark Post Sent: Friday, January 23, 2009 11:17 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: FTP over HiperSockets network On 1/23/2009 at 11:12 AM, Dean, David (I/S) david_d...@bcbst.com wrote: Alan, we put ours in the individual user definition. It works. Should we have put it under TCPIP as you stated below, and what's the dif? Terry wants to use it from z/VM, not Linux. So, the devices need to be given to TCP/IP. Mark Post - Please see the following link for the BlueCross BlueShield of Tennessee E-mail disclaimer: http://www.bcbst.com/email_disclaimer.shtm
Re: FTP over HiperSockets network
On 1/23/09 11:02 AM, Alan Altmark alan_altm...@us.ibm.com wrote: On Friday, 01/23/2009 at 12:20 EST, Martin, Terry R. (CMS/CTR) (CTR) terry.mar...@cms.hhs.gov wrote: Is anyone running FTP on z/VM connecting to FTP server on z/OS over a HiperSockets network? If so can you share with me the steps to take to get this working? The general sequence is: 1. Define a HiperSocket chpid and subchannels. Give the z/VM partition access to three of them (at least) and bring them online. 2. Add DEDICATE statements for the three subchannels to TCPIP's directory entry. 3. Add DEVICE/LINK/HOME/START statements to PROFILE TCPIP. The IFCONFIG command can assist you if you're uncertain about the statements you need. But don't take my word for it! Take a gander at the often-ignored and always-overlooked z/VM Connectivity book in the z/VM library. ;-) Please note that there is an error on p.82 IODEVICE ...,UNIT=IDQ USERPRM=(parmname,OS_num) is incorrect. It should read IODEVICE ...,UNIT=IDQ,CHPARM=(code) where (code) has the OS values in the table. You should also consider using a Linux guest as a firewall/NAT router between the physical hipersocket and a VSWITCH or guest LAN instead of dedicating the connection to the VM stack. Since the number of hipersockets are limited, and setting them up is a royal PITA, a setup like that allows more guest systems to take advantage of the connection w/o reinventing the wheel every time, and provides significant traffic shaping and management capabilities. It will cost you some cycles, but it's worth considering if you need a managed path between the guests and don't want to use an external switch path for network performance or security reasons.
Re: FTP over HiperSockets network
Thanks a lot for the information Alan! Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 terry.mar...@cms.hhs.gov -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Alan Altmark Sent: Friday, January 23, 2009 11:02 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: FTP over HiperSockets network On Friday, 01/23/2009 at 12:20 EST, Martin, Terry R. (CMS/CTR) (CTR) terry.mar...@cms.hhs.gov wrote: Is anyone running FTP on z/VM connecting to FTP server on z/OS over a HiperSockets network? If so can you share with me the steps to take to get this working? The general sequence is: 1. Define a HiperSocket chpid and subchannels. Give the z/VM partition access to three of them (at least) and bring them online. 2. Add DEDICATE statements for the three subchannels to TCPIP's directory entry. 3. Add DEVICE/LINK/HOME/START statements to PROFILE TCPIP. The IFCONFIG command can assist you if you're uncertain about the statements you need. But don't take my word for it! Take a gander at the often-ignored and always-overlooked z/VM Connectivity book in the z/VM library. ;-) Please note that there is an error on p.82 IODEVICE ...,UNIT=IDQ USERPRM=(parmname,OS_num) is incorrect. It should read IODEVICE ...,UNIT=IDQ,CHPARM=(code) where (code) has the OS values in the table. Alan Altmark z/VM Development IBM Endicott
Re: FTP over HiperSockets network
Yes, that is correct Mark! Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 terry.mar...@cms.hhs.gov -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Mark Post Sent: Friday, January 23, 2009 11:17 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: FTP over HiperSockets network On 1/23/2009 at 11:12 AM, Dean, David (I/S) david_d...@bcbst.com wrote: Alan, we put ours in the individual user definition. It works. Should we have put it under TCPIP as you stated below, and what's the dif? Terry wants to use it from z/VM, not Linux. So, the devices need to be given to TCP/IP. Mark Post
FTP over HiperSockets network
Hi Is anyone running FTP on z/VM connecting to FTP server on z/OS over a HiperSockets network? If so can you share with me the steps to take to get this working? Thanks in advance for the help!!! Terry
Re: HiperSockets Question
On Wednesday, 08/20/2008 at 09:54 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: After a lot of effort we found that there was an extra route in the Linux routing table that sent the data bound for one of the z/OS LPARS over to our z/VM where it then was routed by the TCP/IP stack on z/VM to the z/OS LPAR over the HiperSockets LINK. This LINK happened to have a MTU size of 1500. When we removed the bogus route from the Linux routing table causing the SQL queries in question to travel over a HiperSockets LINK with a MTU size of 16348 they were successful. It appears that the difference in the MTU sizes was the problem. My question is does the MTU size mismatch cause the problem? I thought that even if the packets were fragmented they would still get over in full. Yes, it can (but not always) cause a problem if the MTU sizes on the same shared media are different. All MTUs on the same network segment need to have the same MTU. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Question
Thanks Alan! Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Friday, August 22, 2008 9:11 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Question On Wednesday, 08/20/2008 at 09:54 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: After a lot of effort we found that there was an extra route in the Linux routing table that sent the data bound for one of the z/OS LPARS over to our z/VM where it then was routed by the TCP/IP stack on z/VM to the z/OS LPAR over the HiperSockets LINK. This LINK happened to have a MTU size of 1500. When we removed the bogus route from the Linux routing table causing the SQL queries in question to travel over a HiperSockets LINK with a MTU size of 16348 they were successful. It appears that the difference in the MTU sizes was the problem. My question is does the MTU size mismatch cause the problem? I thought that even if the packets were fragmented they would still get over in full. Yes, it can (but not always) cause a problem if the MTU sizes on the same shared media are different. All MTUs on the same network segment need to have the same MTU. Alan Altmark z/VM Development IBM Endicott
HiperSockets Question
Hi Alan, I thought I sent this out last night but it does not look like it went so I am sending it again. If for some reason you did get this I apologize for sending it again. Not to belabor the subject I just want to make sure I understand. Here is what my issue was. I processed many SQL queries of DB2 tables via DB2 connect running on a z/OS LPAR down to a Linux server running on the same machine but a different z/VM LPAR without any issues. Until, we ran across some queries that were larger than all of the others. These queries failed (timed out). We saw in the DB2 connect trace that the size of the queries was about 2.3K. We knew it was a size issue because when we did a 'SELECT *' on the tables, basically reducing the size it worked. After a lot of effort we found that there was an extra route in the Linux routing table that sent the data bound for one of the z/OS LPARS over to our z/VM where it then was routed by the TCP/IP stack on z/VM to the z/OS LPAR over the HiperSockets LINK. This LINK happened to have a MTU size of 1500. When we removed the bogus route from the Linux routing table causing the SQL queries in question to travel over a HiperSockets LINK with a MTU size of 16348 they were successful. It appears that the difference in the MTU sizes was the problem. My question is does the MTU size mismatch cause the problem? I thought that even if the packets were fragmented they would still get over in full. Sorry for being so wordy but I thought I would present to try to help you help me and to also help others that may be just getting into this as I am. Thanks again for all of your help!! Terry Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED]
HiperSockets Question
Hi If I am using HiperSockets and my MTU size is 1500 and a pass a packet size of 2000 will the data in the packet be truncated since it is bigger than 1500? I thought I read that since HiperSockets is memory transfers that the MTU can not be split like it would over a normal transmission pipe, meaning that for a 3K packet size over a MTU 1500 link would not be split into 2 transmissions. Am I correct in the assumption? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED]
Re: HiperSockets Question
On Wednesday, 08/20/2008 at 03:43 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: If I am using HiperSockets and my MTU size is 1500 and a pass a packet size of 2000 will the data in the packet be truncated since it is bigger than 1500? I thought I read that since HiperSockets is memory transfers that the MTU can not be split like it would over a normal transmission pipe, meaning that for a 3K packet size over a MTU 1500 link would not be split into 2 transmissions. Am I correct in the assumption? Yes and no. :-) It won't be truncated, but it will be fragmented. The MTU is the Guardian of the Gate. It prevents the IP layer from sending a packet too large for the underlying interface to handle. So a 3K packet over an MTU 1500 link will result in two fragments, each 1500 bytes. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Question
Packet fragmentation is done in the TCP/IP stack. The stack will chunk the packet up into smaller packets before sending it over the HiperSockets device. Regards, Miguel Delapaz z/VM TCP/IP Development The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU wrote on 08/20/2008 12:40:47 PM: [image removed] HiperSockets Question Martin, Terry R. (CMS/CTR) (CTR) to: IBMVM 08/20/2008 12:43 PM Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU Please respond to The IBM z/VM Operating System Hi If I am using HiperSockets and my MTU size is 1500 and a pass a packet size of 2000 will the data in the packet be truncated since it is bigger than 1500? I thought I read that since HiperSockets is memory transfers that the MTU can not be split like it would over a normal transmission pipe, meaning that for a 3K packet size over a MTU 1500 link would not be split into 2 transmissions. Am I correct in the assumption? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED]
Re: HiperSockets Question
If you wanted to get nitpicky, it would actually need 3 fragments to account for the extra IP header(s). :-) Regards, Miguel Delapaz z/VM TCP/IP Development The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU wrote on 08/20/2008 01:38:48 PM: Yes and no. :-) It won't be truncated, but it will be fragmented. The MTU is the Guardian of the Gate. It prevents the IP layer from sending a packet too large for the underlying interface to handle. So a 3K packet over an MTU 1500 link will result in two fragments, each 1500 bytes. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Question
On Wednesday, 08/20/2008 at 04:44 EDT, Miguel Delapaz/Endicott/[EMAIL PROTECTED] wrote: If you wanted to get nitpicky, it would actually need 3 fragments to account for the extra IP header(s). :-) I *knew* I'd regret not diving into nitty gritty details. I KNEW IT. Mr. Smartypants. Couldn't let it go, could you? We were in the middle of a *point*. It's going to cost you. Not today, not tomorrow, but when you least expect it. No need to look over your shoulder - you won't see me. ;-) Sleep well (but not too deeply). As Khan Noonien Singh said, Revenge is a dish best served cold. Alan da Weasel
Re: HiperSockets Question
At 07:09 PM 8/20/2008 -0400, Alan Altmark wrote: I *knew* I'd regret not diving into nitty gritty details. I KNEW IT. Mr. Smartypants. Couldn't let it go, could you? We were in the middle of a *point*. DTCMPY001S Nobody ever expects the Spanish Inquisition.
Re: HiperSockets Question
Isn't that Sir Alan da Weasel? :) Alan Altmark wrote: Alan da Weasel -- Rich Smrcina VM Assist, Inc. Phone: 414-491-6001 Ans Service: 360-715-2467 rich.smrcina at vmassist.com http://www.linkedin.com/in/richsmrcina Catch the WAVV! http://www.wavv.org WAVV 2009 - Orlando, FL - May 15-19, 2009
Re: HiperSockets Question
On Wednesday, 08/20/2008 at 07:51 EDT, Rich Smrcina [EMAIL PROTECTED] wrote: Isn't that Sir Alan da Weasel? :) Please. Was Zorro, Don Zorro? I think not. Alan da Weasel
Re: HiperSockets Question
He was El Zorro. And Quickdraw McGraw was El Kabong. The 'El' does make a difference. /Tom Kern Alan Altmark wrote: On Wednesday, 08/20/2008 at 07:51 EDT, Rich Smrcina [EMAIL PROTECTED] wrote: Isn't that Sir Alan da Weasel? :) Please. Was Zorro, Don Zorro? I think not. Alan da Weasel
Re: HiperSockets Question
Hi Alan, Not to belabor the subject I just want to make sure I understand. Here is what my issue was. I processed many SQL queries of DB2 tables via DB2 connect running on a z/OS LPAR down to a Linux server running on the same machine but a different z/VM LPAR without any issues. Until, we ran across some queries that were larger than all of the others. These queries failed (timed out). We saw in the DB2 connect trace that the size of the queries was about 2.3K. We knew it was a size issue because when we did a 'SELECT *' on the tables, basically reducing the size it worked. After a lot of effort we found that there was an extra route in the Linux routing table that sent the data bound for one of the z/OS LPARS over to our z/VM where it then was routed by the TCP/IP stack on z/VM to the z/OS LPAR over the HiperSockets LINK. This LINK happened to have a MTU size of 1500. When we removed the bogus route from the Linux routing table causing the SQL queries in question to travel over a HiperSockets LINK with a MTU size of 16348 they were successful. It appears that the difference in the MTU sizes was the problem. My question is does the MTU size mismatch cause the problem? I thought that even if the packets were fragmented they would still get over in full. Sorry for being so wordy but I thought I would present to try to help you help me and to also help others that may be just getting into this as I am. Thanks again for all of your help!! Terry -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Wednesday, August 20, 2008 4:39 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Question On Wednesday, 08/20/2008 at 03:43 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: If I am using HiperSockets and my MTU size is 1500 and a pass a packet size of 2000 will the data in the packet be truncated since it is bigger than 1500? I thought I read that since HiperSockets is memory transfers that the MTU can not be split like it would over a normal transmission pipe, meaning that for a 3K packet size over a MTU 1500 link would not be split into 2 transmissions. Am I correct in the assumption? Yes and no. :-) It won't be truncated, but it will be fragmented. The MTU is the Guardian of the Gate. It prevents the IP layer from sending a packet too large for the underlying interface to handle. So a 3K packet over an MTU 1500 link will result in two fragments, each 1500 bytes. Alan Altmark z/VM Development IBM Endicott
HiperSockets
Hi Is there anything work with sharing the same HiperSockets Network (CHPID) across multiple z/Linux guests down to the different z/OS LPARS? For example, say I have HiperSockts chpid E0 (24 UCBs). I want to define this E0 on both my DEV and VAL LPARS on z/OS. In z/OS can I use the same CHPID for both LPARS and just have different UCBs on the z/Vm z?linux side to address the CHPID? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED]
Re: HiperSockets
You can use the same address triplets in all LPARs. In other words, you can use all 24 addresses in every LPAR. If you have more than one connection to the CHPID from within the *same* LPAR (e.g. VM's TCPIP stack and LINUX guests), then, of course, you must use different triplets there. In z/VM, you can still attach, for example, E000-E002 to TCPIP and then other triplets (E003-E005, E006-E008, ...) to LINUX guests always as virtual E000-E002. That may make LINUX duplication easier. Ivica
Re: HiperSockets and Linux General question
On Sunday, 07/13/2008 at 06:54 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: If I want to set up a HiperSockets link between one of my z/OS LPARS and a z/Linux guest can I just make the necessary changes to my z/OS TCPPARMS on that LPAR with the GATEWAY IP pointing to the z/Linux guest? Then on the z/Linux guest create a route supplying the HiperSockets information associating the IP of the z/OS LPAR? I am assuming that I do not need to define anything for this connection in the TCP/IP PROFILE on the z/VM side is this a correct assumption? As always, when you're trying to decide what must be configured, and how, start with a picture. HiperSockets are LAN segments. Define them to z/OS and Linux as such, just as you would if it were an ethernet. If VM TCP/IP is not in your picture - in the data path, then it need not be updated. It is impossible to give you an absolute answer since I don't know what your network looks like. Alan Altmark z/VM Development IBM Endicott
HiperSockets
Hi Does anyone if there is a restriction that forces you to start the HiperSockets addresses on an even number of the Triplet? Ex: Could I start at EA03 (odd number) for 3 (EA03, EA04, EA05)? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED]
Re: HiperSockets
Thanks Brian! Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Brian Nielsen Sent: Friday, July 11, 2008 1:28 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets There is no current restriction. We use our hipersocket addresses in consecutve groups of 3 without regard to the evenness or oddness of the = real address. Brian Nielsen On Fri, 11 Jul 2008 13:10:58 -0400, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Hi Does anyone if there is a restriction that forces you to start the HiperSockets addresses on an even number of the Triplet? Ex: Could I start at EA03 (odd number) for 3 (EA03, EA04, EA05)? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED]
Re: HiperSockets
Hmm - Thought you could do that, as long as you used virtual addresses and the 1st address was even. Maybe I'm confused with another OS. I work on to many. On Fri, Jul 11, 2008 at 1:51 PM, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Thanks Brian! Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Brian Nielsen Sent: Friday, July 11, 2008 1:28 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets There is no current restriction. We use our hipersocket addresses in consecutve groups of 3 without regard to the evenness or oddness of the = real address. Brian Nielsen On Fri, 11 Jul 2008 13:10:58 -0400, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Hi Does anyone if there is a restriction that forces you to start the HiperSockets addresses on an even number of the Triplet? Ex: Could I start at EA03 (odd number) for 3 (EA03, EA04, EA05)? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -- Mark Pace Mainline Information Systems
Re: HiperSockets
Doesn't a CTC connection require the even/odd address for send/receive? -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology This message (including any attachments) contains confidential information intended for a specific individual and purpose, and its content is protected by law. If you are not the intended recipient, you should delete this message and are hereby notified that any disclosure, copying, or distribution of this transmission, or taking any action based on it, is strictly prohibited. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Mark Pace Sent: Friday, July 11, 2008 1:29 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Hmm - Thought you could do that, as long as you used virtual addresses and the 1st address was even. Maybe I'm confused with another OS. I work on to many. [snip] -- Mark Pace Mainline Information Systems
Re: HiperSockets
Yes, CTC must be an even/odd pair. On Fri, Jul 11, 2008 at 2:30 PM, McKown, John [EMAIL PROTECTED] wrote: Doesn't a CTC connection require the even/odd address for send/receive? -- John McKown Senior Systems Programmer HealthMarkets Keeping the Promise of Affordable Coverage Administrative Services Group Information Technology This message (including any attachments) contains confidential information intended for a specific individual and purpose, and its content is protected by law. If you are not the intended recipient, you should delete this message and are hereby notified that any disclosure, copying, or distribution of this transmission, or taking any action based on it, is strictly prohibited. -- *From:* The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] *On Behalf Of *Mark Pace *Sent:* Friday, July 11, 2008 1:29 PM *To:* IBMVM@LISTSERV.UARK.EDU *Subject:* Re: HiperSockets Hmm - Thought you could do that, as long as you used virtual addresses and the 1st address was even. Maybe I'm confused with another OS. I work on to many. [snip] -- Mark Pace Mainline Information Systems -- Mark Pace Mainline Information Systems
Re: HiperSockets
Current versions of z/Linux don't care either. We have some that start a t even virtual addresses and some at odd virtual addresses. Older versions of Linux did have a restriction, I believe. This topic has been discusse d here before, so if you search you should be able to find them. Brian Nielsen On Fri, 11 Jul 2008 14:28:34 -0400, Mark Pace [EMAIL PROTECTED] wrote: Hmm - Thought you could do that, as long as you used virtual addresses a nd the 1st address was even. Maybe I'm confused with another OS. I work o n to many. On Fri, Jul 11, 2008 at 1:51 PM, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Thanks Brian! Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] O n Behalf Of Brian Nielsen Sent: Friday, July 11, 2008 1:28 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets There is no current restriction. We use our hipersocket addresses in consecutve groups of 3 without regard to the evenness or oddness of th e = real address. Brian Nielsen On Fri, 11 Jul 2008 13:10:58 -0400, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Hi Does anyone if there is a restriction that forces you to start the HiperSockets addresses on an even number of the Triplet? Ex: Could I start at EA03 (odd number) for 3 (EA03, EA04, EA05)? Thank You, Terry Martin Lockheed Martin - Information Technology z/OS z/VM Systems - Performance and Tuning Cell - 443 632-4191 Work - 410 786-0386 [EMAIL PROTECTED] -- Mark Pace Mainline Information Systems
Re: HiperSockets
On Friday, 07/11/2008 at 02:32 EDT, McKown, John [EMAIL PROTECTED] wrote: Doesn't a CTC connection require the even/odd address for send/receive? No. It requires n and n+1. And even that is just an artifact of the device driver. There's nothing special about adjacent subchannels on a CTC. Your IOCP could even configure device number n and n+1 on different CTC chpids. It just doesn't matter... This even thing with OSAs and HiperSockets is the vestigal memory of the original OSA-1s that required the control path to be on an even address. I think that was lifted in the OSA-2 and was definitely gone by the time OSA-Express appeared. It ranks right up there with PORTNAMEs. They have been optional on z/VM and Linux for several years, yet I keep finding NEW configurations that have them coded and people saying they have to match, which they do, but ONLY IF YOU CODE THEM! Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets
Hi Alan, Yes, you are correct: In the HOME address specification I had the z/OS HiperSockets HOME address below the VIPA HOME address. I do not have the z/OS's IP address defined in the z/VM TCP/IP configuration. Sorry for the confusion and thanks for clearing it up. Terry -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Sunday, June 15, 2008 1:04 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets On Friday, 06/13/2008 at 10:08 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: With the help of IBM support we were able to figure out my problem. It turns out that we have a Source VIPA defined and active on the z/OS LPAR. In the HOME address specifications I had the z/VM HOME address below the VIPA HOME address. When I Pinged VM it wanted to respond back via the VIPA HOME IP address which was different than the HOME IP address that I had defined for the z/OS LPAR in TCPIP in VM. As a simple correction I changed the order of the HOME addresses on the z/OS side putting the VM HOME IP address above the VIPA HOME IP address then the PING was successful. I'm glad you got it figured out, but your post is confusing to me. I don't understand why you have z/VM's HOME address defined in you z/OS config, and vice versa. I think you meant In the HOME address specification I had the z/OS HiperSocket HOME address below the VIPA HOME address. Likewise, you would not define z/OS's IP addresses in the z/VM TCP/IP config. But, yes, SourceVIPA can mess things up if you're not running dynamic routing daemons everywhere. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets
On Friday, 06/13/2008 at 10:08 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: With the help of IBM support we were able to figure out my problem. It turns out that we have a Source VIPA defined and active on the z/OS LPAR. In the HOME address specifications I had the z/VM HOME address below the VIPA HOME address. When I Pinged VM it wanted to respond back via the VIPA HOME IP address which was different than the HOME IP address that I had defined for the z/OS LPAR in TCPIP in VM. As a simple correction I changed the order of the HOME addresses on the z/OS side putting the VM HOME IP address above the VIPA HOME IP address then the PING was successful. I'm glad you got it figured out, but your post is confusing to me. I don't understand why you have z/VM's HOME address defined in you z/OS config, and vice versa. I think you meant In the HOME address specification I had the z/OS HiperSocket HOME address below the VIPA HOME address. Likewise, you would not define z/OS's IP addresses in the z/VM TCP/IP config. But, yes, SourceVIPA can mess things up if you're not running dynamic routing daemons everywhere. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets
On Thursday, 06/12/2008 at 06:13 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: I will try to get this information. I am working off of a DUMB terminal for this proof of concept so getting screen shots and such is not happening. Screen shots are not required. Just transcribe the NETSTAT output here and do a crude ASCII art drawing. The purpose is to (a) See what you are TRYING to build, and (2) see what you DID build. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets
Hi With the help of IBM support we were able to figure out my problem. It turns out that we have a Source VIPA defined and active on the z/OS LPAR. In the HOME address specifications I had the z/VM HOME address below the VIPA HOME address. When I Pinged VM it wanted to respond back via the VIPA HOME IP address which was different than the HOME IP address that I had defined for the z/OS LPAR in TCPIP in VM. As a simple correction I changed the order of the HOME addresses on the z/OS side putting the VM HOME IP address above the VIPA HOME IP address then the PING was successful. Thanks, Terry -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Thursday, June 12, 2008 5:25 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets On Thursday, 06/12/2008 at 04:26 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: I have set up a HiprSockets connection from z/VM 5.3 to one of my z/OS LPARS. I can PING the z/OS LPAR but when I try to PINGthe z/VM LPAR from the z/OS LPAR we time out. Is there anything I might be missing? We are using the OBEYFILE to bring the configuration in dynamically. A network problem with no details and no diagram with addresses and subnet masks. If you can provide that, along with the output of NETSTAT HOME GATE from VM and MVS, then we should be able to ferret out the problem. Alan Altmark z/VM Development IBM Endicott
HiperSockets
Hi I have set up a HiprSockets connection from z/VM 5.3 to one of my z/OS LPARS. I can PING the z/OS LPAR but when I try to PING the z/VM LPAR from the z/OS LPAR we time out. Is there anything I might be missing? We are using the OBEYFILE to bring the configuration in dynamically. Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets
Hi Terry. I suspect that your gateway or default statement in the VM TCPIP stack is not correct. Need a network diagram and your TCPIP statements to provide further suggestions. Hans From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Martin, Terry R. (CMS/CTR) (CTR) Sent: June 12, 2008 4:20 PM To: IBMVM@LISTSERV.UARK.EDU Subject: HiperSockets Hi I have set up a HiprSockets connection from z/VM 5.3 to one of my z/OS LPARS. I can PING the z/OS LPAR but when I try to PING the z/VM LPAR from the z/OS LPAR we time out. Is there anything I might be missing? We are using the OBEYFILE to bring the configuration in dynamically. Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets
On Thursday, 06/12/2008 at 04:26 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: I have set up a HiprSockets connection from z/VM 5.3 to one of my z/OS LPARS. I can PING the z/OS LPAR but when I try to PINGthe z/VM LPAR from the z/OS LPAR we time out. Is there anything I might be missing? We are using the OBEYFILE to bring the configuration in dynamically. A network problem with no details and no diagram with addresses and subnet masks. If you can provide that, along with the output of NETSTAT HOME GATE from VM and MVS, then we should be able to ferret out the problem. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets
Hi I will try to get this information. I am working off of a DUMB terminal for this proof of concept so getting screen shots and such is not happening. I will probably need to talk to someone while I am displaying the information at the z/VM and z/OS consoles and rely what I am seeing as we speak. Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Thursday, June 12, 2008 5:25 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets On Thursday, 06/12/2008 at 04:26 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: I have set up a HiprSockets connection from z/VM 5.3 to one of my z/OS LPARS. I can PING the z/OS LPAR but when I try to PINGthe z/VM LPAR from the z/OS LPAR we time out. Is there anything I might be missing? We are using the OBEYFILE to bring the configuration in dynamically. A network problem with no details and no diagram with addresses and subnet masks. If you can provide that, along with the output of NETSTAT HOME GATE from VM and MVS, then we should be able to ferret out the problem. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
On Friday, 06/06/2008 at 05:19 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: I specified the MTU on the Gateway statement! The error was DTCPRS051E ? Line 24: Invalid Packet Size in Gateway Command: 16384 I think the problem is that 16384 exceeds the LARGEENVELOPEPOOLSIZE, which defaults to 8K. No MTU can exceed the L-E-P-Size. Since the GATEWAY entry was then discarded, TCP/IP should have taken the default, which is 4000 for HiperSockets. From the TCP/IP Messages and Codes book: DTCPRS051E Line line: Invalid packet size in GATEWAY command: value Explanation: You specified an invalid max_packet_size in a GATEWAY statement. The size must be a positive integer between 576 (the IPv4 minimum MTU value) and the large envelope size (specified on the LARGEENVELOPEPOOLSIZE statement). If this doesn't jive with what you witnessed, please open a PMR so that we can investigate more fully. I'm not happy with the NETSTAT output you reported. MFS was 24K, and MTU should have been 4000 - subject to documentation errors. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
Thanks Alan, I will verify this and will open a PMR if needed! Thanks, Terry -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Saturday, June 07, 2008 8:45 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup On Friday, 06/06/2008 at 05:19 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: I specified the MTU on the Gateway statement! The error was DTCPRS051E ? Line 24: Invalid Packet Size in Gateway Command: 16384 I think the problem is that 16384 exceeds the LARGEENVELOPEPOOLSIZE, which defaults to 8K. No MTU can exceed the L-E-P-Size. Since the GATEWAY entry was then discarded, TCP/IP should have taken the default, which is 4000 for HiperSockets. From the TCP/IP Messages and Codes book: DTCPRS051E Line line: Invalid packet size in GATEWAY command: value Explanation: You specified an invalid max_packet_size in a GATEWAY statement. The size must be a positive integer between 576 (the IPv4 minimum MTU value) and the large envelope size (specified on the LARGEENVELOPEPOOLSIZE statement). If this doesn't jive with what you witnessed, please open a PMR so that we can investigate more fully. I'm not happy with the NETSTAT output you reported. MFS was 24K, and MTU should have been 4000 - subject to documentation errors. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
Terry, What statement did you specify the MTU on - Gateway or Link? And what was the error you got? Peggy Williams z/VM - TCP/IP Development Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] To .hhs.gov IBMVM@LISTSERV.UARK.EDU Sent by: The IBM cc z/VM Operating SystemSubject [EMAIL PROTECTED] Re: HiperSockets Setup ARK.EDU 06/06/2008 09:42 AM Please respond to The IBM z/VM Operating System [EMAIL PROTECTED] ARK.EDU Yes, Thanks Alan. In the TCPIP PROFILE I did specify 16K but I received an error so I changed it to 1500 which I new would get around the error. When I did the NETSTAT displays I saw a MFS of 4096 and MTU of 32768. So I guess the question is why did I get the error in the TCPIP PROFILE specifying 16K? Am I missing something? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Friday, June 06, 2008 12:13 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup On Thursday, 06/05/2008 at 05:14 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Basically once I got the filename correct and added the PASSWORD for the TCPMAINT 198 disk I was able to get the OBEYFILE command to work. I did receive an error on my first good try. It was balking about the MTU size being invalid. I was using 16384. I changed it to 1500 and tried the OBEYFILE command again and it worked. I noticed on the NETSTAT displays that the MTU actually was 32768 with a Frame Size of 4096. I used a HIPERSOCKETS device that was GENed with CHPARM=40. So basically it did not use the MTU size that I had defined in the PROFILE TCPIP. CHPARM=40 is a MFS of 24K, not 4096. The MTU = MFS - 8K, so the MTU is 16K. (Doc'd in the IOCP book.) Be sure everyone using the same HiperSocket chpid has the same MTU. If you don't, a PING will work, but a large data transfer won't. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
Yes, Thanks Alan. In the TCPIP PROFILE I did specify 16K but I received an error so I changed it to 1500 which I new would get around the error. When I did the NETSTAT displays I saw a MFS of 4096 and MTU of 32768. So I guess the question is why did I get the error in the TCPIP PROFILE specifying 16K? Am I missing something? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Friday, June 06, 2008 12:13 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup On Thursday, 06/05/2008 at 05:14 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Basically once I got the filename correct and added the PASSWORD for the TCPMAINT 198 disk I was able to get the OBEYFILE command to work. I did receive an error on my first good try. It was balking about the MTU size being invalid. I was using 16384. I changed it to 1500 and tried the OBEYFILE command again and it worked. I noticed on the NETSTAT displays that the MTU actually was 32768 with a Frame Size of 4096. I used a HIPERSOCKETS device that was GENed with CHPARM=40. So basically it did not use the MTU size that I had defined in the PROFILE TCPIP. CHPARM=40 is a MFS of 24K, not 4096. The MTU = MFS - 8K, so the MTU is 16K. (Doc'd in the IOCP book.) Be sure everyone using the same HiperSocket chpid has the same MTU. If you don't, a PING will work, but a large data transfer won't. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
Hi Peggy, I specified the MTU on the Gateway statement! The error was DTCPRS051E - Line 24: Invalid Packet Size in Gateway Command: 16384 Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Peggy Williams Sent: Friday, June 06, 2008 10:13 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup Terry, What statement did you specify the MTU on - Gateway or Link? And what was the error you got? Peggy Williams z/VM - TCP/IP Development Inactive hide details for Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 06/06/2008 09:42 AM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject Re: HiperSockets Setup Yes, Thanks Alan. In the TCPIP PROFILE I did specify 16K but I received an error so I changed it to 1500 which I new would get around the error. When I did the NETSTAT displays I saw a MFS of 4096 and MTU of 32768. So I guess the question is why did I get the error in the TCPIP PROFILE specifying 16K? Am I missing something? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Friday, June 06, 2008 12:13 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup On Thursday, 06/05/2008 at 05:14 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Basically once I got the filename correct and added the PASSWORD for the TCPMAINT 198 disk I was able to get the OBEYFILE command to work. I did receive an error on my first good try. It was balking about the MTU size being invalid. I was using 16384. I changed it to 1500 and tried the OBEYFILE command again and it worked. I noticed on the NETSTAT displays that the MTU actually was 32768 with a Frame Size of 4096. I used a HIPERSOCKETS device that was GENed with CHPARM=40. So basically it did not use the MTU size that I had defined in the PROFILE TCPIP. CHPARM=40 is a MFS of 24K, not 4096. The MTU = MFS - 8K, so the MTU is 16K. (Doc'd in the IOCP book.) Be sure everyone using the same HiperSocket chpid has the same MTU. If you don't, a PING will work, but a large data transfer won't. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
Terry, The message DTCOBE005E Unable to read file 'Filename Filetype filemode' means that it cannot read your configuration file for some reason. Do you have the correct filename, filetype and filemode? Peggy Williams 607-429-4063 z/VM - TCP/IP Development Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] To .hhs.gov IBMVM@LISTSERV.UARK.EDU Sent by: The IBM cc z/VM Operating SystemSubject [EMAIL PROTECTED] HiperSockets Setup ARK.EDU 06/04/2008 04:44 PM Please respond to The IBM z/VM Operating System [EMAIL PROTECTED] ARK.EDU Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets Setup
And don't forget the minidisk-read-password like I always do. Bobby Bauer Center for Information Technology National Institutes of Health Bethesda, MD 20892-5628 301-594-7474 From: Peggy Williams [mailto:[EMAIL PROTECTED] Sent: Thursday, June 05, 2008 9:15 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup Terry, The message DTCOBE005E Unable to read file 'Filename Filetype filemode' means that it cannot read your configuration file for some reason. Do you have the correct filename, filetype and filemode? Peggy Williams 607-429-4063 z/VM - TCP/IP Development Inactive hide details for Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 06/04/2008 04:44 PM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject HiperSockets Setup Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets Setup
Hi I will reproduce the error today and send you want I see. The text of the error does not really say a whole lot. I probably have some kind of syntax error but it is not so obvious. Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Ackerman Sent: Wednesday, June 04, 2008 11:04 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup On Wed, 4 Jun 2008 17:44:05 -0500, Mike Walter [EMAIL PROTECTED] w= rote: Terry, Looking in the z/VM 530 TCP/IP Messages and Codes manual, there's no DTCOBE0055E message listed. OBVIOUSLY, it must be self-documenting! That typed with disapproval of self-documenting messages dripping fro= m the keyboard. In fact there's only 1 (count 'em, ONE) numbered OBEYFIL= E message, and three unnumbered messages. DTCOBE0055E must be newer than the -02 manual I'm looking in. Could you please copy/paste the specific message you are seeing? I am setting up HIPERSOCKETS for the first time and I am dealing with th= e syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am tryi= ng to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Interesting little game of telephone here. DTCOBE005E - DTCOBE0055E. B= ut I could find neither one in LookAt. z/VM TCP/IP has numerous undocumented messages. Every time we complain we= get the same response which would you rather have us do, document messages or write = new code. Open an incident with IBM. And include the full message text. Maybe they = will get tired of answering the same questions over and over and document the messages. -OR- Go read the restrictions of what can and cannot be changed with OBEYFILE = in TCP/IP Planning and Customization. Some things cannot be changed, and must NOT b= e in the file supplied to OBEYFILE. In particular, it cannot handle a full PROFILE TCPI= P. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: HiperSockets Setup
Hi Peggy, I just left you a voice message. If you could give me a call back, it would be great! I will update the thread as we talk through the issue! Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Peggy Williams Sent: Thursday, June 05, 2008 9:15 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup Terry, The message DTCOBE005E Unable to read file 'Filename Filetype filemode' means that it cannot read your configuration file for some reason. Do you have the correct filename, filetype and filemode? Peggy Williams 607-429-4063 z/VM - TCP/IP Development Inactive hide details for Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 06/04/2008 04:44 PM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject HiperSockets Setup Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets Setup
Thanks Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Bauer, Bobby (NIH/CIT) [E] Sent: Thursday, June 05, 2008 9:20 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup And don't forget the minidisk-read-password like I always do. Bobby Bauer Center for Information Technology National Institutes of Health Bethesda, MD 20892-5628 301-594-7474 From: Peggy Williams [mailto:[EMAIL PROTECTED] Sent: Thursday, June 05, 2008 9:15 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup Terry, The message DTCOBE005E Unable to read file 'Filename Filetype filemode' means that it cannot read your configuration file for some reason. Do you have the correct filename, filetype and filemode? Peggy Williams 607-429-4063 z/VM - TCP/IP Development Inactive hide details for Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 06/04/2008 04:44 PM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject HiperSockets Setup Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets Setup
On Thursday, 06/05/2008 at 09:20 EDT, Bauer, Bobby (NIH/CIT) [E] [EMAIL PROTECTED] wrote: And don?t forget the minidisk-read-password like I always do. (Pt! Hey, buddy! Over here. Wanna buy a security manager? No more worries about minidisk passwords.) Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
You mean like RACF? It's planned. Budgets/Approval/Papperwork! Bobby Bauer Center for Information Technology National Institutes of Health Bethesda, MD 20892-5628 301-594-7474 -Original Message- From: Alan Altmark [mailto:[EMAIL PROTECTED] Sent: Thursday, June 05, 2008 9:51 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup On Thursday, 06/05/2008 at 09:20 EDT, Bauer, Bobby (NIH/CIT) [E] [EMAIL PROTECTED] wrote: And don?t forget the minidisk-read-password like I always do. (Pt! Hey, buddy! Over here. Wanna buy a security manager? No more worries about minidisk passwords.) Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
On Wednesday, 06/04/2008 at 06:44 EDT, Mike Walter [EMAIL PROTECTED] wrote: Looking in the z/VM 530 TCP/IP Messages and Codes manual, there's no DTCOBE0055E message listed. OBVIOUSLY, it must be self-documenting! That typed with disapproval of self-documenting messages dripping from the keyboard. In fact there's only 1 (count 'em, ONE) numbered OBEYFILE message, and three unnumbered messages. DTCOBE0055E must be newer than the -02 manual I'm looking in. obeyfile x tcpip VM TCP/IP Obeyfile DTCOBE005E Unable to read file 'X TCPIP A' TPMAINT at GDLVMK4(00012); T=0.01/0.01 10:00:30 That would be far better as DTC0BE005E File 'X TCPIP A' not found TPMAINT at GDLVMK4(00028); ..but it is what it is. C'est la vie. If you call us about something and there is a related message which is not documented that is preventing you from discovering the error, we will add it to the book. But please don't call us just to tell us that some message is missing. Chuckie knows where you live and what kind of car you drive, if ya know whaddi mean. If you want to know what to put into an OBEY file, try using ifconfig. ifconfig -show hsi0 10.1.2.3 mask 255.255.255.0 mtu 1500 hipers b000 will show you the statements that need to be in the obey file. (Without the -show it will actually do the obeyfile.) It can be very handy when you are using virtual networks since it will also show you the DEFINEs and COUPLEs. This doesn't help with getting past OBEYFILE file access errors, of course, but it helps with content. Alan Altmark z/VM Development IBM Endicott
Re: HiperSockets Setup
Gets me every time. OBEYFILE fn ft fm (readpassword for the minidisk the file is on. On Thu, Jun 5, 2008 at 9:19 AM, Bauer, Bobby (NIH/CIT) [E] [EMAIL PROTECTED] wrote: And don't forget the minidisk-read-password like I always do. Bobby Bauer Center for Information Technology National Institutes of Health Bethesda, MD 20892-5628 301-594-7474 -- *From:* Peggy Williams [mailto:[EMAIL PROTECTED] *Sent:* Thursday, June 05, 2008 9:15 AM *To:* IBMVM@LISTSERV.UARK.EDU *Subject:* Re: HiperSockets Setup Terry, The message DTCOBE005E Unable to read file '*Filename Filetype filemode*' means that it cannot read your configuration file for some reason. Do you have the correct filename, filetype and filemode? Peggy Williams 607-429-4063 z/VM - TCP/IP Development [image: Inactive hide details for Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]]Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] *Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED]* Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 06/04/2008 04:44 PM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject HiperSockets Setup Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? *Thanks.. Terry* *Terry Martin* * Lockheed Martin - CITIC* * z/OS Performance and Tuning (410) 786-0386 - Office* * (443) 632-4191 - Cell* * [EMAIL PROTECTED] -- Mark Pace Mainline Information Systems
Re: HiperSockets Setup
On Thursday, 06/05/2008 at 05:14 EDT, Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] wrote: Basically once I got the filename correct and added the PASSWORD for the TCPMAINT 198 disk I was able to get the OBEYFILE command to work. I did receive an error on my first good try. It was balking about the MTU size being invalid. I was using 16384. I changed it to 1500 and tried the OBEYFILE command again and it worked. I noticed on the NETSTAT displays that the MTU actually was 32768 with a Frame Size of 4096. I used a HIPERSOCKETS device that was GENed with CHPARM=40. So basically it did not use the MTU size that I had defined in the PROFILE TCPIP. CHPARM=40 is a MFS of 24K, not 4096. The MTU = MFS - 8K, so the MTU is 16K. (Doc'd in the IOCP book.) Be sure everyone using the same HiperSocket chpid has the same MTU. If you don't, a PING will work, but a large data transfer won't. Alan Altmark z/VM Development IBM Endicott
HiperSockets Setup
Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets Setup
Terry, Looking in the z/VM 530 TCP/IP Messages and Codes manual, there's no DTCOBE0055E message listed. OBVIOUSLY, it must be self-documenting! That typed with disapproval of self-documenting messages dripping from the keyboard. In fact there's only 1 (count 'em, ONE) numbered OBEYFILE message, and three unnumbered messages. DTCOBE0055E must be newer than the -02 manual I'm looking in. Could you please copy/paste the specific message you are seeing? Mike Walter Hewitt Associates Any opinions expressed herein are mine alone and do not necessarily represent the opinions or policies of Hewitt Associates. Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 06/04/2008 03:44 PM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject HiperSockets Setup Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message, including any attachments. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. All messages sent to and from this e-mail address may be monitored as permitted by applicable law and regulations to ensure compliance with our internal policies and to protect our business. E-mails are not secure and cannot be guaranteed to be error free as they can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by e-mail.
Re: HiperSockets Setup
Make sure you use the 2007 version of the Hipersockets Redbook. I was never able to get it to work with the 2002 edition. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Martin, Terry R. (CMS/CTR) (CTR) Sent: Wednesday, June 04, 2008 4:44 PM To: IBMVM@LISTSERV.UARK.EDU Subject: HiperSockets Setup Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets Setup
Hi Yes, I have the 2007 version! Any Idea what the error message means? Thanks, Terry From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Kreiter, Chuck Sent: Wednesday, June 04, 2008 9:29 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup Make sure you use the 2007 version of the Hipersockets Redbook. I was never able to get it to work with the 2002 edition. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Martin, Terry R. (CMS/CTR) (CTR) Sent: Wednesday, June 04, 2008 4:44 PM To: IBMVM@LISTSERV.UARK.EDU Subject: HiperSockets Setup Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED]
Re: HiperSockets Setup
All you have provided so far is the message ID, which is not in current doc.. It would be most helpful if you could paste the complete message so that we might have a little more to work with. Mike Walter Hewitt Associates - Original Message - From: Martin, Terry R. (CMS/CTR) (CTR) [EMAIL PROTECTED] Sent: 06/04/2008 10:10 PM AST To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup Hi Yes, I have the 2007 version! Any Idea what the error message means? Thanks, Terry From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Kreiter, Chuck Sent: Wednesday, June 04, 2008 9:29 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: HiperSockets Setup Make sure you use the 2007 version of the Hipersockets Redbook. I was never able to get it to work with the 2002 edition. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Martin, Terry R. (CMS/CTR) (CTR) Sent: Wednesday, June 04, 2008 4:44 PM To: IBMVM@LISTSERV.UARK.EDU Subject: HiperSockets Setup Hi I am setting up HIPERSOCKETS for the first time and I am dealing with the syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am trying to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Thanks.. Terry Terry Martin Lockheed Martin - CITIC z/OS Performance and Tuning (410) 786-0386 - Office (443) 632-4191 - Cell [EMAIL PROTECTED] The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, or if this message has been addressed to you in error, please immediately alert the sender by reply e-mail and then delete this message, including any attachments. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. All messages sent to and from this e-mail address may be monitored as permitted by applicable law and regulations to ensure compliance with our internal policies and to protect our business. E-mails are not secure and cannot be guaranteed to be error free as they can be intercepted, amended, lost or destroyed, or contain viruses. You are deemed to have accepted these risks if you communicate with us by e-mail.
Re: HiperSockets Setup
On Wed, 4 Jun 2008 17:44:05 -0500, Mike Walter [EMAIL PROTECTED] w rote: Terry, Looking in the z/VM 530 TCP/IP Messages and Codes manual, there's no DTCOBE0055E message listed. OBVIOUSLY, it must be self-documenting! That typed with disapproval of self-documenting messages dripping fro m the keyboard. In fact there's only 1 (count 'em, ONE) numbered OBEYFIL E message, and three unnumbered messages. DTCOBE0055E must be newer than the -02 manual I'm looking in. Could you please copy/paste the specific message you are seeing? I am setting up HIPERSOCKETS for the first time and I am dealing with th e syntax of the TCPIP PROFILE for the first time. I am using the HIPERSOCKETS Implementation Guide Red Book to walk through this. I made the changes to the PROFILE TCPIP and the SYSTEM DTCPARMS file. I am tryi ng to do the OBEYFILE PROFILE TCPIP and I am receiving a DTCOBE005E error message. I assume it is some kind of syntax error in the PROFILE but I cannot find the description of the error message. Anyone know where I might find it? Interesting little game of telephone here. DTCOBE005E - DTCOBE0055E. B ut I could find neither one in LookAt. z/VM TCP/IP has numerous undocumented messages. Every time we complain we get the same response which would you rather have us do, document messages or write new code. Open an incident with IBM. And include the full message text. Maybe they will get tired of answering the same questions over and over and document the messages. -OR- Go read the restrictions of what can and cannot be changed with OBEYFILE in TCP/IP Planning and Customization. Some things cannot be changed, and must NOT b e in the file supplied to OBEYFILE. In particular, it cannot handle a full PROFILE TCPI P. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: Hipersockets - xposted to VM-L IBM-Main
Sorry for the late reply, I've been on vacation. (some of you may remember those). I do not see where you would specify a port name on z/OS. I've tried to remove the port names from the z/VM definitions, but thus far it has not made a difference. On Tue, May 13, 2008 at 3:32 AM, Alan Altmark [EMAIL PROTECTED] wrote: On Monday, 05/12/2008 at 01:53 EDT, Stephen Frazier [EMAIL PROTECTED] wrote: My recollection is that the port name must be the same everywhere or absent everywhere. Try removing your port names and see if that works. Port names are meaningful to z/OS, not Linux or z/VM. It is ok if some are using port names and others are not. If you are using port names, then all users must have the same port name. First one in wins. Alan Altmark z/VM Development IBM Endicott -- Mark Pace Mainline Information Systems
Re: Hipersockets - xposted to VM-L IBM-Main
Maybe you are not even using port name on z/OS. I don't know much about z/OS, so I cannot help you find it, but even if you are not using port name in z/OS, you have to be sure that nothing else comes up before the z/OS guest, connects to the hipersocket and changes port name from nothing into something.(1) Did you check your z/VSE stack for a port name? (2) Did you have a chance to reset the hipersocket CHPID (vary off the CHPID from all LPARs *at the same time*, then vary it on to all)? I remember that on z800 hardware port names could differ in one character out of eight (but one only), and microcode check would pass. Eg, port names HSOA1 and HSOB1 were OK, but not HSOB2. Later on I found out that this is in fact how it worked. This applied both to hipersockets and OSA's. From z990 onwards, this slack was removed and from then on all port names must be identical or not used at all. So, if you don't really need port names (and z/VM allows you to not use them from release 4.4 onwards), this exercise of removing them is well worth your while. And if you do (1) and (2) above, you will at least be sure that port name is *not* your problem. Then, we'll dig deeper. Ivica On Tue, May 27, 2008 at 10:28 PM, Mark Pace [EMAIL PROTECTED] wrote: Sorry for the late reply, I've been on vacation. (some of you may remember those). I do not see where you would specify a port name on z/OS. I've tried to remove the port names from the z/VM definitions, but thus far it has not made a difference. On Tue, May 13, 2008 at 3:32 AM, Alan Altmark [EMAIL PROTECTED] wrote: On Monday, 05/12/2008 at 01:53 EDT, Stephen Frazier [EMAIL PROTECTED] wrote: My recollection is that the port name must be the same everywhere or absent everywhere. Try removing your port names and see if that works. Port names are meaningful to z/OS, not Linux or z/VM. It is ok if some are using port names and others are not. If you are using port names, then all users must have the same port name. First one in wins. Alan Altmark z/VM Development IBM Endicott -- Mark Pace Mainline Information Systems
Re: Hipersockets
On Tuesday, 05/27/2008 at 11:35 EDT, Ivica Brodaric [EMAIL PROTECTED] wrote: Maybe you are not even using port name on z/OS. I don't know much about z/OS, so I cannot help you find it, but even if you are not using port name in z/OS, you have to be sure that nothing else comes up before the z/OS guest, connects to the hipersocket and changes port name from nothing into something. A mismatched port name on an OSA creates an initialization error. In any case, HiperSockets do not have port names. On z/OS they are addressed solely by chpid (DEVICE IUTIQDxx). My apologies for creating confusion; I was addressing only the incorrect assertions about port names rather than focusing on the problem at hand. It's been a while and I've forgotten the details, but HiperSocket communication requires: 1. Both LPARs reference the same HiperSocket chpid. Note that z/OS does not allow specification of a HiperSocket chpid that is being used for dynamic XCF. Verify: Get a packet trace to see that packets are being place on and received from the HiperSocket. 2. Both IP stacks have correct local routes. That is, the same subnet and subnet mask with no gateway specification. Verify: Look at the routing table. Make sure the gateway for the HiperSocket subnet is 0.0.0.0. And make sure you don't have overlapping routes. 3. Both stacks are using the same MTU and that MTU is consistent with the MFS value in the chpid definition. Verify: It shouldn't cause a problem with PING, but will result in lost packets for any frame that exceeds the MTU of the receiver. If you want to do all that, fine, but I'd suggest first bringing up both images as z/VM guests and connect them to a HiperSocket Guest LAN instead of real HiperSockets. Be sure to use the CHPID parameter on the NICDEF for a z/OS guest to ensure that the correct virtual chpid is chosen, matching your real HiperSocket chpid number. Compare the routing tables. They should be the same, virtual or real, as far as the HiperSocket interface is concerned. By the way, I'm not cross-posting to IBM-MAIN. Let's try to consolidate discussion in just one place (here). Alan Altmark z/VM Development IBM Endicott
Re: Hipersockets
Alan, A mismatched port name on an OSA creates an initialization error. Correct, but with a caveat explained in the other thread (yes for z990 onwards, before z990 one character could mismatch. I know it sounds weird, but that's the way it worked) In any case, HiperSockets do not have port names. Ummm, no. They can be specified for HiperSocket connection in DEVICE statement in VM and in DEFINE LINK statement in VSE. And after peeking into some z/OS manuals, I found that even z/OS is not oblivious to port names. From z/OS V1R9.0 Communications Server IP Configuration Guide: quote Therefore, there are two types of HiperSockets devices: - DYNAMICXCF HiperSockets device or interface (TRLE IUTIQDIO and an MPC group of subchannel devices). The PORTNAME will be IUTIQDxx, where xx = the IQD CHPID that VTAM(R) uses (for example, IUTIQDFD when using IQD CHPID x'FD'). - A user-defined HiperSockets device or interface (TRLE IUTIQDxx and an MPC group of subchannel devices). The PORTNAME is not applicable for this TRLE. In both cases, the TRLE is dynamically built by VTAM. end quote I presume that we have a user-defined HiperSocket here, but in the response to Q LAN DETAILS command for the virtual LAN that Mark defined to test the connection to the VM stack via virtual HiperSocket there is a following line: Adapter Owner: ZOS19NIC: 0724 Name: *IUTIQDFF* That Name: at the end is port name (not device name). If the portname was not used by z/OS, it should've said Name: UNASSIGNED. So which one is it? DYNAMICXCF or user-defined? And why did it work anyway? Maybe virtual HiperSocket is more lax towards the port names than real HiperSocket, because Mark says that the connection over the virtual HiperSocket worked. I also do not want to overly emphasize this port name thing, but I still think that it is worth while clearing. I think Mark said that he copied z/OS from the LPAR to a guest on VM, changed the home IP address and attached the real HiperSocket device trio to z/OS guest as the same addresses that z/OS expects. So I don't think that we will get any further by doing that again, except that this new copy of z/OS will be unmodified. Maybe still worth a try... Your point 1 seems to be in contradiction with the quote above, but I'll believe you have your reasons. Your point about MFS and MTU is great. I can't see nothing wrong in NETSTAT HOME and NETSTAT GATE responses provided in the other thread, but I noticed there that packet size is 57344 (56K) for IUTLNK1 link. How does that work? Does z/OS automatically adjust the MTU of the interface when the device is activated? This would indicate that CHPID is defined with OS=C0 (64K MFS). Mark, when you defined a virtual HiperSocket, you used default MFS of 16K as I see in QUERY LAN response. Try with MFS 64K operand in DEFINE LAN. Just to remove any difference between the real and virtual HiperSocket... Ivica Brodaric
Re: Hipersockets - xposted to VM-L IBM-Main
Hi All, Sorry to change direction here, but has anyone run into the following error while trying to get FCP to work between z/VM 5.3 and RedHat Linux 4.6. We are connecting to a IBM DS8100. We present the FCP device to the Linux guest and it sees it. But they receive the error when they try to do their thing. [EMAIL PROTECTED] etc]# more zfcp.conf 0.0.5500 0x01 0x5005076306088385 0x00 0x40204000 0.0.5700 0x02 0x5005076306188385 0x00 0x40204000 Here are the lines from the /var/log/messages file from the linux guest: May 23 17:37:23 e49l124d sshd(pam_unix)[1873]: session opened for user root by (uid=0) May 23 17:38:26 e49l124d kernel: scsi1 : zfcp May 23 17:38:26 e49l124d kernel: zfcp: error: Point-to-point fibrechannel configuration detected at adapter 0.0.5500 unsupported, shutting down adapter May 23 17:38:34 e49l124d kernel: scsi2 : zfcp May 23 17:38:34 e49l124d kernel: zfcp: error: Point-to-point fibrechannel configuration detected at adapter 0.0.5700 unsupported, shutting down adapter May 23 17:52:29 e49l124d kernel: scsi3 : zfcp May 23 17:52:29 e49l124d kernel: zfcp: error: Point-to-point fibrechannel configuration detected at adapter 0.0.5500 unsupported, shutting down adapter May 23 18:04:19 e49l124d kernel: zfcp: adapter 0.0.5700: device gone May 23 18:04:19 e49l124d kernel: zfcp: adapter 0.0.5500: device gone May 23 18:08:11 e49l124d kernel: zfcp: adapter 0.0.5700: operational again May 23 18:08:11 e49l124d kernel: zfcp: error: Point-to-point fibrechannel configuration detected at adapter 0.0.5700 unsupported, shutting down adapter May 23 18:08:19 e49l124d kernel: zfcp: adapter 0.0.5500: operational again May 23 18:08:19 e49l124d kernel: zfcp: adapter erp failed on adapter 0.0.5500 May 23 18:08:20 e49l124d kernel: zfcp: error: Point-to-point fibrechannel configuration detected at adapter 0.0.5500 unsupported, shutting down adapter May 23 18:17:58 e49l124d kernel: scsi4 : zfcp May 23 18:17:58 e49l124d kernel: zfcp: adapter erp failed on adapter 0.0.5500 May 23 18:30:21 e49l124d kernel: zfcp: error: Point-to-point fibrechannel configuration detected at adapter 0.0.5500 unsupported, shutting down adapter Thanks, Terry From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Mark Pace Sent: Wednesday, April 23, 2008 10:41 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: Hipersockets - xposted to VM-L IBM-Main Does it make a difference that Hipersockets are real devices dedicated to the z/OS guest and the OSA connection is a VSWITCH? The OSA connection on the VSWITCH works, the real Hipersockets do not. On Wed, Apr 23, 2008 at 12:08 AM, Alan Altmark [EMAIL PROTECTED] wrote: On Tuesday, 04/22/2008 at 03:21 EDT, Mark Pace [EMAIL PROTECTED] wrote: But the Hipersockets are a different issue. I attached 3 address, (even-odd-even) to the guest as the same addresses used on the original z/OS, (As an aside, the addresses aren't important to z/OS. He only cares about the chpid.) EZZ4313I INITIALIZATION COMPLETE FOR DEVICE IUTIQDFF but I can't ping any other address on the Hipersocket network. And nothing can ping it. All the messages from netstat,dev look exactly like the one on the z/OS that works, except of course, for the all zeros in the counts. I agree with Mark Wheeler that your VSWITCH is of the non-routing flavor. (Another problem with z/OS not supporting layer 2 OSA...send your cards and letters to them.) By setting the VSWITCH to PRIROUTER, be aware that you cannot, then, share the OSA with some other host who also believes they deserve PRIROUTER. Alan Altmark z/VM Development IBM Endicott -- Mark Pace Mainline Information Systems
Re: Hipersockets - xposted to VM-L IBM-Main
On Monday, 05/12/2008 at 01:53 EDT, Stephen Frazier [EMAIL PROTECTED] wrote: My recollection is that the port name must be the same everywhere or absent everywhere. Try removing your port names and see if that works. Port names are meaningful to z/OS, not Linux or z/VM. It is ok if some are using port names and others are not. If you are using port names, then all users must have the same port name. First one in wins. Alan Altmark z/VM Development IBM Endicott
