Re: Postscript printing in RSCS
>I've studied the Exit Customization book, the TCP/IP Solutions for VM/ESA
>Redbook ("Enterprise Printing") and I'm still confused how to tie the LPR
>link & LPRXFORM CONFIG file together. The Redbook describes setting up an
>LPRXFORM CONFIG and shows several examples of FORM, FORMSUB & PREFIX
>values to be used with PS printers. However it says there should be an
>entry for each form in the LPDXMANY CONFIG file. I don't have any LPD's
>defined, just LPR's. Do I just set up a config file for LPRXFORM and point
>to it in the EPARM of the LPR link (C=LPRXFORM)?
I am assuming you are sending the z/OS files via an NJE link; so you
would not want or need to setup an LPD link. The LPRXFORM exits
are used to setup form blocks during RSCS initialization, which are
selected via the spool file FORMS. They are not used by the LPRXPSE
exit, so you would not want to code them in a C=LPRXFORM EPARM
statement.
>
>Then, in the files I send to RSCS from z/OS, they have plain text and
>Postscript programs included in one file, (i.e. JESMSG, JESJCL, JESLOG,
>(all text) followed by a postscript program to print a special invoice
>form). Does LPRXPSE send a program that will handle the text as well as
>the imbeded postscript program?
Do each of these files have separate NJE or dataset headers? Can you
change the form on each file? LPRXPSE doesn't handle an imbeded
postscript file in the middle of a plain text file.
Best Regards,
Les Geer
IBM z/VM and Linux Development
Re: Postscript printing in RSCS
I've studied the Exit Customization
book, the TCP/IP Solutions for VM/ESA Redbook ("Enterprise Printing")
and I'm still confused how to tie the LPR link & LPRXFORM CONFIG file
together. The Redbook describes setting up an LPRXFORM CONFIG and shows
several examples of FORM, FORMSUB & PREFIX values to be used with PS
printers. However it says there should be an entry for each form in the
LPDXMANY CONFIG file. I don't have any LPD's defined, just LPR's. Do I
just set up a config file for LPRXFORM and point to it in the EPARM of
the LPR link (C=LPRXFORM)?
Then, in the files I send to RSCS from
z/OS, they have plain text and Postscript programs included in one file,
(i.e. JESMSG, JESJCL, JESLOG, (all text) followed by a postscript program
to print a special invoice form). Does LPRXPSE send a program that
will handle the text as well as the imbeded postscript program?
--
Bob Henry * Mgr., Rochester IT * SunGard Higher Education * 3000 Ridge
Rd. E., Rochester, NY 14622
Tel 585-339-2366 * Mobile 585-330-5240 * Fax 610-578-3014 * [EMAIL PROTECTED]
* www.sungardhe.com
CONFIDENTIALITY: This email (including any attachments) may contain confidential,
proprietary and privileged information, and unauthorized disclosure or
use is prohibited. If you received this email in error, please notify the
sender and delete this email from your system. Thank you
"Les Geer (607-429-3580)"
<[EMAIL PROTECTED]>
Sent by: The IBM z/VM Operating System
03/16/2006 03:11 PM
Please respond to
The IBM z/VM Operating System
To
IBMVM@LISTSERV.UARK.EDU
cc
Subject
Postscript printing in RSCS
>I'm routing z/OS output to LPR printers in RSCS
and haven't had any
>success getting postscript to work. The printer is postscript capable
(an
>HP 8150). I'm using exit LPRXPSE defined as:
>PARM ROC815WP EXIT=LPRXPSE ITO=0 SEC=NO #
> HOST=149.24.51.28
#
> PRINTER=raw #
> USER=YES
>
>What else do I need to do to initiate the postscript function? I'm
>researching this in the various Redbooks, RSCS manuals, etc. but hope
>someone can give me a quick jump-start.
>
>Also, when I route normal text output (i.e. JESMSGLG, JESJCL, etc.)
along
>
>with the postscript output will the printer handle that ok?
>
LPRXPSE sends print jobs to postscript capable printers. The print
job
can itself be a postscript program, or it can be a plain text file
which LPRXPSE wraps a postscript program around it.
LPRXPSE attempts to tell if the file is a postscript program,
however it might not determine this early enough in processing to
terminate the postscript program it sent, and hence the printer
might continue to run wrapper portscript program instead of
terminating.
The best way to deal with this is to tell LPRXPSE what type of file
this is via the forms.
P+ASCII tells LPRXPSE the file is a streaming postscript file in ASCII.
Best Regards,
Les Geer
IBM z/VM and Linux Development
Re: the subject line
And if that program could insure that the message actually make sense … I had better not go there or most posts will be rejected. Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED]On Behalf Of Mike Walter Sent: Friday, March 17, 2006 1:27 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: the subject line Now if he could programmatically confirm that the subject and the note text match , and make sense for subsequent searching... :-) Mike Walter Hewitt Associates The opinions expressed herein are mine alone, not my employer's.
Re: the subject line
I'd suggest taking advantage of list personalization options - in this case, the "SubjectHdr" setting. Via the web interface (http://listserv.uark.edu/scripts/wa.exe?REPORT&z=3) check the options available under "Mail Header Style". Selecting "LISTSERV-style, with list name in subject" [SUBJECTHDR] will cause the string "[IBMVM]" to be prefixed to the subject line on all mail from the list to your subscription. I use this, and like it quite a lot -- it makes sorting and filtering in-box contents via Thunderbird somewhat easier. Kind regards, -dan. Steve Gentry wrote: Hello. Could you guys (and gals) please include subject lines? I've made it a habit, for a long time now, not to open any email that doesn't have a subject line. Mr. Moderator, is there a way you can require that the poster have a subject line and if not, don't post the message? Thanks, Steve G. *snip!*
Re: the subject line
ok, Thanks for checking into it. Steve G. Mike Walter <[EMAIL PROTECTED]> Sent by: The IBM z/VM Operating System 03/17/2006 04:27 PM Please respond to The IBM z/VM Operating System To: IBMVM@LISTSERV.UARK.EDU cc: Subject: Re: the subject line Steve, Good habit! But in Mike Kinnear's case he was actually trying to send a listserve "command" and sent it to the list instead of the listserve. Even then I usually include the command as the Subject just so *I* have a clue later. No doubt you've noticed the minor flood of listserve commands (including SIGNOFF and LOGOFF) being sent to the discussion list instead of the listserve address itself - probably few of us enter listserve commands on a regular basis so it's easy to explain the high tide. But the idea that a "Subject:" be required or the note be returned with an explanatory message is a good one. No doubt in Dan's copious free time he'll look right into that. Now if he could programmatically confirm that the subject and the note text match , and make sense for subsequent searching... :-) Mike Walter Hewitt Associates The opinions expressed herein are mine alone, not my employer's. "Steve Gentry" <[EMAIL PROTECTED]> Sent by: "The IBM z/VM Operating System" 03/17/2006 01:14 PM Please respond to "The IBM z/VM Operating System" To IBMVM@LISTSERV.UARK.EDU cc Subject Re: the subject line Hello. Could you guys (and gals) please include subject lines? I've made it a habit, for a long time now, not to open any email that doesn't have a subject line. Mr. Moderator, is there a way you can require that the poster have a subject line and if not, don't post the message? Thanks, Steve G. "Kinnear, Mike" <[EMAIL PROTECTED]> Sent by: The IBM z/VM Operating System 03/17/2006 02:09 PM Please respond to The IBM z/VM Operating System To: IBMVM@LISTSERV.UARK.EDU cc: Subject: thanks. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Duane Weaver Sent: Friday, March 17, 2006 1:10 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Folks Send your SET command to [EMAIL PROTECTED] not to the list at IBMVM@LISTSERV.UARK.EDU At 02:05 PM 3/17/2006, you wrote: "SET IBMVM DIGEST" DISCLAIMER: This message and accompanying documents are covered by the Electronic Communications Privacy Act, 18 U.S.C. ?? 2510-2521, and contains information intended for the specified individual(s) only. This information is confidential. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, copying, or the taking of any action based on the contents of this information is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail, and delete the original message. 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.
Re: VM Performance Question
You got me. I keep forgetting that the 2 GB I/O boundry in z/VM 5.1 and below isn't the same as executing code. Neither of these, have ever bitten me. The VM folks have been very good about keeping ahead of the curve, as far as I have been concerned. Tom Duerbusch THD Consulting >>> Dennis.L.O'[EMAIL PROTECTED] 3/17/2006 1:01 PM >>> Correction to item 1: z/VM 5.1 and below can support more than 2G of central storage in the LPAR, as long as you're on zSeries hardware. They just have to move pages below 2G when CP needs to do something with them, such as I/O. Dennis "Do you think it's a good idea, letting an Arab country take over our ports? This is like letting Bill Clinton be the manager of a Hooters." -- Jay Leno -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Tom Duerbusch Sent: Friday, March 17, 2006 09:19 To: IBMVM@LISTSERV.UARK.EDU Subject: Re: VM Performance Question I don't know for sure if it really helps, but I have been saying the same thing for years. Setting expanded storage is a good thing for an actively paging system. Also by reducing the amount of central storage, would cause increased paging. But if you are not actively paging, why have expanded storage. But this is only good for: 1. If you are on z/VM 5.1 or under and have 2 GB or less in the LPAR. If you have more than 2 GB, then you have to configure expanded storage to use the rest, so it is a moot point. 2. You are not actively paging. IMHO, a few pages a second or a spike when something relatively rare happens (a new guest being IPL'ed), CICS being cycled, xedit a million line printout, etc. But when you start actively paging, then, for performance reasons, you do need to start configuring expanded storage. I have no proof, but I doubt that you will see any difference (a performance monitor will), unless you are paging a lot. I always question rules of thumb and other performance tidbits. What conditions do these solve, and am I even close to one of those conditions. There are many things that really only for large or heavy use, shops. Tom Duerbusch THD Consulting >>> [EMAIL PROTECTED] 3/15/2006 4:21 PM >>> The link below is a VM Performance article with the ending "Bottom Lines" recommendations. Has anyone done this and did it in fact improve performance? Bottom Lines |--- ---| | | |For VM/ESA, if you have plenty of storage, little paging, a well tuned MDC, a | |consistent load, and a robust DASD paging configuration, then all real storage| |is most likely the best case. Otherwise, consider configuring some storage as | | expanded storage. A ballpark starting point is 25% of processor storage. You | |must configure anything above 2GB as expanded storage. | |--- ---| |--- ---| | | | For z/VM, you should still configure some storage as expanded. The 25% value | |may still be a good starting point. Most systems do not need more than 2GB of | | expanded storage regardless of the total storage available. Systems with no | | constraint below 2GB, can use less expanded storage. Constraint below 2GB | | often is indicated by significant paging to DASD and a large number of pages | |available above 2GB (as seen by QUERY FRAMES command or your favorite | | performance tool). In that case, you will want to add more expanded storage. | | You may also be able to free-up expanded storage by limiting the amount of | | expanded storage in use by minidisk cache (MDC) via the CP command SET MDC | | XSTORE or with a system configuration statement. | |--- ---| www.vm.ibm.com/perf/tips/storconf.html TIA Jan Canavan
Re: ITO parameterfor LPR link
>"Use of autostart and ITO really is a personal choice. If you have=20 >considerable number of links, you would not want them all active=20 >at a time. This was more important back when RSCS only supported=20 >16M. However, GCS still has 16M limitations. But this would be=20 >for a rather large (100+) links." > >Right now, I have 72 SNA3270P links defined, all STARTed. I plan to >convert all to LPR. There shouldn't be an issue with GCS virtual storage >is this case, right? How much does 1 link consume in terms of virtual >storage? Correct there should not be a storage issue as LPR type links should not consume GCS storage differently than the SNA3270P links. Note this is storage GCS requires for managing tasks, not storage RSCS uses (which is 31-bit storage). Best Regards, Les Geer IBM z/VM and Linux Development
Re: FTP and VMFTP
On Thursday, 03/16/2006 at 05:49 EST, "Hughes, Jim - OIT" <[EMAIL PROTECTED]> wrote: > We are running ZVM 3.1 and TCP/IP 3A0. > > We use VMFTP quite heavily here. We have a modification to VMFTP named > NODELAY. This NODELAY fix changes the original start up command from > "FTP 0" to FTP 0.0.0.0" > > If I issue a regular FTP command with no arguments, I get prompted > straight off to enter the address of the FTP server. No delay at all. > > If I enter the command FTP 0.0.0.0, there is a delay followed by a > message "unknown host:". > > Perhaps I am just slow because after 2 years of ZVM and TCPIP 3A0 > running in production, I noticed tonight all the VMFTP commands have > about a 3-6 second delay before anything inside the VMFTP macro. I > suspect its because FTP is locating the ftp server at 0.0.0.0. > > Have any of you noticed this behavior too? > > If so, how did you get around the delay when looking for 0.0.0.0??? > > Thanks in advance. > > Please make me look stupid and tell me it's a tcpip configuration > option. It would make my day. The excessive delay with "FTP 0" was seen when the name server config in TCPIP DATA was bad or there was a routing loop in the network. The delay was RESOLVERTIMEOUT x RESOLVERRETRIES x the number of bad NSINTERADDR entries. Turn on TRACE RESOLVER in TCPIP DATA and use TRACE option on the FTP command to get a full picture of what's happening. Here's what I see with TRACE (but no resolver trace): 16:26:17 ftp 0.0.0.0 (trace exit 16:26:17 VM TCP/IP FTP Level 520 16:26:17 Translate Table: STANDARD 16:26:17 about to call BeginTcpIp 16:26:17 Connecting to 0.0.0.0, port 21 16:26:17 SysAct 0 21 0 CC -1 16:26:17 ==> Active open to host 0.0.0.0 port 21 from host 0 port 65535 16:26:17 TcpOpen error: unspecified foreign address in active open 16:26:17 Unable to connect to 0.0.0.0 16:26:17 TcpOpen error: unspecified foreign address in active open 16:26:17 SysHalt has been Called 16:26:17 Ready(1); T=0.01/0.02 16:26:17 Alan Altmark z/VM Development IBM Endicott
Re: the subject line
Steve, Good habit! But in Mike Kinnear's case he was actually trying to send a listserve "command" and sent it to the list instead of the listserve. Even then I usually include the command as the Subject just so *I* have a clue later. No doubt you've noticed the minor flood of listserve commands (including SIGNOFF and LOGOFF) being sent to the discussion list instead of the listserve address itself - probably few of us enter listserve commands on a regular basis so it's easy to explain the high tide. But the idea that a "Subject:" be required or the note be returned with an explanatory message is a good one. No doubt in Dan's copious free time he'll look right into that. Now if he could programmatically confirm that the subject and the note text match , and make sense for subsequent searching... :-) Mike Walter Hewitt Associates The opinions expressed herein are mine alone, not my employer's. "Steve Gentry" <[EMAIL PROTECTED]> Sent by: "The IBM z/VM Operating System" 03/17/2006 01:14 PM Please respond to "The IBM z/VM Operating System" To IBMVM@LISTSERV.UARK.EDU cc Subject Re: the subject line Hello. Could you guys (and gals) please include subject lines? I've made it a habit, for a long time now, not to open any email that doesn't have a subject line. Mr. Moderator, is there a way you can require that the poster have a subject line and if not, don't post the message? Thanks, Steve G. "Kinnear, Mike" <[EMAIL PROTECTED]> Sent by: The IBM z/VM Operating System 03/17/2006 02:09 PM Please respond to The IBM z/VM Operating System To: IBMVM@LISTSERV.UARK.EDU cc: Subject: thanks. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Duane Weaver Sent: Friday, March 17, 2006 1:10 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Folks Send your SET command to [EMAIL PROTECTED] not to the list at IBMVM@LISTSERV.UARK.EDU At 02:05 PM 3/17/2006, you wrote: "SET IBMVM DIGEST" DISCLAIMER: This message and accompanying documents are covered by the Electronic Communications Privacy Act, 18 U.S.C. ?? 2510-2521, and contains information intended for the specified individual(s) only. This information is confidential. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, copying, or the taking of any action based on the contents of this information is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail, and delete the original message. 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.
Re: the subject line
Hello. Could you guys (and gals) please include subject lines? I've made it a habit, for a long time now, not to open any email that doesn't have a subject line. Mr. Moderator, is there a way you can require that the poster have a subject line and if not, don't post the message? Thanks, Steve G. "Kinnear, Mike" <[EMAIL PROTECTED]> Sent by: The IBM z/VM Operating System 03/17/2006 02:09 PM Please respond to The IBM z/VM Operating System To: IBMVM@LISTSERV.UARK.EDU cc: Subject: thanks. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Duane Weaver Sent: Friday, March 17, 2006 1:10 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Folks Send your SET command to [EMAIL PROTECTED] not to the list at IBMVM@LISTSERV.UARK.EDU At 02:05 PM 3/17/2006, you wrote: "SET IBMVM DIGEST" DISCLAIMER: This message and accompanying documents are covered by the Electronic Communications Privacy Act, 18 U.S.C. ?? 2510-2521, and contains information intended for the specified individual(s) only. This information is confidential. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, copying, or the taking of any action based on the contents of this information is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail, and delete the original message.
Re: step order for adding links with mproute?
> I've assumed (because the book doesn't say) that it should be: > > 1) Add a new Interface statement to MPROUTE CONFIG > Step 1a) issue SMSG MPROUTE RECONFIG to tell MPRoute to check MPROUTE CONFIG for new interfaces. > 2) Construct an OBEYFILE containing the new Device, Link, Home and (null) > Gateway statements > > 3) Execute the OBEYFILE command against the file > > 4) Start the new link > > 5) Bring up the guest > Regards, Miguel Diaz z/VM TCP/IP Development
Re: ITO parameterfor LPR link
David & Les, Makes sense about special forms and physical presense. As to Les reply that "Use of autostart and ITO really is a personal choice. If you have considerable number of links, you would not want them all active at a time. This was more important back when RSCS only supported 16M. However, GCS still has 16M limitations. But this would be for a rather large (100+) links." Right now, I have 72 SNA3270P links defined, all STARTed. I plan to convert all to LPR. There shouldn't be an issue with GCS virtual storage is this case, right? How much does 1 link consume in terms of virtual storage? Thanks, Mike -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of David Boyes Sent: March 16, 2006 8:10 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: ITO parameterfor LPR link On Wed, Mar 15, 2006 at 10:11:14AM -0500, Horlick, Michael wrote: > Thanks for the explanation. I see for ITO the possibilities are : > > Why would anyone want to have the link inactive? Most common reason I've seen is printers that use expensive special forms, or printers in secure areas that require you to activate them only for specific reasons, and you may have to have some official physically present to certify the printout.
step order for adding links with mproute?
Hi, I've been having trouble with my TCP/IP stacks in z/VM 4.4.0 ever since we enabled the MPROUTE server. Sometimes when I add a new point-to-point link, the MPROUTE server starts announcing itself as the default router for our whole network. I've thought several times that I've figured out how to prevent this from happening, but eventually it gets me again. What is the proper step order to follow in adding a link to the stack? I've assumed (because the book doesn't say) that it should be: 1) Add a new Interface statement to MPROUTE CONFIG 2) Construct an OBEYFILE containing the new Device, Link, Home and (null) Gateway statements 3) Execute the OBEYFILE command against the file 4) Start the new link 5) Bring up the guest If this is not right, what is? Regards, Leslie Turriff VM Systems Programmer Central Missouri State University Room 400 Ward Edwards Building Warrensburg MO 64093 [EMAIL PROTECTED] [EMAIL PROTECTED] 660.543.4285 660.580.0523
[no subject]
thanks. From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Duane WeaverSent: Friday, March 17, 2006 1:10 PMTo: IBMVM@LISTSERV.UARK.EDUSubject: FolksSend your SET command to[EMAIL PROTECTED]not to the list at IBMVM@LISTSERV.UARK.EDUAt 02:05 PM 3/17/2006, you wrote: "SET IBMVM DIGEST" DISCLAIMER:This message and accompanying documents are covered by the Electronic Communications Privacy Act, 18 U.S.C. ?? 2510-2521, and contains information intended for the specified individual(s) only. This information is confidential. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, copying, or the taking of any action based on the contents of this information is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail, and delete the original message.
[no subject]
Folks Send your SET command to [EMAIL PROTECTED] not to the list at IBMVM@LISTSERV.UARK.EDU At 02:05 PM 3/17/2006, you wrote: "SET IBMVM DIGEST"
[no subject]
"SET IBMVM DIGEST" DISCLAIMER:This message and accompanying documents are covered by the Electronic Communications Privacy Act, 18 U.S.C. ?? 2510-2521, and contains information intended for the specified individual(s) only. This information is confidential. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, copying, or the taking of any action based on the contents of this information is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail, and delete the original message.
Re: VM Performance Question
Correction to item 1: z/VM 5.1 and below can support more than 2G of central storage in the LPAR, as long as you're on zSeries hardware. They just have to move pages below 2G when CP needs to do something with them, such as I/O. Dennis "Do you think it's a good idea, letting an Arab country take over our ports? This is like letting Bill Clinton be the manager of a Hooters." -- Jay Leno -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Tom Duerbusch Sent: Friday, March 17, 2006 09:19 To: IBMVM@LISTSERV.UARK.EDU Subject: Re: VM Performance Question I don't know for sure if it really helps, but I have been saying the same thing for years. Setting expanded storage is a good thing for an actively paging system. Also by reducing the amount of central storage, would cause increased paging. But if you are not actively paging, why have expanded storage. But this is only good for: 1. If you are on z/VM 5.1 or under and have 2 GB or less in the LPAR. If you have more than 2 GB, then you have to configure expanded storage to use the rest, so it is a moot point. 2. You are not actively paging. IMHO, a few pages a second or a spike when something relatively rare happens (a new guest being IPL'ed), CICS being cycled, xedit a million line printout, etc. But when you start actively paging, then, for performance reasons, you do need to start configuring expanded storage. I have no proof, but I doubt that you will see any difference (a performance monitor will), unless you are paging a lot. I always question rules of thumb and other performance tidbits. What conditions do these solve, and am I even close to one of those conditions. There are many things that really only for large or heavy use, shops. Tom Duerbusch THD Consulting >>> [EMAIL PROTECTED] 3/15/2006 4:21 PM >>> The link below is a VM Performance article with the ending "Bottom Lines" recommendations. Has anyone done this and did it in fact improve performance? Bottom Lines |--- ---| | | |For VM/ESA, if you have plenty of storage, little paging, a well tuned MDC, a | |consistent load, and a robust DASD paging configuration, then all real storage| |is most likely the best case. Otherwise, consider configuring some storage as | | expanded storage. A ballpark starting point is 25% of processor storage. You | |must configure anything above 2GB as expanded storage. | |--- ---| |--- ---| | | | For z/VM, you should still configure some storage as expanded. The 25% value | |may still be a good starting point. Most systems do not need more than 2GB of | | expanded storage regardless of the total storage available. Systems with no | | constraint below 2GB, can use less expanded storage. Constraint below 2GB | | often is indicated by significant paging to DASD and a large number of pages | |available above 2GB (as seen by QUERY FRAMES command or your favorite | | performance tool). In that case, you will want to add more expanded storage. | | You may also be able to free-up expanded storage by limiting the amount of | | expanded storage in use by minidisk cache (MDC) via the CP command SET MDC | | XSTORE or with a system configuration statement. | |--- ---| www.vm.ibm.com/perf/tips/storconf.html TIA Jan Canavan
Re: DASD I/O performance VM 44 vs. VM 52
Right, they are. But they are one or the other, not both. So $25K a piece for a pair of FICON. And $25K a piece for a pair of FCP. That is, if you need to run both mainframe and scsi attached devices. And then, you have the cost on the dasd side. One dasd subsystem with FICON. Another DASD subsystem with FCP. Or a box that can be LPARed with both FICON and FCP connections. At the decision point, where we replace our processor and dasd, adding FCP was just additional cost in which we wouldn't actually see any benefit. That may change in the future, but in July 2005, it was just additional cost. I'm not knocking FCP. I can see it as being a lower overhead connection. But the cost/benefit just wasn't there, yet. Our older VSE/ESA 2.3 systems wouldn't be able to use it unless we converted all VSE dasd from CKD to FBA. Just not worth the hassle for VSE systems that should be migrated to VSE/ESA 2.7 and above over the next 2 years. Tom Duerbusch THD Consulting >>> [EMAIL PROTECTED] 3/17/2006 12:19 PM >>> That's strange. I thought the FCP and FICON adapters were one in the same with different microcode. The IOCDS setting of TYPE=FCP (?) was what caused the different microcode to be loaded. There are different FICON cards though aren't there? FICON Express etc. I wonder if this was the difference. -Original Message- One other facet to this discussioncost On the z/890, the FICON adapters were $25K each. The FCP adapters were $25K each. Then, you need the adapters on the storage unit. We went to an IBM DS6800. The Business Partner was somewhat pushing the FCP adapters for our growing Linux workload. The DS6800 would be LPARed into a FICON attached side for CKD data, and a FCP side for scsi attached access. I took one look at the cost and couldn't justify, at this time, FCP. We can always add it in the future. And I expect when we replace the z/890, FCP will be just one of those "standard" things. Yep, would still be optional, but everyone orders it.
unsubscribe
Re: DASD I/O performance VM 44 vs. VM 52
That's strange. I thought the FCP and FICON adapters were one in the same with different microcode. The IOCDS setting of TYPE=FCP (?) was what caused the different microcode to be loaded. There are different FICON cards though aren't there? FICON Express etc. I wonder if this was the difference. -Original Message- One other facet to this discussioncost On the z/890, the FICON adapters were $25K each. The FCP adapters were $25K each. Then, you need the adapters on the storage unit. We went to an IBM DS6800. The Business Partner was somewhat pushing the FCP adapters for our growing Linux workload. The DS6800 would be LPARed into a FICON attached side for CKD data, and a FCP side for scsi attached access. I took one look at the cost and couldn't justify, at this time, FCP. We can always add it in the future. And I expect when we replace the z/890, FCP will be just one of those "standard" things. Yep, would still be optional, but everyone orders it.
Re: DASD I/O performance VM 44 vs. VM 52
One other facet to this discussioncost On the z/890, the FICON adapters were $25K each. The FCP adapters were $25K each. Then, you need the adapters on the storage unit. We went to an IBM DS6800. The Business Partner was somewhat pushing the FCP adapters for our growing Linux workload. The DS6800 would be LPARed into a FICON attached side for CKD data, and a FCP side for scsi attached access. I took one look at the cost and couldn't justify, at this time, FCP. We can always add it in the future. And I expect when we replace the z/890, FCP will be just one of those "standard" things. Yep, would still be optional, but everyone orders it. If we were pushing the I/O thru on the Linux side, perhaps running a second IFL, then the cost/performance issue would be moot. But right now, on my system, FCP doesn't resolve any issues. I would like to say that I have better uses for the $100k, but the real issue was the extra cost would sink the prospects for buying the z/890 and DS6800. Tom Duerbusch THD Consulting >>> [EMAIL PROTECTED] 3/16/2006 7:51 AM >>> Rob, IBM has published some numbers for FBA emulation on SCSI for Linux guests: "Linux Disk I/O Alternatives" at http://www.vm.ibm.com/perf/reports/zvm/html/520lxd.html it's clear there's a CPU overhead penalty for FBA emulation; but not clear if the pain would be noticeable. This e-mail, including any attachments, may be confidential, privileged or otherwise legally protected. It is intended only for the addressee. If you received this e-mail in error or from someone who was not authorized to send it to you, do not disseminate, copy or otherwise use this e-mail or its attachments. Please notify the sender immediately by reply e-mail and delete the e-mail from your system. -Original Message- From: VM/ESA and z/VM Discussions [mailto:[EMAIL PROTECTED] On Behalf Of Rob van der Heij Sent: Thursday, March 16, 2006 3:31 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: DASD I/O performance VM 44 vs. VM 52 On 3/15/06, Romanowski, John (OFT) <[EMAIL PROTECTED]> wrote: > On VM 5.2 I'm considering trying FBA emulation on SCSI for Linux guests > to simplify each Linux guest's configuration and save staff time. > > If the additional CPU overhead doesn't hurt much it'll be worth it. I have no numbers from such a configuration (but would be interested to see them) so I cannot tell what the impact of this will be. I normally avoid to express an opinion about things I have not tried and measured myself. All I can tell is that the FBA emulation on SCSI was meant for shops that want their Linux data on SCSI to "simplify configuration and save staff time" ;-) because they have an investment in the "open storage" area. For those shops it would be a PITA to define S/390 I/O as well just to hold the basic VM system. Now if you would say that Open FCP for Linux on z/VM would greatly benefit from virtualizing that support on z/VM, then I could not agree more with you. In addition to connecting to and managing the Open FCP data, I could also see some fun in having a virtual SAN living on VM paging space. And I am sure that this is not the first time our friends in z/VM planning see this idea... Rob -- Rob van der Heij Velocity Software, Inc
Re: VM Performance Question
I don't know for sure if it really helps, but I have been saying the same thing for years. Setting expanded storage is a good thing for an actively paging system. Also by reducing the amount of central storage, would cause increased paging. But if you are not actively paging, why have expanded storage. But this is only good for: 1. If you are on z/VM 5.1 or under and have 2 GB or less in the LPAR. If you have more than 2 GB, then you have to configure expanded storage to use the rest, so it is a moot point. 2. You are not actively paging. IMHO, a few pages a second or a spike when something relatively rare happens (a new guest being IPL'ed), CICS being cycled, xedit a million line printout, etc. But when you start actively paging, then, for performance reasons, you do need to start configuring expanded storage. I have no proof, but I doubt that you will see any difference (a performance monitor will), unless you are paging a lot. I always question rules of thumb and other performance tidbits. What conditions do these solve, and am I even close to one of those conditions. There are many things that really only for large or heavy use, shops. Tom Duerbusch THD Consulting >>> [EMAIL PROTECTED] 3/15/2006 4:21 PM >>> The link below is a VM Performance article with the ending "Bottom Lines" recommendations. Has anyone done this and did it in fact improve performance? Bottom Lines |--| | | |For VM/ESA, if you have plenty of storage, little paging, a well tuned MDC, a | |consistent load, and a robust DASD paging configuration, then all real storage| |is most likely the best case. Otherwise, consider configuring some storage as | | expanded storage. A ballpark starting point is 25% of processor storage. You | |must configure anything above 2GB as expanded storage. | |--| |--| | | | For z/VM, you should still configure some storage as expanded. The 25% value | |may still be a good starting point. Most systems do not need more than 2GB of | | expanded storage regardless of the total storage available. Systems with no | | constraint below 2GB, can use less expanded storage. Constraint below 2GB | | often is indicated by significant paging to DASD and a large number of pages | |available above 2GB (as seen by QUERY FRAMES command or your favorite | | performance tool). In that case, you will want to add more expanded storage. | | You may also be able to free-up expanded storage by limiting the amount of | | expanded storage in use by minidisk cache (MDC) via the CP command SET MDC | | XSTORE or with a system configuration statement. | |--| www.vm.ibm.com/perf/tips/storconf.html TIA Jan Canavan
Re: VM's SMTP server
On Friday, 03/17/2006 at 12:57 GMT, "HARROP, Roy" <[EMAIL PROTECTED]> wrote: > Some time ago, our network people changed the address of their mail > server. > > I duly changed all references that I could find in the VM system to > point to the new address. > > Now VM's SMTP recognises the new IPMAILER address and sends mail > destined for the real Outside World to the correct place (and such mail > gets delivered), it continues sending inter-company mail to the old > external server. > > I've obviously missed a reference - but can't see where. > > I've been through all the 191 disks and SFS directories that are defined > to the SMTP and TCPIP machines (as well as, in desperation) FTPSERVE, > NAMESERV, etc, etc but can find no reference (except one that's > commented out) to the old mailserver. > > Can anyone suggest where I might look now? Please!! SMTP only has one configuration statement for mail relays: IPMAILERADDRESS. Remember that IPMAILERADDRESS is only used when the target hostname cannot be resolved via DNS. Addresses that *can* be resolved are sent directly by SMTP, without regard as to whether or not the target host is actually in your network. When SMTP starts, the server configuration is displayed. Look for "ip mailer address" in the console log. Alan Altmark z/VM Development IBM Endicott
VM's SMTP server
Hi, Some time ago, our network people changed the address of their mail server. I duly changed all references that I could find in the VM system to point to the new address. Now VM's SMTP recognises the new IPMAILER address and sends mail destined for the real Outside World to the correct place (and such mail gets delivered), it continues sending inter-company mail to the old external server. I've obviously missed a reference - but can't see where. I've been through all the 191 disks and SFS directories that are defined to the SMTP and TCPIP machines (as well as, in desperation) FTPSERVE, NAMESERV, etc, etc but can find no reference (except one that's commented out) to the old mailserver. Can anyone suggest where I might look now? Please!! Regards, Roy Harrop ** This email and any files transmitted with it are confidential. If you are not the intended recipient, please notify our Help Desk. Email [EMAIL PROTECTED] immediately. You should not copy or use this email or attachment(s) for any purpose nor disclose their contents to any other person. NATS computer systems may be monitored and communications carried on them recorded, to secure the effective operation of the system and for other lawful purposes.
Re: Trying to ftp a file from operatns
Where are you FTPing to and what directory are you cd into ? If this is a "local" ftp (not to IBM) then the login or cd should have shown you that you have a R/O access to the disk, something like: 230-MAINT logged in; working directory = MAINT 191 (ReadOnly) 230 write access currently unavailable If you are FTPing to IBM, then be sure you are in /s390/toibm/vm ___ James Vincent Systems Engineering Consultant Nationwide Services Co., Technology Infrastructure Engineering Mainframe, z/VM and z/Linux Support One Nationwide Plaza 3-25-02 Columbus OH 43215-2220 U.S.A Voice: (614) 249-5547Fax: (614) 677-7681 mailto:[EMAIL PROTECTED] The IBM z/VM Operating System wrote on 03/17/2006 07:38:32 AM: > IBMVM@LISTSERV.UARK.EDU > > Thanks Jim,I got my "A" commands mixed up. But now I have another problem > When I issue the put I get the following: > > put prb0001.dump0001.d > >>>SITE FIXrecfm 4096 > 500 'SITE FIXRECFM 4096': command not understood. > >>>PORT 66,144,176,244,4,2 > 200 PORT command successful. > >>>STOR prb0001.dump0001 > 553 prb0001.dump0001: Permission denied. (upload) > Command: > > Any clues?? > > thanks > Mace
Re: Trying to ftp a file from operatns
forget it I found I was in the wrong area At IBM thank for the help Larry
Re: Trying to ftp a file from operatns
Thanks Jim,I got my "A" commands mixed up. But now I have another problem When I issue the put I get the following: put prb0001.dump0001.d >>>SITE FIXrecfm 4096 500 'SITE FIXRECFM 4096': command not understood. >>>PORT 66,144,176,244,4,2 200 PORT command successful. >>>STOR prb0001.dump0001 553 prb0001.dump0001: Permission denied. (upload) Command: Any clues?? thanks Mace
Re: Trying to ftp a file from operatns
Are you doing something like: CP LINK TCPMAINT 592 592 RR ACCESS 592 C If you try to "attach" 592, I can believe you would be getting that error. ATTACH is for real devices for CP and ACCESS is for CMS disks. If you post the commands and errors you are getting, that would help us nail it down for you. ___ James Vincent Systems Engineering Consultant Nationwide Services Co., Technology Infrastructure Engineering Mainframe, z/VM and z/Linux Support One Nationwide Plaza 3-25-02 Columbus OH 43215-2220 U.S.A Voice: (614) 249-5547Fax: (614) 677-7681 mailto:[EMAIL PROTECTED] Larry Macioce <[EMAIL PROTECTED] ES.LIQUOR.STA From TE.OH.US> Larry Macioce <[EMAIL PROTECTED] Sent by: The S> IBM z/VM To Operating IBMVM@LISTSERV.UARK.EDU System cc <[EMAIL PROTECTED] RV.UARK.EDU>Subject Trying to ftp a file from operatns 03/17/2006 07:18 AM Please respond to The IBM z/VM Operating System <[EMAIL PROTECTED] RV.UARK.EDU> I know I need to get acces to tcpmaint I have tried to link to the 592 disk and then att to that disk but I'm getting a msg that 592 doesn't exist help thanks Larry
Trying to ftp a file from operatns
I know I need to get acces to tcpmaint I have tried to link to the 592 di sk and then att to that disk but I'm getting a msg that 592 doesn't exist help thanks Larry
