Discussion:
using REXX to spawn a Java program
(too old to reply)
Steve Austin
2018-10-08 08:12:13 UTC
Permalink
I have some questions regarding the above.

I'm specifying an environment variable of "_BPX_SHAREAS=MUST". Before each spawn of the Java program I have a "syscalls('on') and after a "syscalls("off"). I'm noticing a significant delay at the "syscalls('off'); it can be as much as 2 minutes. If I remove the "syscalls('off')" I don't get the delay.


1) syscalls('off') is removing the unix environment, but why the delay?

2) By removing the "syscalls('off')" I'm retaining the unix environment, but is the Java JVM also retained for reuse? I'm guessing not, but it would be nice if it were.

3) The documentation I've found so far is pretty general. Is there documentation somewhere that describes the mechanics in detail?

Thanks

Steve
--
This e-mail message has been scanned and cleared by Google Message Security
and the UNICOM Global security systems. This message is for the named
person's use only. If you receive this message in error, please delete it
and notify the sender.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Joe Monk
2018-10-08 12:29:10 UTC
Permalink
This kinda gives you a clue:

"syscalls('ON') ensures that the SYSCALL host command environment is
available in your REXX environment. If the call detects that SYSCALL is not
available in your environment, it dynamically adds it."

"Performance characteristics for dynamically added host commands are not as
good as for host commands that are included in the initial environment:
Every time a command is directed to the SYSCALL host command environment,
the TSO/E REXX support loads the module for the SYSCALL host command."

Page 13

https://www-01.ibm.com/servers/resourcelink/svc00100.nsf/pages/zOSV2R3SA232283/$file/bpxb600_v2r3.pdf

So every time you turn off syscalls, and subsequently turn it back on, you
are dynamically adding host support... this is probably the cause of your
performance issues.

Joe
Post by Steve Austin
I have some questions regarding the above.
I'm specifying an environment variable of "_BPX_SHAREAS=MUST". Before each
spawn of the Java program I have a "syscalls('on') and after a
"syscalls("off"). I'm noticing a significant delay at the "syscalls('off');
it can be as much as 2 minutes. If I remove the "syscalls('off')" I don't
get the delay.
1) syscalls('off') is removing the unix environment, but why the delay?
2) By removing the "syscalls('off')" I'm retaining the unix
environment, but is the Java JVM also retained for reuse? I'm guessing not,
but it would be nice if it were.
3) The documentation I've found so far is pretty general. Is there
documentation somewhere that describes the mechanics in detail?
Thanks
Steve
--
This e-mail message has been scanned and cleared by Google Message Security
and the UNICOM Global security systems. This message is for the named
person's use only. If you receive this message in error, please delete it
and notify the sender.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Paul Gilmartin
2018-10-08 13:57:32 UTC
Permalink
Post by Steve Austin
I have some questions regarding the above.
I'm specifying an environment variable of "_BPX_SHAREAS=MUST". Before each spawn of the Java program I have a "syscalls('on') and after a "syscalls("off"). I'm noticing a significant delay at the "syscalls('off'); it can be as much as 2 minutes. If I remove the "syscalls('off')" I don't get the delay.
"Doctor, it hurts when I do this. "

Why do you use SYSCALLS("OFF")? It's not recommended:

http://www2.marist.edu/htbin/wlvtype?MVS-OE.35369

Date: Mon, 2 Jun 2003 20:08:28 -0400
From: William Schoen <***@us.ibm.com>
Subject: Re: SYSCALLS() Query?

I do not recommend using sycalls 'OFF'. It may clean up more than you want
yet not everything from syscalls 'ON'.

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Steve Austin
2018-10-08 14:47:01 UTC
Permalink
Thanks for the link I found it useful and interesting. As to why I used syscall('off'), the doc I worked from said it was not usually necessary, rather than not recommended, so it seemed tidier to do so. I'll certainly remove it. Incidentally the problem only manifested when I switched from _BPX_SHAREAS=NO to _BPX_SHAREAS=MUST/REUSE/YES. From the description it sounds like REUSE is what I should use.

-----Original Message-----
From: IBM Mainframe Discussion List [mailto:IBM-***@LISTSERV.UA.EDU] On Behalf Of Paul Gilmartin
Sent: Monday, October 8, 2018 2:57 PM
To: IBM-***@LISTSERV.UA.EDU
Subject: Re: using REXX to spawn a Java program
Post by Steve Austin
I have some questions regarding the above.
I'm specifying an environment variable of "_BPX_SHAREAS=MUST". Before each spawn of the Java program I have a "syscalls('on') and after a "syscalls("off"). I'm noticing a significant delay at the "syscalls('off'); it can be as much as 2 minutes. If I remove the "syscalls('off')" I don't get the delay.
"Doctor, it hurts when I do this. "

Why do you use SYSCALLS("OFF")? It's not recommended:

http://www2.marist.edu/htbin/wlvtype?MVS-OE.35369

Date: Mon, 2 Jun 2003 20:08:28 -0400
From: William Schoen <***@us.ibm.com>
Subject: Re: SYSCALLS() Query?

I do not recommend using sycalls 'OFF'. It may clean up more than you want
yet not everything from syscalls 'ON'.

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions, send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
--
This e-mail message has been scanned and cleared by Google Message Security
and the UNICOM Global security systems. This message is for the named
person's use only. If you receive this message in error, please delete it
and notify the sender.

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to ***@listserv.ua.edu with the message: INFO IBM-MAIN
Loading...