[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: REGION=0M and LSQA


Barbara,

>
I don't think anyone says that software is perfect. WHat I personally find 
sad is that just about every Java programmer doesn't know and doesn't care 
what the underlying architecture does and hence is incapable of even 
analyzing anything. 
We currently have a problem where a java application has its main thread 
silently terminating, leaving all other threads active but doing nothing. 
The application is incapable of determining *why* that thread terminates 
and is playing a guessing game. 'It must be too few storage.' Well, even 
increasing the storage caused the hang ('this is not a hang').

And did I mention that that address space has region=0M and even gets 
region=0M (as in everything available above and below - nothing is used 
below, the usage above is negligible).

One thing that rubs me (and probably others) the wrong way is the 
implication that a storage leak (or whatever the current mnemonic is) is 
something to be tolerated and encouraged by specifying region=0M (your 
wording "... I am not sure why customers would want to run Java without 
REGION=0M, as without it , it is possible to cause a JVM abend. ..." came 
across like this to me). 

I am with Ed on this: Give me a good unconditional getmain so I can set a 
slip trap in abend878/80a/whatever and get good doc. And I have an 
abending application that doesn't hang around probably exposing other bugs 
(like why the hell doesn't the main thread take all other threads with it 
as they can't do anything anyway.)
>

1. You have really misunderstood me, the reason for REGION=0M is not to 
tolerate a leak, it is to ensure that the JVM has enough storage for 
normal operation and to avoid "avoidable" 878 abends etc. I wish I had 
never mentioned "leaks" now.

2. Why not raise a PMR if you have a problem?

3. Many of the problems that I see are the result of an abend, e.g. 878-10
 

------------------------------------------------------------------------------------------------------------------
Mike Poil
Java z/OS Level 3 Service
IBM United Kingdom Limited, Hursley Park, Winchester SO21 2JN
Internal: 246824  External: +44 (0)1962 816824 
Java debugging: http://www.ibm.com/developerworks/java/jdk/diagnosis/
------------------------------------------------------------------------------------------------------------------






Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU






----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to listserv@xxxxxxxxxxx with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html


This mailing list archive is a service of Copilot Consulting.