Is there a file limit size of 2GB in JDK 1.5

[zigzagdna]

I am on HP UNIX 11.2 which is 64 bit operating system. I am using JDK
1.5 and using Java's File i/o to write my trace information to a file.
I noticed that even though my program is running, it stopped writing
any more information once trace file became 2GB.

It cannot be HP UNIX limitation which is 64 bit, also file systems
have been set to allow files larger than 2GB. I am surprise that
java's i/o is 32 bit (yes java's version is most likely 32 bit).

 

[Roedy Green]

It cannot be HP UNIX limitation which is 64 bit, also file systems
have been set to allow files larger than 2GB. I am surprise that
java's i/o is 32 bit (yes java's version is most likely 32 bit).

hmm. Java itself uses longs (64 bits) for file offsets. There is some
native code that java.io. uses to hook into the OS. It might be
designed for an older 32-bit HP API. Or it may be that partitions are
old, 32-bit even under the new 64-bit os. You may have to reformat
them to free up 64 bit power.

Try an experiment with C to see if it too has this limit on that
partition.

Consider that Java 1.5 has been declared obsolete for quite some time
now. JDK 1.7 is out. Try your code with 1.7 and see how it works.

 

[Tom]

I am on HP UNIX 11.2 which is 64 bit operating system. I am using JDK
1.5 and using Java's File i/o to write my trace information to a file. I
noticed that even though my program is running, it stopped writing any
more information once trace file became 2GB.

It cannot be HP UNIX limitation which is 64 bit, also file systems have
been set to allow files larger than 2GB. I am surprise that java's i/o
is 32 bit (yes java's version is most likely 32 bit).

Is there a process limit in force? What does ulimit -a say ?

 

[Arne Vajhøj]

hmm. Java itself uses longs (64 bits) for file offsets. There is some
native code that java.io. uses to hook into the OS. It might be
designed for an older 32-bit HP API. Or it may be that partitions are
old, 32-bit even under the new 64-bit os. You may have to reformat
them to free up 64 bit power.

Try an experiment with C to see if it too has this limit on that
partition.

Consider that Java 1.5 has been declared obsolete for quite some time
now. JDK 1.7 is out. Try your code with 1.7 and see how it works.

HP-UX != Windows

HP-UX Java 1.5 is not obsolete. I don't even think HP has announced an
EOL date. HP-UX Java 1.4.2 will be obsolete May 1st 2012.

http://bizsupport2.austin.hp.com/bc/docs/support/SupportManual/c02690385/c02690385.pdf

HP-UX Java 1.7 is not available yet.

Arne

 

[supercalifragilisticexpialadiamaticonormalizeringe]

HP-UX != Windows

HP-UX Java 1.5 is not obsolete. I don't even think HP has announced an
EOL date. HP-UX Java 1.4.2 will be obsolete May 1st 2012.

That just means that HP-UX is behind the times, not that Java 1.5 isn't.

 

[Arne Vajhøj]

That just means that HP-UX is behind the times, not that Java 1.5 isn't.

It means that Java 1.5 is not obsolete and fully supported by the
vendor on HP-UX.

Which is the relevant facts.

Arne

 

[Lew]

It means that Java 1.5 is not obsolete and fully supported by the
vendor on HP-UX.

Which is the relevant facts.

Oracle has declared only their own version of Java 5 obsolete. Other vendors are free to do otherwise. Arne is, as usual, correct.

 

[thoolen]

On 22/09/2011 9:13 AM, "Jane Doe", an obvious murphy sock, wrote:
NaN> Newsgroups: comp.lang.java.programmer

NaN> You ever find Derbysire (Seamus) being solid on Fact?

Who is "Derbyshire (Seamus)", murphy? There is nobody in this newsgroup
using that alias.

NaN> I will save you the keystrokes.

What does your classic erroneous presupposition that you're saving
people keystrokes by adding one more off-topic article for them to mark
read or hit control-K on have to do with Java, murphy?

NaN> No.

What does your denial have to do with Java, murphy?

"I had 'volunteered (years back) to support those who do endeavor
to provide free Free Usenet access, support those who offered
subscription based Free Usenet access, nothing more than
cooperation expected in return for what has been many
thousands of hours of work. I note most of those I joined with
are either deceased, severely disabled, or plain ole' MIA..
now it is my Time. ...

You just read my last. ...

For those who think they see me in future times I can only wish
you severe Tinnitus in your dreams. For those who know me
well (eMail, whatever) and see me, know I will be smiling also.
It is to you I say "adieu mein frenz and adios .. grazie' [hugs]
for all the Good Times! May you and yours always bear well
with all Life brings you".

/0ut"
--murphy

http://www.uffnet.com/kookkamp/goodbye.htm

And people wonder why I call them Famous Last Words.

P.S. You forgot to include a copy of your famous paranoid "Lits o' Haet"
in your post, murphy. Still suffering from memory problems, murphy?

 

[supercalifragilisticexpialadiamaticonormalizeringe]

supercalifragilisticexpialadiamaticonormalizeringelimatisticantations

It means that Java 1.5 is not obsolete and fully supported by the
vendor on HP-UX.

Which just means that that vendor is behind the times. Of course, since
it's a proprietary Unix vendor, that's hardly surprising.

 

[Jane Doe]

It means that Java 1.5 is not obsolete and fully supported by the
vendor on HP-UX.

Which is the relevant facts.

Arne

You ever find Derbysire (Seamus) being solid on Fact?
I will save you the keystrokes.

No.

 

[Lew]

You ever find Derbysire (Seamus) being solid on Fact?
I will save you the keystrokes.

No.

Please stay on topic here, and take your personal vendettas elsewhere.

 

[Jane Doe]

Please stay on topic here, and take your personal vendettas elsewhere.

I can appreciate your point.
However from what I have witnessed first hand you (particularly)
are not mature enough in the Usenet sense to apply that same advice
to the posts Derbyshire leads you "up the garden path" with.
Even when you *know* his posts contains factual mistakes you
grant the post credibility by an exchange which implies the
disagreement is up for discussion.
You ignore the other choice.
Tell him to **** off. Or ignore the post.
That's "maturity" in dealing with the Usenet child.

Like, how many times is it required to *show* Derbyshire
is running multiple names in cljp?
Programmers are that dense?
Or just you?

 

[Lew]

I can appreciate your point.
However from what I have witnessed first hand you (particularly)
are not mature enough in the Usenet sense to apply that same advice
to the posts Derbyshire leads you "up the garden path" with.

Plonk.

 

[Arne Vajhøj]

Which just means that that vendor is behind the times. Of course, since
it's a proprietary Unix vendor, that's hardly surprising.

No.

It is completely standard that server vendors support
a Java version for more years than SUN/Oracle supports the free
version.

Arne

 

[supercalifragilisticexpialadiamaticonormalizeringe]

Yes.

It is completely standard that server vendors support
a Java version for more years than SUN/Oracle supports the free
version.

It may be completely standard, but it is also the case that Java 1.5 is
outdated and lacking in numerous features and other enhancements that
are found in Java 6 and Java 7. HP should really migrate to Java 7 in
preference to supporting Java 1.5.

But ultimately it's their (inferred) lack of support for Java 7, rather
than their support for 1.5, that is at issue.

 

[thoolen]

On 22/09/2011 5:40 PM, "Jane Doe", an obvious murphy sock, wrote:
NaN> Newsgroups: comp.lang.java.programmer

NaN> I can appreciate your point.

What does your classic unsubstantiated and erroneous claim have to do
with Java, murphy? You have never shown any capacity for "appreciating"
any "point" involving your messages being off-topic or in other ways
unwelcome or violating of social norms, murphy, as evidenced by the
slimy trail of gigabytes of off-topic paranoid blather, punctuated by
occasional flamewars, that you have left all over Usenet like some giant
radiation-mutated snail.

NaN> However from what I have witnessed first hand you (particularly)
NaN> are not mature enough in the Usenet sense to apply that same advice
NaN> to the posts Derbyshire leads you "up the garden path" with.

Who is "Derbyshire", murphy? There is nobody in this newsgroup using
that alias.

NaN> Even when you *know* his posts contains factual mistakes you
NaN> grant the post credibility by an exchange which implies the
NaN> disagreement is up for discussion.

How ironic, seeing as your posts contain numerous factual mistakes,
expressed and implied.

NaN> You ignore the other choice.
NaN> Tell him to **** off. Or ignore the post.

How ironic, coming from someone who seems to be incapable himself of
ignoring the same posts.

NaN> That's "maturity" in dealing with the Usenet child.

Who is "the Usenet child", murphy? There is nobody in this newsgroup
using that alias.

NaN> Like, how many times is it required to *show* Derbyshire
NaN> is running multiple names in cljp?

Who is "Derbyshire", murphy? There is nobody in this newsgroup using
that alias.

NaN> Programmers are that dense?

It isn't a sign of density when someone disbelieves or disregards you,
murphy. In fact, it is a sign of density when someone listens to you and
takes what you said seriously, murphy.

NaN> Or just you?

What does your question of Bloch have to do with Java, murphy?

"I had 'volunteered (years back) to support those who do endeavor
to provide free Free Usenet access, support those who offered
subscription based Free Usenet access, nothing more than
cooperation expected in return for what has been many
thousands of hours of work. I note most of those I joined with
are either deceased, severely disabled, or plain ole' MIA..
now it is my Time. ...

You just read my last. ...

For those who think they see me in future times I can only wish
you severe Tinnitus in your dreams. For those who know me
well (eMail, whatever) and see me, know I will be smiling also.
It is to you I say "adieu mein frenz and adios .. grazie' [hugs]
for all the Good Times! May you and yours always bear well
with all Life brings you".

/0ut"
--murphy

http://www.uffnet.com/kookkamp/goodbye.htm

And people wonder why I call them Famous Last Words.

P.S. You forgot to include a copy of your famous paranoid "Lits o' Haet"
in your post, murphy. Still suffering from memory problems, murphy?

 

[Arne Vajhøj]

It may be completely standard, but it is also the case that Java 1.5 is
outdated and lacking in numerous features and other enhancements that
are found in Java 6 and Java 7. HP should really migrate to Java 7 in
preference to supporting Java 1.5.

I would assume HP knows more about their customers needs than you.

But ultimately it's their (inferred) lack of support for Java 7, rather
than their support for 1.5, that is at issue.

The issue you raised was about Java 1.5.

Arne

 

[supercalifragilisticexpialadiamaticonormalizeringe]

I would assume HP knows more about their customers needs than you.

Unless their customers are very unusual (some sort of nouveau-Amish that
decided to freeze their technology level at 2002 or something?), then
it's pretty much a given that they want (at least the option of) the
latest and greatest.

This is much as if HP were shipping computers running Windows 98 instead
of Vista and/or Windows 7. It doesn't matter if they keep supporting it
-- Microsoft isn't anymore, and is the software's ultimate source, and
furthermore the world has largely moved on, making compatibility an
issue; and Windows 98 is unstable and full of bugs and security holes
that a) are by now very widely known and b) are fixed in newer versions
of Windows.

I doubt you'd think it at all sensible for HP to ship computers running
Windows 98 and refuse to support any newer version of Windows, even if
they provided support for the obsolescent version they did ship. So why
do you think differently when it's obsolescent versions of the JRE and JDK?

The issue you raised was about Java 1.5.

Obviously not, since I just explicitly said that it's lack of support
for Java 7 that's the (greater?) sin.

P.S. What got it into your head to drag this old, settled argument back
up out of the group's archive months later and suddenly start disputing
things again? Did you get bored and decide we needed a good
old-fashioned flamewar around here to make things entertaining or
something? Because I'm sure you're aware that the group's expressed
purpose is neither flamage nor your personal entertainment but
good-faith discussion of Java programming...also, you chose the wrong
person to try to pick a fight with. If you want a descent into serious
incivility, without being seen as the one to descend to that level first
yourself, I won't oblige. I'd suggest you take Twisted's name in vain,
but even he has apparently mellowed out; I doubt you'd get more of a
rise out of him these days than a polite, but firm, "none of the nasty
things are true" type of response. (But please don't even try it. I
don't think we would benefit by the group being disrupted with a
flamewar or a serious attempt to provoke one.)

 

[spk]

things are true" type of response. (But please don't even try it. I
don't think we would benefit by the group being disrupted with a
flamewar or a serious attempt to provoke one.)

... seriously _you_ "we"[sic] threatening a "flamewar"[sic]
How many hands have you got - sock free?

Do not begin what you cannot finish, Paul.
One name per newsgroup.

Flames are old bait to you.
http://tinyurl.com/8xe5bpy
Display any maturity or even a different stroke.
Whatever, your shtick grows very old very fast.

 

[Roedy Green]

Java has used longs 64 bit addressing for files since day 1. Some
older platforms had their own limits, but they did not come from Java.

that gives you files up to 9,223,372,036,854,775,807 bytes.
9 exabytes. Not likely to be a problem any time soon.
--
Roedy Green Canadian Mind Products
http://mindprod.com
Capitalism has spurred the competition that makes CPUs faster and
faster each year, but the focus on money makes software manufacturers
do some peculiar things like deliberately leaving bugs and deficiencies
in the software so they can soak the customers for upgrades later.
Whether software is easy to use, or never loses data, when the company
has a near monopoly, is almost irrelevant to profits, and therefore
ignored. The manufacturer focuses on cheap gimicks like dancing paper
clips to dazzle naive first-time buyers. The needs of existing
experienced users are almost irrelevant. I see software rental as the
best remedy.