RE: tmpnam() not unique

Thanks for the information Scott.

I think I'll just add the job number to the filename.

Unfortunately, there is going to be numerous programs where this 'fix' is required !



-----Original Message-----
From: RPG400-L [mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of Scott Klement
Sent: 14 October 2014 15:03
To: RPG programming on the IBM i (AS/400 and iSeries)
Subject: Re: tmpnam() not unique

This link is for Z/OS.

I know that tmpnam() is not considered unique on FreeBSD (an open Unix similar to Linux that I've worked on), and they recommend calling
mkstemp() instead.

When I was looking for something similar on IBM i, I got the impression that tmpnam() worked nicely on IBM i and it wasn't the same problem. So it's disappointing to hear that it isn't, in fact, unique.

Usually (at least on other platforms) the problem is the time between calling tmpnam() and the time it takes you to create the file in the /tmp directory. Since it's possible that another job could call
tmpnam() in between your job calling tmpnam() and actually creating the unique file in /tmp. The unique filename is checked against the /tmp diretory, but if you haven't created it yet, it could duplicate.

But, I thought IBM had somehow worked around that on IBM i. Sigh.


On 10/14/2014 8:53 AM, McGovern, Sean wrote:

No, I've not seen anything that states the filename is based on timestamp, but it wouldn't surprise me if it is.

I've just looked at this link...

http://publib.boulder.ibm.com/infocenter/zvm/v6r1/index.jsp?topic=/com
.ibm.zvm.v610.edclv/tmpnam.htm

...but can't tell whether the filename should be unique across the LPAR.

"Produces a valid file name that is not the same as the name of any existing file..."
"The tmpnam() function produces a different name each time it is called within a module..."
"When the calling application is an OpenExtensions C program, tmpnam() returns a unique file name..."

...and don't know how to write something to test it !

We had some issues which I've investigated, and my conclusion is that tmpnam() has returned the same filename in two separate jobs.


-----Original Message-----
From: RPG400-L [mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of
Scott Klement
Sent: 14 October 2014 14:45
To: RPG programming on the IBM i (AS/400 and iSeries)
Subject: Re: tmpnam() not unique

That would be very disappointing... IBM says that tmpnam() will generate a unique filename.

Does it say somewhere that this is based on the timestamp?? (which
certainly isn't unique!)

On 10/14/2014 8:19 AM, McGovern, Sean wrote:

I've believe I have encountered an issue with my use of tmpnam().

If 2 jobs use tmpnam() at the same timestamp moment, I *think* the same filename is returned (though I don't know how to test this scenario).

I haven't coded to handle this possibility, and end up with two jobs trying to create the same temporary IFS file.

I was thinking that a solution could be just prefix/suffix the job number to the filename returned by tmpnam().

Does that seem like a feasible solution ?

--
This is the RPG programming on the IBM i (AS/400 and iSeries)
(RPG400-L) mailing list To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx Before posting, please take a
moment to review the archives at http://archive.midrange.com/rpg400-l.

--
This is the RPG programming on the IBM i (AS/400 and iSeries) (RPG400-L) mailing list To post a message email: RPG400-L@xxxxxxxxxxxx To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives at http://archive.midrange.com/rpg400-l.