Synchronizing file access between threads in a single program in java is trivial. Java has a built-in synchronized methods to implement critical sections. I'm not sure how well it would scale though as I guess there implemented by calling the pthread API.
On 20/04/2012, at 9:01 PM, "Farley, Peter x23353" <[email protected]> wrote: > The OP originally stated that he had two STC processes that updated the same > z/Unix file at the same time and wanted to prevent this. Are two STC's both > considered z/Unix processes under the same kernel when they are running on > the same LPAR? If so, ISTM that locking and unlocking the file around the > write code will work as a mechanism to prevent simultaneous writes. > > But what about two STC's running on different LPAR's in the same sysplex? > Does z/Unix file locking work across LPAR's, or only within the same LPAR? > > Just curious. > > ISTR that simultaneous writes to a common file are classically solved in most > unix systems by implementing a daemon process that is the only process > allowed to write to the file, and other processes send messages to the daemon > to accomplish the write function, like the syslog daemon for system messages. > > The new question is whether Java on z/Unix provides access to any file > locking mechanism at all, as the OP also just said it was a Java application > and he wants to share the file among tasks of the *same* program. That would > argue that using whatever the Java equivalent of the C flockfile function is > would be the solution to use. > > But can z/Unix Java do that? Or does he have to write a JNI function to be > called from his z/Unix Java code to implement a file lock? Or should he just > implement a third process/task under the same program for the file writes and > have both of the other tasks just send messages to the write process? > > Peter > > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf > Of Walt Farrell > Sent: Friday, April 20, 2012 8:27 AM > To: [email protected] > Subject: Re: USS File Integrity > > On Fri, 20 Apr 2012 11:18:45 +0800, David Crayford <[email protected]> > wrote: > >> Of course, fcntl() can be used to implement byte-range-locking. So in >> theory you could use it to implement >> row-level locking in a dictionary library. ENQ is not that granular. > > ENQ is as granular as the application wants to make it, depending on how > clever the application programmer is at encoding information into the RNAME > the application will use. > > The key point about UNIX files, though, is that all the locking is advisory, > and controlled by the applications that use the file. If they all implement > the same locking mechanism (whatever that may be), the locking will work. If > they don't, it won't work. > -- > > > This message and any attachments are intended only for the use of the > addressee and may contain information that is privileged and confidential. If > the reader of the message is not the intended recipient or an authorized > representative of the intended recipient, you are hereby notified that any > dissemination of this communication is strictly prohibited. If you have > received this communication in error, please notify us immediately by e-mail > and delete the message and any attachments from your system. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
