Win knows when files are open - but more research needs to be done on how(if) Windows actually locks files - the word processors seem to use a 'flag' file. (Which makes for interesting issues if you don't close the file on a flashdrive before transferring the flashdrive to another computer.  )

oops, meant 2012, not 2015.

I know that with Win server 2003, 2008, 2015 when using word processing programs (WordPerfect, Word) We get file lock notifications (File in use/File can be opened but must be saved under a different name) as they see the file is open elsewhere. In the Windows 2015 control panel there is a section that allows you to see the open files. So I know there is something built into Win Server that indicates an open file. Not sure if the OS locks it or if its built into the other programs to see the open file and operate as it is "Locked"?

Red Barchetta wrote on Oct 10^th, 2016 at 7:36pm:


From the literature it seems that Windows takes file-locking 'under advisement' rather than just doing what it's told. But I've also noticed that for a lot of programs, file locking involves writing a 'hidden' file to the HD to to identify files locked for editing. Probably not a good solution on a server. 

Better now, for some reason it was set to 60%. I think I was told a long time ago that the WIN file locking did not actually work.

You can adjust your personal text size in Edit Profiles -> Options.

The fopen and fclose were designed way back when to handle file locking issues when programming for both *nix and Win file systems.  And then things like counting how files were opened and closed for debuging got tucked into those functions.

The normal perl way is: open the file using perl's 'open', lock the file using one of the locking protocols (which should lock the file against being worked on by another process while being written to) and then close the file (which should automatically release the lock.)

The problem for YaBB was (and probably still is) that *nix and Windows use(d) different methods for locking files.

This is one of the reasons we keep looking at moving data storage to database - the locking mechanisms are supposed to be better.

Is the  fopen and fclose strictly a perl thing, and/or is it a server/host issue?

(and why is my text so small?)

Computer is back working, I have almost all my data back, I'm hard at work on getting YaBB 2.7.0 to comply with the strict pragma. I'm hoping to get the strict version into the SourceForge SVN sometime next week. NOTE: this test version does not have any file locking as I still have to get the fopen and fclose functions to behave under the strict pragma. (I'm also not sure how well they actually work at locking files but Debug pulls data out of those functions so they (maybe) are stil needed.)

Other changes in this version - many variables and subroutines have been converted to lowercase (This has has been the recommendation for many years and helps prevent some of the issues with variables that Monni and others have uncovered.) I'll post a list of the changes later.

Luckily this is on the desktop - so the OS was okay and the programs weren't damaged. But there's about 500 gb of music that I may or may not be able to recover.
And there's a lot of 0-length files I'll have to recreate. 
Edited:

That's a lot of data.

But I know what that's like. I whittled mine down to just under 180 GB of essential data on my new working laptop. My older laptop had about 930 GB.

One of the positives of only copying essential data onto the new laptop is that my other data, which are frequently accessed, are readily available on two backup drives. So for now at least, my new laptop is lean, and optimized.

Considering all of the work that I've done over the years, I'm a bit surprised that I only have about 1.3 TB of data that's copied, and backed up on various external drives. Then again, a lot of work was done with partners, so the media holding it is dependent on who has the rights.

To this day, I'm still amazed at how much data Google stores. The last I read on this claimed around 10-15 exabytes. But I think the NSA holds the most, or surely near the top. Imagine keeping all of that data backed up, although in practice I think it's basically a series of relatively small backups that occur simultaneously.

In any case, the relatively large amount of data that I have is one of the reasons I keep it backed up incrementally. Mostly. When it comes time to start over with a fresh, clean backup, that's when I feel the pain of how much time it takes.

Hang in there.

250 gb of data takes a heck of a long time to copy from a HDD that won't pass Check Disk. 

OUCH! Been there; done that.

Not long ago this almost happened again, but this time I had the luck to have seen the signs of a major failure before it actually happened. I now have a new laptop computer because of what happened, and just in the nick of time, too.

But even with my good luck this time, it still wasn't a particularly pleasant task to basically recreate my work space on the new laptop. I do have a desktop that more or less duplicates my work space, but I was still on edge about it all for a good week or so.

So I feel your pain. Good luck with getting everything back in order. This too shall pass.

I've just had a major computer failure that took out my data drive and the most recent backup.

It may be a while before I can get back to work.