On Jan 13, 2012, at 14:10 46, Cowboy wrote:

> Certainly not the most desirable solution.

A timestamp in the DB would also greatly bloat the MySQL binlogs -- not good 
for replication setups (especially those pulling across slow WAN links).


> How much time before a lock is considered stale ?
> 10 miliseconds ?
> 24 hours ?

Neither.  Part of the lock includes the ID of the process that created it.  If 
that process is still alive, then _prima facie_ the lock is considered valid 
and hence honored.  If not, then it's stale and can be safely broken.

This is why we'd need system-wide IPC.


> What if someone has opened a file for edit, forgot to close,
> and left for the weekend ? It could happen.

Likely.  Virtually certain in fact, if past experience is any guide.  :)


> Forseeing the possibilities, and unintended consequences,
> is the hard part.

Always is.  Hence this list.

Cheers!


|-------------------------------------------------------------------------|
| Frederick F. Gleason, Jr. |               Chief Developer               |
|                           |               Paravel Systems               |
|-------------------------------------------------------------------------|
|  Harrisberger's Fourth Law of the Lab:                                  |
|       Experience is directly proportional to the amount of equipment    |
|       ruined.                                                           |
|-------------------------------------------------------------------------|

_______________________________________________
Rivendell-dev mailing list
[email protected]
http://lists.rivendellaudio.org/mailman/listinfo/rivendell-dev

Reply via email to