Cache File Locking
I’m performing some complex SQL selects to generate some of these cached modules.
Some of them are taking longer than I would like, but priorities demand that I concentrate
on tasks other than database performance. I simply added some cron tasks to automate the
refreshing of modules in coordination with their tag REFRESH tag. The cron tasks simply
request the page from the web server.
Some of them are taking longer than I would like, but priorities demand that I concentrate
on tasks other than database performance. I simply added some cron tasks to automate the
refreshing of modules in coordination with their tag REFRESH tag. The cron tasks simply
request the page from the web server.
However, I was concerned with multiple users hitting the server initiating redundant
SQL executions, which would unnecessarily increase the load on the database.
SQL executions, which would unnecessarily increase the load on the database.
I used a simple locking solution. If the script determines that a cache file needs
to be updated, it creates a lock file that indicates the cache file is already being
generated. Subsequent requests for that same cache file notice that the lock file exists
and wait until that lock file is removed by the initial script and then grabs the cache
file when it finishes.
to be updated, it creates a lock file that indicates the cache file is already being
generated. Subsequent requests for that same cache file notice that the lock file exists
and wait until that lock file is removed by the initial script and then grabs the cache
file when it finishes.
Hopefully, with these enhancements, this solution becomes even more powerful while
making it easier to implement for the developer.
making it easier to implement for the developer.
You should be able to cut and paste it into your own solution.
