Fixed
Details
Assignee
Michael OffnerMichael OffnerReporter
Colby LitnakColby Litnak(Deactivated)Priority
CriticalFix versions
New Issue warning screen
Before you create a new Issue, please post to the mailing list first https://dev.lucee.org
Once the issue has been verified, one of the Lucee team will ask you to file an issue
Sprint
NoneAffects versions
Details
Details
Assignee
Michael Offner
Michael OffnerReporter
Colby Litnak
Colby Litnak(Deactivated)Priority
Fix versions
New Issue warning screen
Before you create a new Issue, please post to the mailing list first https://dev.lucee.org
Once the issue has been verified, one of the Lucee team will ask you to file an issue
Sprint
None
Affects versions
Created 20 November 2015 at 22:25
Updated 9 April 2019 at 08:13
Resolved 29 March 2019 at 20:54
In windows there is an outgoing TCP connection limit of 5000 (though configurable)
We have a server with many instances of our application on it. It is a beefy server and it can handle it in terms of resources.
Our application sometimes have a scheduled task that we run every 30 seconds. Sometimes when our servers are some load those scheduled tasks get a little bit behind, meaning a scheduled task starts when the last one isn't finished, and when the conditions are just right our server uses all 5000 connections and most things do not work. Upon investigation we find that many of the connections are scheduled tasks getting behind.
We have a fix that has prevents lucee's scheduled tasks from starting when the scheduled task is already running, Obviously, this isn't the only thing we changed to improve the stability of our app, but we feel this (or something like it) is a good change. I am logging the issue so you are aware, and hope you would consider the pending pull request.