Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

We are on a Windows Server 2003 VPS running ColdFusion 9, Apache 2.2, and MySQL. We host approximately 75-85 sites on this server running on a shared codebase(homegrown CMS).

At around 1:05-1:10 EST, none of the sites were responsive. Attempting to visit any of the sites resulted in an immediate browser error - site could not be found. My initial assumption was that something had happened with Apache, so I restarted it. This did not resolve the problem, so we restarted ColdFusion. Once ColdFusion came back up, the sites returned to normal.

Now I am in the process of diagnosing the problem, and the logs have not been helpful. All of the CF logs do not have any entries in that time period, so I went to Apache. The only thing I can find in Apache's error log is:

[Thu Apr 07 13:07:17 2011] [warn] Server ran out of threads to serve requests. Consider raising the ThreadsPerChild setting

There's nothing else in the Apache log that indicates other issues.

I did a little research on this, and it appears that ThreadsPerChild can be configured by uncommenting the include from httpd.conf and then modifying httpd-mpm.conf. I'm a little hesitant to start playing with these settings because I'm not sure if this is even the root cause of the problem.

A quick search of error.log shows that we receive this error message 2-4 times per day, and this is the first time (that I am aware of) that we have had a problem like this.

So, I guess my question ultimately is: Is this error message something that has the potential to bring ColdFusion down, or am I looking in the wrong place?

Any suggestions that could be offered would be greatly appreciated.

share|improve this question

1 Answer 1

up vote 0 down vote accepted

That Apache error is not likely to be the issue.

Since you didn't specify which servlet engine your CF is running on, I'm gonna assume it's the default, jrun -- which, to put it lightly, is not the most stable piece of software.

Do some digging for any other logs (java logs in particular; back to the jrun thing) that could have anything from ColdFusion. The most frequent causes that I've seen of these silent hangs are issues with database connections (ODBC especially, got any of those?) and (of course!) heap memory exhaustion.

share|improve this answer
    
Thanks for the quick and informative response. You're correct that we are using jrun. I scoured the logs in ColdFusion9\runtime\logs for any entries around the time that the problem occurred, but there are zero entries between 13:00 and 13:20(when CF was finally restarted). Our JVM is configured to use 512mb of heap memory, but currently is running around 725mb. None of our DSNs are using ODBC drivers, most are using SeeFusion - MySQL. A couple are using MySQL(4/5) and Microsoft SQL Server. –  Sean Walsh Apr 7 '11 at 19:39

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.