Thread

Hello there,

Until recently there hasn’t been any issues with our system, but since last week our PHP cache has been rapidly filling up and causing PHP errors within the EE admin. When looking through the list of system cache entries, mod.channel.php appears to be the culprit. The memory allocation was increased to handle a larger load, but didn’t seem to solve the problem (it just takes a little longer for the error to occur). The site was properly optimized and templated using the recommended best practices and like I said, hasn’t had any problems since the launch early this year.

We are still using build 20101220, so I’m wondering if there’s a chance that build 20110411 would fix this? I wasn’t able to find any other support threads that referred to the same issue. Any insight on what could be causing this and what to do about it would be much appreciated.

Thanks,
Daniel

Hi Daniel,

How much PHP Memory do you currently have allocated?

What specific errors have you been getting?

Hey John,

Thanks for responding. According to php.ini 512M is allocated, but APC is showing only 96M shared memory is being used for PHP cache. The error I get is something along the lines of “Unable to allocate memory for pool” but I recently cleared the cache so I am not getting the error at the moment.

Here are some of the errors:

A PHP Error was encountered
Severity: Warning
Message: include_once() [function.include-once]: Unable to allocate memory for pool.
Filename: core/Loader.php
Line Number: 779

A PHP Error was encountered
Severity: Warning
Message: include() [function.include]: Unable to allocate memory for pool.
Filename: core/EE_Lang.php
Line Number: 167

A PHP Error was encountered
Severity: Warning
Message: require() [function.require]: Unable to allocate memory for pool.
Filename: core/Common.php
Line Number: 137

Hi Daniel,

I don’t think that this is related to EE at all but more likely an APC issue. I found this that may indicate what the problem is. Please contact your host and ask them to look into this for you.

Cheers

Greg

Thanks. This was very helpful. I will reach out to my host and report back once I find out what is causing this.

Thanks, Daniel. Keep us posted with what you hear from your host.

Kevin

Resolved! Hosting support suggested an increase of the apc.ttl (it was set at zero for some reason) and there have been no errors since setting it to 7200. Apparently APC received an update recently that happened to change some its behavior. Special thanks to Greg for the tip.

Awesome, glad to see you’re squared away!