Viewing my access logs for the last few days, I have seen high-speed bursts of access attempts on one web resource per day. For example, yesterday was an image, today is a PHP document.
Today's burst lasted about twenty seconds totaling 106 requests. Nearly all accesses show the same transfer size (because it's the exact same resource they're hitting) but what worries me is toward the end of the access burst, the transfer size is slightly higher for one access, then the next is nearly twice the transfer size, then goes back to normal.
Log excerpt:
175.xx.xx.x - - [15/Oct/2024:17:05:33 -0700] "GET /business/contact.php HTTP/1.1" 200 5291 175.xx.xx.x - - [15/Oct/2024:17:05:33 -0700] "GET /business/contact.php HTTP/1.1" 200 5291 175.xx.xx.x - - [15/Oct/2024:17:05:33 -0700] "GET /business/contact.php HTTP/1.1" 200 5291 175.xx.xx.x - - [15/Oct/2024:17:05:33 -0700] "GET /business/contact.php HTTP/1.1" 200 5310 <<< larger than normal 175.xx.xx.x - - [15/Oct/2024:17:05:34 -0700] "GET /business/contact.php HTTP/1.1" 200 10481 <<< very large 175.xx.xx.x - - [15/Oct/2024:17:05:34 -0700] "GET /business/contact.php HTTP/1.1" 200 5291 <<< back to normal 175.xx.xx.x - - [15/Oct/2024:17:05:34 -0700] "GET /business/contact.php HTTP/1.1" 200 5291 175.xx.xx.x - - [15/Oct/2024:17:05:35 -0700] "GET /business/contact.php HTTP/1.1" 200 5291 Because apache is showing different sizes toward the end of these access bursts, does this mean a bug or vulnerability is being exploited?
mod_php-enabled Apache servers, PHP itself was called viaSetHandlerand the the decision to call it was performed based on MIME type of the requested resource, which in turn was influenced not only by the last "extension" but by any, so a fileimage.php.jpgcould contain a PHP script (in a JPEG comment, or not be an image at all) which was interpreted.