https://bz.apache.org/bugzilla/show_bug.cgi?id=70320
--- Comment #5 from abernal <[email protected]> --- Thank you for testing this so thoroughly. Since the same traffic shows up with nginx, it is not specific to httpd: both servers simply open and read the files from the CIFS share. Your capture also lines up with that: the burst at 17:47:15 comes right after the HTTPS connection from 23.110.173.186, which is the GET / at 17:47:16 in your access log. The packets in that burst are small SMB requests and responses, which look like metadata operations of the kernel CIFS client rather than file content. So the CIFS mount options (cache=, actimeo=) or the NAS itself are where I would look next, and keeping the files on local disk would avoid it altogether. I'll leave this open for now in case you find something that points back to httpd. -- You are receiving this mail because: You are the assignee for the bug. --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
