So, I’ve been meaning to do this for a while. I’m sure your all familiar with this crappy oneliner where you’ll check the hits using wc -l against a webserver log for a given hour and use for i in seq or similar to get the last 10 minutes data, or the last 59 minutes. But getting hours and days is a bit harder. You need some additional nesting, and, it isn’t difficult for you to do at all!<\/p>\n
\r\nfor i in 0{1..9}; do echo "16\/Jan\/2017:06:$i"; grep "16\/Jan\/2017:06:$i" \/var\/log\/httpd\/soemsite.com-access_log | wc -l ; done\r\n<\/pre>\nI improved this one, quite significantly by adding an additional for j, for the hour, and adding an additional 0 in front of {1..9}, this properly is matching the Apache2 log format and allows me to increment through all the hours of the day. to present ;-D All that is missing is some error checking when the last date in the file is, im thinking a tail and grep for the timecode from the log should be sufficient.<\/p>\n
Here is the proud oneliner I made for this!<\/p>\n
\r\nfor j in 0{1..9}; do for i in 0{1..9} {10..59}; do echo "16\/Jan\/2017:$j:$i"; grep "16\/Jan\/2017:06:$i" \/var\/log\/httpd\/website.com-access_log | wc -l ; done; done\r\n<\/pre>\n","protected":false},"excerpt":{"rendered":"So, I’ve been meaning to do this for a while. I’m sure your all familiar with this crappy oneliner where you’ll check the hits using wc -l against a webserver log for a given hour and use for i in … Continue reading