My Apache (Linux/CentOS) is behind a load balancer (AWS ELB). So in my Apache access_log, there are tons of useless logs, like:
- - - [13/Jan/2016:23:38:02 +0800] "GET / HTTP/1.1" 200 16 "-" "ELB-HealthChecker/1.0"
- - - [13/Jan/2016:23:42:39 +0800] "OPTIONS * HTTP/1.0" 200 - "-" "Apache (internal dummy connection)"
These two logs heavily flooded in my access_log. Actually these two are totally useless for me (since they are just internal health checkers only), but affecting/confusing to my log analyzers, like awstats. So i kinda don't want to have this kind of log entries.
Is there a way to configure Apache to WHAT NOT TO log?
Related
Yesterday we faced a strange behavior when reading access log of Apache httpd. An example of below:
207.46.13.135 - - [25/Sep/2022:15:28:28 +0700] "GET / HTTP/1.1" 302 287 (core.c/0/translate_name) - 140
This is a normal access entry: x.x.x.x - - [26/Sep/2022:14:16:57 +0700] "GET /corp/L003/consumer/theme/vn.ssc.css HTTP/1.1" 200 1043 (core.c/0/handler) - 830
We have directives to proxy and redirect the request going to the system. But why this does not redirect (the return code 302 is understandable when in debug mod but why we don't get it when having in production log) – > We suspected that these IPs used some kind of engines to flood the web server, only to response status but not the content.
I setup an ELK stack in AWS Elasticsearch. I'm ingesting Apache logs into ELK, one of the apache fields is the vhost. In the vhost field it will have www.domain.com and domain.com, I'd like to combine those so I can setup accurate searches and visualize data by vhost. Currently it separates www.domain.com and domain.com as separate values.
123.456.7.89 - - [13/Feb/2017:18:56:19 +0000] thisdomain.com "GET /about.html HTTP/1.0" 200 1446 "-" "-" Server=aws8 SSL=- 634713 0
123.456.7.89 - - [13/Feb/2017:18:58:19 +0000] www.thisdomain.com "GET /services.html HTTP/1.0" 200 1446 "-" "-" Server=aws8 SSL=- 634713 0
I entered this data into Elasticsearch setup in AWS which gave me my field definitions. Also works in Cloudwatch.
[ip, user, username, timestamp, vhost, request, status_code, bytes, referrer, browser, server, ssl, timems, times]
I am running a website with siginup/login process. and I like to insert user ids when the user signs up or in to my website.
My apache access.log is as in the following:
115.137.10.87 - - [26/Aug/2013:07:38:52 +0900] "GET /api/game/ HTTP/1.1" 200 1402 "-" "android-async-http/1.4.3 (http://loopj.com/android-async-http)"
115.137.10.87 - - [26/Aug/2013:07:38:59 +0900] "GET /api/premium/brand HTTP/1.1" 200 721 "-" "android-async-http/1.4.3 (http://loopj.com/android-async-http)"
115.137.10.87 - - [26/Aug/2013:07:38:59 +0900] "GET /api/brand/ HTTP/1.1" 200 2510 "-" "android-async-http/1.4.3 (http://loopj.com/android-async-http)"
115.137.10.87 - - [26/Aug/2013:07:39:00 +0900] "GET /api/game/ HTTP/1.1" 200 1402 "-" "android-async-http/1.4.3 (http://loopj.com/android-async-http)"
I can clearly see what http request was sent from what ip address.
But if I put the user id (eg. $userid) to this log, it would be wonderful log, and I will be able to get a good statistic information for making decision.
If it is possible, is there any way to do this?
Any advice will be really appreciated.
The log will be like this:
115.137.10.87 - - [26/Aug/2013:07:38:52 +0900] "***[USERID]*** GET /api/game/ HTTP/1.1" 200 1402 "-" "android-async-http/1.4.3 (http://loopj.com/android-async-http)"
If you are using HTTP Basic or Digest authentication, then the user name is derived from an HTTP header and can be inserted into the log by using %u in a LogFormat directive in your apache config. If you have a login system that does not include the username in a header (likely), then the only other option I know of would be to somehow set an environment variable which Apache could then write out with %{FOOBAR}e. See http://httpd.apache.org/docs/2.2/logs.html
Otherwise, you're probably better off writing out your own log file from your application code (or use an application level logging library - Log4j, or whatever.)
I am currently getting 500 errors from Apache using a alarming probe shell script that has been provided to myself.
Unfortunately I have not been able to get to the bottom of why the script generates a 500 error when attempting to access content locally on the server but using other methods like wget and telnet works fine.
The following are the Apache access log entries for each of the attempts:
Using Wget
127.0.0.1 - "" [19/Mar/2013:14:31:44 +1100] "GET /index.html HTTP/1.1" 200 1635 "-" "Wget/1.13.3" "-"
Using Telnet
127.0.0.1 - "" [20/Mar/2013:13:12:11 +1100] "GET /index.html HTTP/1.1" 200 1635 "-" "-" "-"
Using the Probe Scripts
127.0.0.1 - - [19/Mar/2013:14:33:56 +1100] "GET /index.html HTTP/1.1" 500 - "-" "" "-"
The only difference I can see is that the probe has a - instead of a "" in the user agent (3rd item) which either way tells me it wasn't passed in any of the instances (as this is expected since there is no authentication).
I've bumped up the logging for everything in Apache and can't figure out what is amiss. There is no processing involved, it's a static file, and I have attempted with other file types too, like images to no avail.
Does anyone have any ideas or has seen something similar?
Thanks,
Tony
In Apache access.log, I am used to this kind of access log line:
127.0.0.1 - frank [10/Oct/2000:13:55:36 -0700] "GET /apache_pb.gif HTTP/1.0" 200 2326
I was checking some apache access logs this morning and found something I'm not used to:
192.168.1.10- - [20/Feb/2013:00:00:45 +0000] "POST /form/... 404 200 252 "-" "-" 435835
There are multiple status code. Does-it mean the request was sended multiple times (something like a failed/retry mechanism?