这个是自己写的一个使用awk分析nginx.access.log的脚本，已经开源。欢迎大家 start 或者提issue。
github: https://github.com/Guojunzhou-git/awklogs
awklogs
awklogs
Using awk to analyze nginx.access.log, based on all possible fields.
Bugs Report: php20141104@163.com
usage:
awklogs.sh
Usage: awklogs [--option=value] [...]
--log_type
nginx.access analysis nginx.access.log
json analysis json format file
--log_file
/path/to/xxx.log input path of log file
--analysis_field
request_uri analysis access.log file based on request_uri field, give result of:
- 1. how many request every uri makes
- 2. how many request with every http code
- 3. request_time time_range with every http_code
remote_addr analysis access.log file based on remote_addr field, give result of:
- 1. how many remote_addr have sent request
- 2. how many request every remote_addr makes
- 3. how many bytes every remote_addr used
time_local analysis access.log file based on time_local field, give result of:
- 1. how many request every second makes(concurrency)
- 2. when comes the top N concurrency
- 3. QPS of every seconds
xxx analysis json format file, analysis_field need to be given
--filter_timestamp_start add rows filter time_local start, timestamp like 1500000000
--filter_timestamp_end add rows filter time_local end, timestamp like 1500000000,
ignored when timestamp_end < timestamp_start
--filter_datetime_start add rows filter time_local start, datetime with format "%Y-%m-%d_%H:%M:%S" like 2019-01-01_01:01:01
--filter_datetime_end add rows filter time_local end, datetime with format "%Y-%m-%d_%H:%M:%S" like 2019-01-01_01:01:01,
ignored when datetime_end < datetime_start
--filter_request_uri only analysis the special REQUEST_URI
log type
1. --log_type=nginx.access
analysis nginx.access.log based on default nginx log formay.
log_format main '$remote_addr - $remote_user [$time_local] "$request" '
'$status $body_bytes_sent "$http_referer" '
'"$http_user_agent" "$http_x_forwarded_for" "$upstream_response_time" "$request_time"';
2. --log_type=json
analysis json format log file.
analysis_field
analysis log file based on field
1. request_uri based analysis [--log_type=nginx.access]
analysis access.log file based on request_uri field
awklogs.sh --log_type=nginx.access --log_file=test/teacher.cd.singsound.com_access.log --analysis_field=request_uri
----------------------------------------------------------------------------------------------------------------
request_uri total status count <100ms <200ms <500ms <1000ms <2000ms >2000ms
----------------------------------------------------------------------------------------------------------------
/report.php 2689
200 2687 0 23 0 2652 0 12
499 2 0 0 0 2 0 0
----------------------------------------------------------------------------------------------------------------
/ 2356
500 7 0 0 0 7 0 0
200 2345 0 48 3 2252 7 33
499 4 0 0 0 2 0 2
----------------------------------------------------------------------------------------------------------------
2. remote_addr based analysis [--log_type=nginx.access]
analysis access.log file based on remote_addr field
awklogs.sh --log_type=nginx.access --log_file=test/teacher.cd.singsound.com_access.log --analysis_field=remote_addr
----------------------------------------------------------
remote_addr request byte_sum human_sum
----------------------------------------------------------
124.42.59.17 12713 191631083 182.754 MB
183.136.190.62 102 3097813 2.9543 MB
112.124.3.0 65 0 0 B
222.92.255.178 57 251073 245.188 KB
3. time_local based analysis [--log_type=nginx.access]
analysis access.log file based on time_local field
awklogs.sh --log_type=nginx.access --log_file=test/teacher.cd.singsound.com_access.log --analysis_field=time_local
------------------------------------------------------------------
datetime concurrency avg_time(ms) QPS
------------------------------------------------------------------
2019-06-04 02:40:47 28 0 -
2019-06-06 03:43:26 18 0 -
2019-05-20 17:43:54 7 45.5714 153
2019-05-20 17:43:55 7 41.5714 168
2019-05-20 17:43:56 7 36.1429 193
2019-05-28 09:49:14 6 77.8333 77
4. json field count analysis [--log_type=json]
analysis json format log file based on function field.
awklogs.sh --log_type=json --log_file=test/log_10000.json --analysis_field=function
------------------------------------------------
field count
------------------------------------------------
"event_upload_offline_evalute_audio_... 6341
"send" 1145
"start" 919
"on_event_dev_status_result" 634
"on_message" 420
"handle" 420
"event_submit_answers_run" 58
"event_submit_category_run" 58
"on_connect" 2
"do_answer_submit_post" 2
"on_disconnect" 1
------------------------------------------------
10000 rows analysised, 0 rows ignored. 6 seconds used.
filter
add condition to filter rows
1. time_local based filter: --filter_timestamp_start & --filter_timestamp_end
second timestamp like 1560000000
awklogs.sh --log_type=nginx.access --log_file=test/teacher.cd.singsound.com_access.log --analysis_field=remote_addr --filter_timestamp_start=1560000000 --filter_timestamp_end=1561000000
2009 rows analysised.
----------------------------------------------------------
remote_addr request byte_sum human_sum
----------------------------------------------------------
124.42.59.17 1547 12700363 12.112 MB
222.92.255.178 57 251073 245.188 KB
117.136.31.217 42 161848 158.055 KB
113.88.12.202 42 176401 172.267 KB
113.88.14.85 31 169646 165.67 KB
2. time_local based filter: --filter_datetime_start & --filter_datetime_end
datetime with format "%Y-%m-%d_%H:%M:%S" like 2019-06-08_21:20:00
awklogs.sh --log_type=nginx.access --log_file=test/teacher.cd.singsound.com_access.log --analysis_field=remote_addr --filter_datetime_start=2019-06-08_21:20:00 --filter_datetime_end=2019-06-20_11:06:40
2009 rows analysised.
----------------------------------------------------------
remote_addr request byte_sum human_sum
----------------------------------------------------------
124.42.59.17 1547 12700363 12.112 MB
222.92.255.178 57 251073 245.188 KB
117.136.31.217 42 161848 158.055 KB
113.88.12.202 42 176401 172.267 KB
113.88.14.85 31 169646 165.67 KB
3. request_uri based filter: --filter_request_uri
only analysis the special REQUEST_URI
awklogs.sh --log_type=nginx.access --log_file=test/api2.cd.singsound.com_access.log --analysis_field=request_uri --filter_request_uri=/business/frontDynamic/Config
3982 rows analysised. <1 seconds used.
----------------------------------------------------------------------------------------------------------------
request_uri total status count <100ms <200ms <500ms <1000ms <2000ms >2000ms
----------------------------------------------------------------------------------------------------------------
/business/frontDynamic/Config 3982
200 3976 40 3676 12 0 0 248
499 5 2 3 0 0 0 0
400 1 1 0 0 0 0 0
----------------------------------------------------------------------------------------------------------------
result
analysis result output
1. analysised rows
output how many rows analysised filtered by filters
2. used_time
output how many seconds used during this analysis
网友评论