[Contrib] Python-based alternative for clustercheck and xinetd

[Raghavendra D Prabhu]

This bug is to track the python-based script contributed by
dbusby here - https://github.com/Oneiroi/clustercheck

[David Busby (d-busby)]

Updated now has basic flag set / unset to prevent cache storming,

`ab -c 200 -n 1000 http://localhost:8000`

This is ApacheBench, Version 2.3 <$Revision: 1430300 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking localhost (be patient)
Completed 100 requests
Completed 200 requests
Completed 300 requests
Completed 400 requests
Completed 500 requests
Completed 600 requests
Completed 700 requests
Completed 800 requests
Completed 900 requests
Completed 1000 requests
Finished 1000 requests

Server Software: BaseHTTP/0.3
Server Hostname: localhost
Server Port: 8000

Document Path: /
Document Length: 57 bytes

Concurrency Level: 200
Time taken for tests: 0.494 seconds
Complete requests: 1000
Failed requests: 999
   (Connect: 0, Receive: 0, Length: 999, Exceptions: 0)
Write errors: 0
Non-2xx responses: 1000
Total transferred: 183007 bytes
HTML transferred: 50007 bytes
Requests per second: 2023.19 [#/sec] (mean)
Time per request: 98.854 [ms] (mean)
Time per request: 0.494 [ms] (mean, across all concurrent requests)
Transfer rate: 361.58 [Kbytes/sec] received

Connection Times (ms)
              min mean[+/-sd] median max
Connect: 0 0 1.1 0 12
Processing: 1 3 21.5 2 482
Waiting: 1 3 21.5 1 482
Total: 2 3 22.0 2 494

Percentage of the requests served within a certain time (ms)
  50% 2
  66% 2
  75% 2
  80% 2
  90% 2
  95% 2
  98% 2
  99% 3
 100% 494 (longest request)

[Jay Janssen (jay-janssen)]

Heh -- and how much did 'Com_show_status' and 'Com_show_variables' grow on the server? ;)

[David Busby (d-busby)]

Repo created: https://github.com/Oneiroi/clustercheck gist will no longer be maintained

[David Busby (d-busby)]

@Jay

retest single PXC local node, fresh install not configured

dbusby@icleus-oneiroi-co-uk ~/Documents/Projects/Github/clustercheck$ sudo service mysql restart 1 ↵ master
Shutting down MySQL (Percona XtraDB Cluster).... [ OK ]
Starting MySQL (Percona XtraDB Cluster)......... [ OK ]
 mysql -Bse 'show global status like "Com_show_status"; show global status like "Com_show_variables"' master
Com_show_status 1
Com_show_variables 0

python clustercheck.py
...
localhost.localdomain - - [13/May/2013 12:45:16] "GET / HTTP/1.0" 503 -
localhost.localdomain - - [13/May/2013 12:45:18] "GET / HTTP/1.0" 503 -
----------------------------------------
Exception happened during processing of request from ('127.0.0.1', 36304)
Traceback (most recent call last):
  File "/usr/lib64/python2.7/SocketServer.py", line 284, in _handle_request_noblock
    self.process_request(request, client_address)
  File "/usr/lib64/python2.7/SocketServer.py", line 310, in process_request
    self.finish_request(request, client_address)
  File "/usr/lib64/python2.7/SocketServer.py", line 323, in finish_request
    self.RequestHandlerClass(request, client_address, self)
  File "/usr/lib64/python2.7/SocketServer.py", line 638, in __init__
    self.handle()
  File "/usr/lib64/python2.7/BaseHTTPServer.py", line 340, in handle
    self.handle_one_request()
  File "/usr/lib64/python2.7/BaseHTTPServer.py", line 328, in handle_one_request
    method()
  File "clustercheck.py", line 48, in do_GET
    curs = conn.cursor()
UnboundLocalError: local variable 'conn' referenced before assignment
----------------------------------------
^CTraceback (most recent call last):
  File "clustercheck.py", line 98, in <module>
    httpd.serve_forever()
  File "/usr/lib64/python2.7/SocketServer.py", line 225, in serve_forever
    r, w, e = select.select([self], [], [], poll_interval)
KeyboardInterrupt
...
dbusby@icleus-oneiroi-co-uk ~/Documents/Projects/Github/clustercheck$ mysql -Bse 'show global status like "Com_show_status"; show global status like "Com_show_variables"' 1 ↵ master
Com_show_status 3
Com_show_variables 0

Patching this bug, lines 41 -> 46 presently handle a connection error to MySQL but at line 48 the conn var is attempted to be used regardless of failiure.

[David Busby (d-busby)]

Fix in place, albeit quick and dirty fix: https://github.com/Oneiroi/clustercheck/commit/8e327ac21e17c0c7b7a1eb2746693010e127ee12

...
localhost.localdomain - - [13/May/2013 12:52:32] "GET / HTTP/1.0" 503 -
localhost.localdomain - - [13/May/2013 12:52:32] "GET / HTTP/1.0" 503 -
localhost.localdomain - - [13/May/2013 12:53:09] "GET / HTTP/1.0" 503 -
----------------------------------------
^CTraceback (most recent call last):
  File "clustercheck.py", line 102, in <module>
    httpd.serve_forever()
  File "/usr/lib64/python2.7/SocketServer.py", line 225, in serve_forever
    r, w, e = select.select([self], [], [], poll_interval)
KeyboardInterrupt
dbusby@icleus-oneiroi-co-uk ~/Documents/Projects/Github/clustercheck$ mysql -Bse 'show global status like "Com_show_status"; show global status like "Com_show_variables"' 1 ↵ ✹master
Com_show_status 3
Com_show_variables 0

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PXC-1342