Porting loganalyser tools to pandas + jupyter notebooks

In [11]:

    

%matplotlib inline
from loganalyser import jvm, plot


    

In [12]:

    

jvm = reload(jvm)
plot = reload(plot)

gc = jvm.parse('data/grass-sheep.log', 'Europe/Berlin')
#gc = jvm.parse('data/FReportServer_gc.log', 'Europe/Berlin')
gc.head() # note timestamps are in UTC! (workaround for to_json)


    

    
Out[12]:






  

    

      

      
edenSize
      
edenUsed
      
elapsed
      
fromSize
      
fromUsed
      
fullGC
      
permSize
      
permUsed
      
tenuredSize
      
tenuredUsed
      
timestamp
      
toSize
      
toUsed
      
type
    
  
  

    

      
0
      
15360
      
15360
      
0.0070220
      
2560
      
0
      
False
      
21248
      
2730
      
41024
      
0
      
2013-07-18 09:24:42.150
      
2560
      
0
      
GC
    
    

      
1
      
15360
      
0
      
0
      
2560
      
2508
      
False
      
21248
      
2730
      
41024
      
12758
      
2013-07-18 09:24:42.150
      
2560
      
0
      
NaN
    
    

      
2
      
15360
      
15206
      
0.0130340
      
2560
      
2508
      
False
      
21248
      
2734
      
41024
      
12758
      
2013-07-18 09:24:42.290
      
2560
      
0
      
GC
    
    

      
3
      
30720
      
0
      
0
      
2560
      
0
      
False
      
21248
      
2734
      
41024
      
27516
      
2013-07-18 09:24:42.290
      
2560
      
2534
      
NaN
    
    

      
4
      
30720
      
0
      
0.0200690
      
2560
      
0
      
True
      
21248
      
2734
      
41024
      
27516
      
2013-07-18 09:24:42.304
      
2560
      
2534
      
Full GC
    
  

In [13]:

    

gc.dtypes


    

    
Out[13]:





edenSize               object
edenUsed                int64
elapsed                object
fromSize               object
fromUsed                int64
fullGC                   bool
permSize               object
permUsed               object
tenuredSize            object
tenuredUsed            object
timestamp      datetime64[ns]
toSize                 object
toUsed                  int64
type                   object
dtype: object

In [14]:

    

# find out boundaries to select a timeframe -> use ms since epoch
# in a next version boundaries are provided by ipython widgets!
import datetime
from pytz import timezone
print gc.ix[20]
# proper way with tz http://www.saltycrane.com/blog/2009/05/converting-time-zones-datetime-objects-python/
ts = timezone('Europe/Berlin').localize(datetime.datetime.strptime('2013-07-18 11:24:45.356000', '%Y-%m-%d %H:%M:%S.%f'))
print int(ts.strftime("%s")) * 1000


    

    




edenSize                            61632
edenUsed                            61015
elapsed                         0.0230580
fromSize                            44992
fromUsed                                0
fullGC                              False
permSize                            21248
permUsed                             2734
tenuredSize                        133824
tenuredUsed                         48587
timestamp      2013-07-18 09:24:45.356000
toSize                              51328
toUsed                              44542
type                                   GC
Name: 20, dtype: object
1374139485000

In [15]:

    

plot.set_styles(['jvm'])


    

    
Out[15]:

In [16]:

    

plot = reload(plot)
plot.draw_graph('jvm', gc.to_json(orient='records'))


    

    
Out[16]:

In [ ]: