Reconcile descrepencies in document counts within elasticsearch

es-doc-accounting.py

from requests import get 
from json import dumps, loads
from prettytable import PrettyTable


src_server = 'http://localhost:9200'

def docCount(idx=None, typ=None):
    url = src_server
    if idx:
        url += "/%s" % idx
    if typ:
        url += "/%s" % typ
    url += "/_count" 
    res = loads(get(url).content)
    return res['count'] if 'count' in res else 0



tbl = PrettyTable(['Index', 'Count on Index', 'Docs In Types'])
tbl.add_row(['entire cluster', docCount(), 'n/a'])
tbl.add_row(['=====','=====','====='])
src_state=loads(get("%s/_cluster/state" % (src_server), timeout=600).content)
docsCountedInIndexes = 0
docsCountedInTypes = 0
for idx in src_state['metadata']['indices']:
    docsInIndex = docCount(idx)
    docsCountedInIndexes += docsInIndex
    docsInTypes = 0
    for typ in src_state['metadata']['indices'][idx]['mappings']:
        docsInTypes += docCount(idx, typ)
    docsCountedInTypes += docsInTypes
    tbl.add_row([idx, docsInIndex, docsInTypes])

tbl.add_row(['-----','------','-----'])
tbl.add_row(['Totals', docsCountedInIndexes, docsCountedInTypes])

print(tbl)