Calculate Ceph BlueStore OSD overhead per object

ceph-bluestore-overhead.py

#!/usr/bin/env python
'''
Talk to local Ceph OSDs running with BlueStore and calculate the overhead
per object in the RocksDB database from BlueStore.

Author: Wido den Hollander <[email protected]>
'''

import json
from os import listdir
from os.path import isfile, join
from subprocess import check_output

SOCKET_DIR = '/var/run/ceph'

def get_local_osds():
    mypath = '/var/run/ceph/'
    sockets = [f for f in listdir(SOCKET_DIR) if (f.startswith('ceph-osd') and f.endswith('asok'))]
    osds = list()
    for socket in sockets:
        osds.append(socket.split('.')[1])

    return osds

def osd_perf_dump(osd_id):
    cmd = ['ceph', 'daemon', 'osd.' + str(osd_id), 'perf', 'dump']
    return json.loads(check_output(cmd))


def get_osd_perf():
    perf = dict()
    for osd_id in get_local_osds():
        perf[osd_id] = osd_perf_dump(osd_id)

    return perf

if __name__ == '__main__':
    perf = get_osd_perf()
    for osd_id, perf in perf.items():
        if 'bluestore' in perf:
            onodes = perf['bluestore']['bluestore_onodes']
            stat_bytes_used = perf['osd']['stat_bytes_used']
            db_used_bytes = perf['bluefs']['db_used_bytes']
            overhead = db_used_bytes / onodes
            avg_obj_size = stat_bytes_used / onodes

            print('osd.{0} onodes={1} db_used_bytes={2} avg_obj_size={3} overhead_per_obj={4}'.format(osd_id, onodes, db_used_bytes, avg_obj_size, overhead))

For mixed (filestore / bluestore) osd-host (ignore non bluestore osds):

--- ceph-bluestore-overhead.py_orig     2018-04-27 16:28:42.063312979 +0700
+++ ceph-bluestore-overhead.py  2018-04-27 16:28:01.035236995 +0700
@@ -37,10 +37,11 @@ def get_osd_perf():
 if __name__ == '__main__':
     perf = get_osd_perf()
     for osd_id, perf in perf.items():
-        onodes = perf['bluestore']['bluestore_onodes']
-        stat_bytes_used = perf['osd']['stat_bytes_used']
-        db_used_bytes = perf['bluefs']['db_used_bytes']
-        overhead = db_used_bytes / onodes
-        avg_obj_size = stat_bytes_used / onodes
+       if 'bluestore' in perf:
+            onodes = perf['bluestore']['bluestore_onodes']
+            stat_bytes_used = perf['osd']['stat_bytes_used']
+            db_used_bytes = perf['bluefs']['db_used_bytes']
+            overhead = db_used_bytes / onodes
+            avg_obj_size = stat_bytes_used / onodes
 
-        print('osd.{0} onodes={1} db_used_bytes={2} avg_obj_size={3} overhead_per_obj={4}'.format(osd_id, onodes, db_used_bytes, avg_obj_size, overhead))
\ No newline at end of file
+            print('osd.{0} onodes={1} db_used_bytes={2} avg_obj_size={3} overhead_per_obj={4}'.format(osd_id, onodes, db_used_bytes, avg_obj_size, overhead))