| 1 |
54585
|
sandro.lab
|
import boto
|
| 2 |
|
|
import boto.s3.connection
|
| 3 |
|
|
import sys
|
| 4 |
54591
|
sandro.lab
|
import pymongo
|
| 5 |
54585
|
sandro.lab
|
from pymongo import MongoClient
|
| 6 |
|
|
import os.path
|
| 7 |
54586
|
sandro.lab
|
import time
|
| 8 |
54585
|
sandro.lab
|
|
| 9 |
|
|
|
| 10 |
54586
|
sandro.lab
|
|
| 11 |
|
|
def file_exist(bucket, file_path):
|
| 12 |
54587
|
sandro.lab
|
c =bucket.get_key(file_path)
|
| 13 |
|
|
if c is not None and c.size > 0:
|
| 14 |
|
|
return True
|
| 15 |
|
|
return False
|
| 16 |
54586
|
sandro.lab
|
|
| 17 |
54585
|
sandro.lab
|
def exportItemForMongoCollection(obsId, db, bucket, log_file):
|
| 18 |
|
|
destination_collection =db[ 's3_'+obsId[:36]]
|
| 19 |
|
|
source_collection = db[ obsId[:36]]
|
| 20 |
|
|
print 'Start to exporting objectStore :%s '%obsId
|
| 21 |
|
|
i = 0
|
| 22 |
54595
|
sandro.lab
|
skip = 0
|
| 23 |
54585
|
sandro.lab
|
last_percentage = 0
|
| 24 |
|
|
total = source_collection.estimated_document_count()
|
| 25 |
|
|
for item in source_collection.find(no_cursor_timeout=True):
|
| 26 |
|
|
fs_path = item['fsPath']
|
| 27 |
|
|
objectId = item['id']
|
| 28 |
54593
|
sandro.lab
|
dest_item = destination_collection.find_one({'id':objectId})
|
| 29 |
54591
|
sandro.lab
|
if dest_item is None:
|
| 30 |
|
|
if os.path.isfile(fs_path):
|
| 31 |
|
|
i += 1
|
| 32 |
|
|
if not file_exist(bucket, '%s/%s'%(obsId,objectId)):
|
| 33 |
|
|
key = bucket.new_key('%s/%s'%(obsId,objectId))
|
| 34 |
|
|
try:
|
| 35 |
|
|
key.set_contents_from_filename(fs_path)
|
| 36 |
|
|
except Exception as e:
|
| 37 |
54600
|
sandro.lab
|
for k in range(10):
|
| 38 |
|
|
try:
|
| 39 |
|
|
time.sleep(10)
|
| 40 |
|
|
key.set_contents_from_filename(fs_path)
|
| 41 |
|
|
except Exception as e:
|
| 42 |
|
|
print "Error on saving object on S3"
|
| 43 |
|
|
print e
|
| 44 |
|
|
print "Sleeping 10 seconds and retry"
|
| 45 |
|
|
|
| 46 |
54591
|
sandro.lab
|
item.pop('_id', None)
|
| 47 |
|
|
item.pop('fsPath')
|
| 48 |
|
|
item['uri'] = 's3://%s/%s'%(bucket.name, key.name)
|
| 49 |
|
|
destination_collection.insert_one(item)
|
| 50 |
54595
|
sandro.lab
|
if i % 100 == 0:
|
| 51 |
|
|
print "Exported %i/%i (skipped)%i "%(i, total, skip)
|
| 52 |
54591
|
sandro.lab
|
else:
|
| 53 |
|
|
log_file.writeline('Missing file for objectStoreid: %s ObjectId:%s path: %s'%(obsId, objectId, fs_path))
|
| 54 |
54595
|
sandro.lab
|
else:
|
| 55 |
|
|
skip += 1
|
| 56 |
54591
|
sandro.lab
|
|
| 57 |
54585
|
sandro.lab
|
|
| 58 |
54594
|
sandro.lab
|
def start_import(metadataCollection, bucket, log_file, done_file, skip_store):
|
| 59 |
54585
|
sandro.lab
|
client = MongoClient()
|
| 60 |
|
|
db = client['objectStore']
|
| 61 |
|
|
metadataCollection = db[metadataCollection]
|
| 62 |
54597
|
sandro.lab
|
print skip_store
|
| 63 |
54585
|
sandro.lab
|
for item in metadataCollection.find(no_cursor_timeout=True):
|
| 64 |
|
|
obsId = item['obsId']
|
| 65 |
54594
|
sandro.lab
|
if obsId not in skip_store:
|
| 66 |
54598
|
sandro.lab
|
destination_collection = db[ 's3_'+obsId[:36]]
|
| 67 |
54597
|
sandro.lab
|
destination_collection.create_index([('id',pymongo.ASCENDING)])
|
| 68 |
54594
|
sandro.lab
|
exportItemForMongoCollection(obsId, db, bucket, log_file)
|
| 69 |
|
|
done_file.write('{}\n'.format(obsId))
|
| 70 |
54585
|
sandro.lab
|
|
| 71 |
|
|
|
| 72 |
|
|
|
| 73 |
|
|
|
| 74 |
|
|
|
| 75 |
|
|
if __name__=='__main__':
|
| 76 |
|
|
args = sys.argv
|
| 77 |
|
|
if not len(args) is not 3 :
|
| 78 |
|
|
print "Error on applying script usage: python scriptName.py s3cfgPath objectstoreBucket"
|
| 79 |
|
|
f = open(args[1])
|
| 80 |
|
|
props = {}
|
| 81 |
|
|
for line in f:
|
| 82 |
|
|
d =line.split('=')
|
| 83 |
|
|
if len(d) == 2:
|
| 84 |
|
|
props[d[0].strip()] = d[1].strip()
|
| 85 |
54594
|
sandro.lab
|
skip_store =[]
|
| 86 |
|
|
if os.path.isfile('store_done'):
|
| 87 |
|
|
f = open('store_done')
|
| 88 |
54596
|
sandro.lab
|
skip_store =[line.strip() for line in f if len(line) >0]
|
| 89 |
54585
|
sandro.lab
|
|
| 90 |
|
|
bname = args[2]
|
| 91 |
|
|
conn = boto.connect_s3(aws_access_key_id = props['access_key'], aws_secret_access_key = props['secret_key'], host = props['host_base'], calling_format = boto.s3.connection.OrdinaryCallingFormat())
|
| 92 |
|
|
bucket = conn.get_bucket(bname, validate=True)
|
| 93 |
|
|
log_file = open('s3_migration.log', 'w')
|
| 94 |
54597
|
sandro.lab
|
done_file = open('store_done_1', 'wb')
|
| 95 |
54594
|
sandro.lab
|
start_import('metadataObjectStore',bucket, log_file, done_file, skip_store)
|
| 96 |
54585
|
sandro.lab
|
log_file.close()
|