added elastic-snapshot script (#193)

* added elastic-snapshot script

* added project to the projects list

* updated requirements.txt
This commit is contained in:
Joe 2020-10-20 23:35:58 -04:00 committed by GitHub
parent 6279e83f26
commit 21b89e112a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
4 changed files with 174 additions and 0 deletions

View File

@ -30,6 +30,7 @@ So far, the following projects have been integrated to this repo:
|[Directory organizer](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/Directory-organizer) | [Athul P](https://github.com/athulpn) | |[Directory organizer](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/Directory-organizer) | [Athul P](https://github.com/athulpn) |
|[DOH DIG](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/DOH-Dig/) | [Ryan](https://github.com/awsumco) | |[DOH DIG](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/DOH-Dig/) | [Ryan](https://github.com/awsumco) |
|[English Theasaurus](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/English_Theasaurus/) | [Ansh Dhingra](https://github.com/anshdhinhgra47) | |[English Theasaurus](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/English_Theasaurus/) | [Ansh Dhingra](https://github.com/anshdhinhgra47) |
|[Elasticsearch snapshot](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/elastic-snapshot) | [Joe Ryan](https://github.com/joeryan) |
|[Excel Files Merger](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/Excel_Files_Merger) | [Andrei N](https://github.com/Andrei-Niculae)| |[Excel Files Merger](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/Excel_Files_Merger) | [Andrei N](https://github.com/Andrei-Niculae)|
|[Excel to List](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/Excel_to_ListofList) | [Nitish Srivastava](https://github.com/nitish-iiitd)| |[Excel to List](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/Excel_to_ListofList) | [Nitish Srivastava](https://github.com/nitish-iiitd)|
|[Extended_ip_address_info](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/extended_ip_address_info) | [hafpaf](https://github.com/hafpaf)| |[Extended_ip_address_info](https://github.com/hastagAB/Awesome-Python-Scripts/tree/master/extended_ip_address_info) | [hafpaf](https://github.com/hafpaf)|

View File

@ -0,0 +1,31 @@
### elastic-snapshot.py
This python3 script helps create snapshots in an elasticsearch cluster. Setting up elasticsearch and creating a snapshot repository are covered in the [elasticsearch documentation](https://www.elastic.co/guide/en/elasticsearch/reference/current/snapshot-restore.html) If a snapshot is already in progress it will sleep for 5 minutes and try again.
- requires the requests library to be installed
- The default behavior is to get a list of snapshots for localhost and the backups snapshot repository
- snapshots are created with at least a minimum of meta data: {taken_by: username, taken_because: 'Snapshot updated <timestamp> for: not specified}
- This script assumes indices are named with dates in the form indexname-YYYY.MM.DD so it can use the period to select indices
- `--testing` flag allows a comma separated list of indices to be supplied. In this case, the period will be used only for the snapshot name.
- optional args and default values:
```
$ python3 elastic-snapshot.py [optional args]
(-a|--action=list) snapshot action to perform (list, create, restore)
(-h|--hostname=localhost:9200)
(-i test* |--index=test*) index name pattern to work on
(-l backups|--location=backups) snapshot repository name
(-n|--noop) do not actually perform the API call but print it to stdout
(-p|--period=2020.01) period for snapshot, unless --testing every index matching pattern indexname*-<period>* is implied
(-r|--reason='not specified') snapshot reason for inclusion in metadata
(-t|--testing=False) allows specifying multiple index names in --index
(-v|--verbose=False) get verbose output to std out
```
- examples:
```
$ python3 elastic-snapshot.py
# returns a list of the snapshots from localhost:9200
$ python3 elastic-snapshot.py --action create --index filebeat* --period 2020.08
# creates a snapshot named filebeat-2020.08-<timestamp> that includes all indices that match the pattern filebeat*-2020.08*
$ python3 elastic-snapshot.py -a create -i metricbeat* -p 2020.08 -l my_snapshots -v
# creates a snapshot named metricbeat-2020.08-<timestamp> in the snapshot repository named my_snapshots. This snapshot will contain any index that matches the pattern metricbeat*-2020.08. Verbose output will be printed to standard output
```

View File

@ -0,0 +1,137 @@
#!/usr/bin/python3
''' elastic-snapshot.py elasticsearch snapshot utility
default action is to get a list of snapshots for localhost on port 9200 and the backups location
optional args:
(-a list |--action=list) snapshot action to perform
(-H localhost:9200 |--hostname=localhost:9200)
(-i testing |--index=testing) index name to work on
(-l backups |--location=backups) location
(-n |--noop) execute script as noop, print web calls and exit.
(-p 2020.01 |--period=2020.01) period of snapshot to take, will have * appended to select additional indices
(-r|--reason=daily) snapshot reason
(-v|--verbose) increase standard output verbosity
'''
import json, os, requests, time
import datetime as dt
from getpass import getuser
def get_snapshot_listing(options):
url = "http://{host}/_cat/snapshots/{sp}?v".format(
sp=options.snapshot_pool, host=options.host_name
)
if options.noop:
return url
else:
#do the real stuff
r = requests.get(url)
if r.status_code == 200:
return r.text
else:
return "Error in making request. received {}: {}".format(r.status_code, r.text)
def create_snapshot(options):
''' creates a snapshot with the specified options using this format of web API request
curl -X PUT "localhost:9200/_snapshot/my_repository/snapshot_2?wait_for_completion=true&pretty" -H 'Content-Type: application/json' -d'
{ "indices": "index_1,index_2",
"ignore_unavailable": true,
"include_global_state": false,
"metadata": {
"taken_by": "user123",
"taken_because": "backup before upgrading" } } '''
indexname = options.index_name
period = options.period
timestamp = dt.datetime.today().strftime('%Y%m%d_%H%M')
# use special testing period for indices to snapshot if defined
if 'testing' in options and options.testing:
indices = indexname
snapshot_name = "{i}-{p}-{d}".format(i="-".join(indexname.split(',')[0].split('*')), p=period,
d=timestamp)
else:
snapshot_name = "{i}-{p}-{d}".format(i=indexname.strip('*'), p=period,
d=timestamp)
indices = "{i}-{p}*".format(i = indexname, p=period)
url = "http://{host}/_snapshot/{sp}/{sn}".format(
sp=options.snapshot_pool, host=options.host_name, sn=snapshot_name
)
payload = { "indices": indices,
"ignore_unavailable": "true",
"include_global_state": "false",
"metadata": {
"taken_by": getuser(),
"taken_because": "Snapshot updated {d} for: {r}".format(d=timestamp,
r=options.reason)
} }
print("create snapshot {} with payload: {}".format(snapshot_name, payload))
if options.noop:
print("Would have called url: {url}\n with payload: {pl}".format(url=url, pl=payload))
return 200
else:
sleep_count = 0
while check_snapshot_inprogress(options):
sleep_count += 1
if options.verbose:
print("A snapshot is running currently, script has slept {} times so far".format(
sleep_count
))
time.sleep(300)
if sleep_count > 20:
print("Snapshot still running, exceeded 10 sleep cycles")
exit(2)
snap_result = requests.put(url, json=payload)
if snap_result.status_code == 200 and options.verbose:
print("Requested: {}".format(snap_result.json()))
else:
print("Error encountered {code}: {response}".format(code=snap_result.status_code,
response=snap_result.json()))
return snap_result.status_code
def restore_snapshot(options):
print("TODO: restore snapshot with options: {}".format(options))
def check_snapshot_inprogress(options):
snapshot_list = get_snapshot_listing(options)
for line in [line for line in snapshot_list.split('\n')]:
if 'IN_PROGRESS' in line.split():
if options.verbose:
print("snapshot in progress: {}".format(line))
return 'IN_PROGRESS' in snapshot_list
if __name__ == '__main__':
import argparse
parser = argparse.ArgumentParser(description='''Command line utility to take snapshots of elasticsearch indices.
Snapshot is created or updated.''')
parser.add_argument('--action', '-a', action='store', dest='action', default='list',
help="what snapshot action to perform: list, create")
parser.add_argument('--host', '-H', action='store', dest='host_name', default='localhost:9200',
help="elasticsearch host")
parser.add_argument('--index', '-i', action='store', dest='index_name', default='test*',
help="Name of index to snapshot")
parser.add_argument('--location', '-l', action='store', dest='snapshot_pool', default='backups',
help="Name of snapshot location")
parser.add_argument('--noop', '-n', action='store_true', default=False, help="just print expceted call")
parser.add_argument('--period', '-p', action='store', dest='period', default='2020.01',
help="period of snapshots to take")
parser.add_argument('--reason', '-r', action='store', dest='reason',
default='not specified', help="text description about why snapshot was taken")
parser.add_argument('--testing','-t', action='store_true', default=False,
help="allows multiple indices to be listed for --index instead of default <index_name>-<period> for testing")
parser.add_argument('--verbose', '-v', action='store_true', default=False, help="Show verbose output for script")
options = parser.parse_args()
action = options.action
if action == 'list':
print(get_snapshot_listing(options))
elif action == 'create':
create_snapshot(options)
else:
print("{action} is not a valid option.\nTry -h|--help for assistance".format(action=action))

View File

@ -0,0 +1,5 @@
certifi==2020.6.20
chardet==3.0.4
idna==2.10
requests==2.24.0
urllib3==1.25.11