Curator: Remote end closed connection without response curator

[ep4sh]

To submit a bug or report an issue

Expected Behavior

2019-01-09 11:10:06,701 INFO      All indices appear to have been restored.
2019-01-09 11:10:06,702 INFO      Action ID: <22>, "restore" completed.

Actual Behavior

2019-08-02 03:39:05,457 INFO      Restoring indices "['applicationlog-trace-2019.01.01']" from snapshot: logs-weeks-13_2018.12.31_2019.01.09
2019-08-02 03:43:16,676 ERROR     Failed to complete action: restore.  <class 'curator.exceptions.FailedExecution'>: Exception encountered.  Rerun with loglevel DEBUG and/or check Elasticsearch logs for more information. Exception: ConnectionError(('Connection aborted.', RemoteDisconnected('Remote end closed connection without response',))) caused by: ProtocolError(('Connection aborted.', RemoteDisconnected('Remote end closed connection without response',)))

Steps to Reproduce the Problem

1.

curator --config curator.yml restore.yml

1. In the restore.py :

client = es.Elasticsearch([{'host': HOST, 'port': 443, 'use_ssl': True, 'ssl-no-validate': False, 'timeout': 600}])

1. In the config.yml :

client:
  hosts: bla-bla.bla.com
  port: 443
  url_prefix:
  use_ssl: True
  ssl_no_validate: False
  timeout: 30

logging:
  loglevel: INFO
  logformat: default
  blacklist: ['elasticsearch', 'urllib3']

Specifications

Centos 7.6
virualenv 3
elasticsearch 7.0.2
elasticsearch-curator 5.7.6

Context (Environment)

Detailed Description

My ES host bla-bla.bla.com working and response all the requests.
4 months ago my restore script worked fine, but I have no changes by this time (checked via git)
Possible changes are versions of the elasticsearch and elasticsearch-curator packages, but the same time snapshot creating with curator works fine (with similar settings)

[ep4sh]

LOL)
I have Elasticsearch cluster v6.4, so I have to use 6.x.x version of the library elasticsearch:
https://pypi.org/project/elasticsearch/
Please close the issue.

[untergeek]

Are you sure that's all it is? Curator 5.7.6 _should_ just work with older versions of Elasticsearch using the 7.x version of the Elasticsearch python module.

[ep4sh]

@untergeek Hi, Aaron. Thank you for the reply.

Yes, in my case it was the solution (I used elasticsearch==6.4.0)

From the elasticsearch client pypi page:

Compatibility

The library is compatible with all Elasticsearch versions since 0.90.x but you have to use a matching major version:
For Elasticsearch 6.0 and later, use the major version 6 (6.x.y) of the library.