How to use the dcicutils.es_utils.create_es_client function in dcicutils

def status_of_elasticsearch_indices(connection, **kwargs):
    check = CheckResult(connection, 'status_of_elasticsearch_indices')
    ### the check
    client = es_utils.create_es_client(connection.ff_es, True)
    indices = client.cat.indices(v=True).split('\n')
    split_indices = [ind.split() for ind in indices]
    headers = split_indices.pop(0)
    index_info = {} # for full output
    warn_index_info = {} # for brief output
    for index in split_indices:
        if len(index) == 0:
            continue
        index_info[index[2]] = {header: index[idx] for idx, header in enumerate(headers)}
        if index_info[index[2]]['health'] != 'green' or index_info[index[2]]['status'] != 'open':
            warn_index_info[index[2]] = index_info[index[2]]
    # set fields, store result
    if not index_info:
        check.status = 'FAIL'
        check.summary = 'Error reading status of ES indices'
        check.description = 'Error reading status of ES indices'

def elastic_search_space(connection, **kwargs):
    """ Checks that our ES nodes all have a certain amount of space remaining """
    check = CheckResult(connection, 'elastic_search_space')
    full_output = {}
    client = es_utils.create_es_client(connection.ff_es, True)
    # use cat.nodes to get id,diskAvail for all nodes, filter out empties
    node_space_entries = filter(None, [data.split() for data in client.cat.nodes(h='id,diskAvail').split('\n')])
    check.summary = check.description = None
    full_output['nodes'] = {}
    for _id, remaining_space in node_space_entries:
        if 'gb' not in remaining_space:
            if 'mb' not in remaining_space:
                check.status = 'FAIL'
                check.summary = check.description = 'At least one of the nodes in this env has no space remaining'
            else:
                check.status = 'WARN'
                check.summary = check.description = 'At least one of the nodes in this env is low on space'
        full_output['nodes'][_id.strip()] = { 'remaining_space': remaining_space }
    if check.summary is None:
        check.status = 'PASS'
        check.summary = check.description = 'All nodes have >1gb remaining disk space'

def indexing_records(connection, **kwargs):
    check = CheckResult(connection, 'indexing_records')
    client = es_utils.create_es_client(connection.ff_es, True)
    namespaced_index = connection.ff_env + 'indexing'
    # make sure we have the index and items within it
    if (not client.indices.exists(namespaced_index) or
        client.count(index=namespaced_index).get('count', 0) < 1):
        check.summary = check.description = 'No indexing records found'
        check.status = 'PASS'
        return check

    res = client.search(index=namespaced_index, doc_type='indexing', sort='uuid:desc', size=1000,
                        body={'query': {'query_string': {'query': '_exists_:indexing_status'}}})
    delta_days = datetime.timedelta(days=3)
    all_records = res.get('hits', {}).get('hits', [])
    recent_records = []
    warn_records = []
    for rec in all_records:
        if rec['_id'] == 'latest_indexing':

def __init__(self, index=None, doc_type='result'):
        self.es = es_utils.create_es_client(HOST, use_aws_url=True)
        self.index = index
        if index and not self.index_exists(index):
            self.create_index(index)
        self.doc_type = doc_type

def wipe_build_indices(connection, **kwargs):
    """ Wipes all indices on the FF-Build ES env """
    check = CheckResult(connection, 'wipe_build_indices')
    check.status = 'PASS'
    check.summary = check.description = 'Wiped all test indices'
    BUILD_ES = 'search-fourfront-builds-uhevxdzfcv7mkm5pj5svcri3aq.us-east-1.es.amazonaws.com:80'
    client = es_utils.create_es_client(BUILD_ES, True)
    full_output = client.indices.delete(index='*')
    if full_output['acknowledged'] != True:
        check.status = 'FAIL'
        check.summary = check.description = 'Failed to wipe all test indices, see full output'
    check.full_output = full_output
    return check

#     staging_deploy = CheckResult(staging_conn, 'staging_deployment').get_primary_result()
    #     if staging_deploy['status'] != 'PASS':
    #         check.summary = 'Staging deployment is running - skipping'
    #         return check

    if get_stage_info()['stage'] != 'prod':
        check.summary = check.description = 'This check only runs on Foursight prod'
        return check

    time_limit = 270  # 4.5 minutes
    t0 = time.time()
    check.full_output = {}  # purged items by item type
    search = '/search/?type=TrackingItem&tracking_type=download_tracking&status=deleted&field=uuid&limit=300'
    search_res = ff_utils.search_metadata(search, key=connection.ff_keys)
    search_uuids = [res['uuid'] for res in search_res]
    client = es_utils.create_es_client(connection.ff_es, True)
    # a bit convoluted, but we want the frame=raw, which does not include uuid
    # use get_es_metadata to handle this. Use it as a generator
    for to_purge in ff_utils.get_es_metadata(search_uuids, es_client=client, is_generator=True,
                                             key=connection.ff_keys):
        if round(time.time() - t0, 2) > time_limit:
            break
        purge_properties = to_purge['properties']
        purge_properties['uuid'] = to_purge['uuid']  # add uuid to frame=raw
        try:
            purge_res = ff_utils.purge_metadata(to_purge['uuid'], key=connection.ff_keys)
        except Exception as exc:
            purge_status = 'error'
            purge_detail = str(exc)
        else:
            purge_status = purge_res['status']
            purge_detail = purge_properties if purge_status == 'success' else purge_res