sbcli-pre 1.2.5__zip → 1.2.7__zip

sbcli_pre-1.2.5/simplyblock_core/mgmt_node_ops.py

@@ -1,80 +0,0 @@
-# coding=utf-8
-import json
-import logging
-import uuid
-
-import docker
-
-from simplyblock_core import utils
-from simplyblock_core.controllers import mgmt_events
-from simplyblock_core.kv_store import DBController
-from simplyblock_core.models.mgmt_node import MgmtNode
-
-logger = logging.getLogger()
-
-
-def add_mgmt_node(mgmt_ip, cluster_id=None):
-    db_controller = DBController()
-    hostname = utils.get_hostname()
-    node = db_controller.get_mgmt_node_by_hostname(hostname)
-    if node:
-        logger.error("Node already exists in the cluster")
-        return False
-
-    node = MgmtNode()
-    node.uuid = str(uuid.uuid4())
-    node.hostname = hostname
-    node.docker_ip_port = f"{mgmt_ip}:2375"
-    node.cluster_id = cluster_id
-    node.mgmt_ip = mgmt_ip
-    node.status = MgmtNode.STATUS_ONLINE
-    node.write_to_db(db_controller.kv_store)
-
-    mgmt_events.mgmt_add(node)
-    logger.info("Done")
-    return True
-
-
-def list_mgmt_nodes(is_json):
-    db_controller = DBController()
-    nodes = db_controller.get_mgmt_nodes()
-    data = []
-    output = ""
-
-    for node in nodes:
-        logging.debug(node)
-        logging.debug("*" * 20)
-        data.append({
-            "UUID": node.get_id(),
-            "Hostname": node.hostname,
-            "IP": node.mgmt_ip,
-            "Status": node.status,
-        })
-
-    if not data:
-        return output
-
-    if is_json:
-        output = json.dumps(data, indent=2)
-    else:
-        output = utils.print_table(data)
-    return output
-
-
-def remove_mgmt_node(uuid):
-    db_controller = DBController()
-    snode = db_controller.get_mgmt_node_by_id(uuid)
-    if not snode:
-        logger.error("can not find node")
-        return False
-
-    logging.info("Removing mgmt node")
-    snode.remove(db_controller.kv_store)
-
-    logger.info("Leaving swarm...")
-    node_docker = docker.DockerClient(base_url=f"tcp://{snode.docker_ip_port}", version="auto")
-    node_docker.swarm.leave()
-
-    mgmt_events.mgmt_remove(snode)
-    logging.info("done")
-

sbcli_pre-1.2.5/simplyblock_core/scripts/apply_dashboard.sh

@@ -1,22 +0,0 @@
-#!/bin/bash
-
-TD=$(dirname -- "$(readlink -f -- "$0")")
-
-# Grafana Password
-export grafanaPassword=$1
-
-# Grafana username
-GF_ADMIN_USER=admin
-
-HOST=0.0.0.0:3000
-
-DASHBOARDS="${TD}/dashboards"
-for dashboard in "${DASHBOARDS}/cluster.json" "${DASHBOARDS}/devices.json" "${DASHBOARDS}/nodes.json" "${DASHBOARDS}/lvols.json" "${DASHBOARDS}/pools.json" "${DASHBOARDS}/node-exporter.json"; do
-    echo -e "\nUploading dashboard: ${dashboard}"
-    curl -X POST -H "Content-Type: application/json" \
-        -d "@${dashboard}" \
-        "http://${GF_ADMIN_USER}:${grafanaPassword}@${HOST}/api/dashboards/import"
-    echo ""
-done
-
-echo "Cluster deployment complete."

sbcli_pre-1.2.5/simplyblock_core/services/health_check_service.py

@@ -1,136 +0,0 @@
-# coding=utf-8
-import logging
-
-import time
-import sys
-from datetime import datetime
-
-
-from simplyblock_core.controllers import health_controller, storage_events, device_events
-from simplyblock_core.models.storage_node import StorageNode
-from simplyblock_core.rpc_client import RPCClient
-from simplyblock_core import constants, kv_store
-
-# Import the GELF logger
-from graypy import GELFUDPHandler
-
-def set_node_health_check(snode, health_check_status):
-    snode = db_controller.get_storage_node_by_id(snode.get_id())
-    if snode.health_check == health_check_status:
-        return
-    old_status = snode.health_check
-    snode.health_check = health_check_status
-    snode.updated_at = str(datetime.now())
-    snode.write_to_db(db_store)
-    storage_events.snode_health_check_change(snode, snode.health_check, old_status, caused_by="monitor")
-
-
-def set_device_health_check(cluster_id, device, health_check_status):
-    if device.health_check == health_check_status:
-        return
-    nodes = db_controller.get_storage_nodes()
-    for node in nodes:
-        if node.nvme_devices:
-            for dev in node.nvme_devices:
-                if dev.get_id() == device.get_id():
-                    old_status = dev.health_check
-                    dev.health_check = health_check_status
-                    node.write_to_db(db_store)
-                    device_events.device_health_check_change(
-                        dev, dev.health_check, old_status, caused_by="monitor")
-
-
-# configure logging
-logger_handler = logging.StreamHandler(stream=sys.stdout)
-logger_handler.setFormatter(logging.Formatter('%(asctime)s: %(levelname)s: %(message)s'))
-gelf_handler = GELFUDPHandler('0.0.0.0', constants.GELF_PORT)
-logger = logging.getLogger()
-logger.addHandler(gelf_handler)
-logger.addHandler(logger_handler)
-logger.setLevel(logging.DEBUG)
-
-# get DB controller
-db_store = kv_store.KVStore()
-db_controller = kv_store.DBController()
-
-logger.info("Starting health check service")
-while True:
-    cluster_id = ""
-    cl = db_controller.get_clusters()
-    if cl:
-        cluster_id = cl[0].get_id()
-
-    snodes = db_controller.get_storage_nodes()
-    if not snodes:
-        logger.error("storage nodes list is empty")
-
-    for snode in snodes:
-        logger.info("Node: %s, status %s", snode.get_id(), snode.status)
-
-        if snode.status not in [StorageNode.STATUS_ONLINE, StorageNode.STATUS_UNREACHABLE]:
-            logger.info(f"Node status is: {snode.status}, skipping")
-            continue
-
-        # 1- check node ping
-        ping_check = health_controller._check_node_ping(snode.mgmt_ip)
-        logger.info(f"Check: ping mgmt ip {snode.mgmt_ip} ... {ping_check}")
-
-        # 2- check node API
-        node_api_check = health_controller._check_node_api(snode.mgmt_ip)
-        logger.info(f"Check: node API {snode.mgmt_ip}:5000 ... {node_api_check}")
-
-        if snode.status == StorageNode.STATUS_OFFLINE:
-            set_node_health_check(snode, ping_check & node_api_check)
-            continue
-
-        # 3- check node RPC
-        node_rpc_check = health_controller._check_node_rpc(
-            snode.mgmt_ip, snode.rpc_port, snode.rpc_username, snode.rpc_password)
-        logger.info(f"Check: node RPC {snode.mgmt_ip}:{snode.rpc_port} ... {node_rpc_check}")
-
-        # 4- docker API
-        node_docker_check = health_controller._check_node_docker_api(snode.mgmt_ip)
-        logger.info(f"Check: node docker API {snode.mgmt_ip}:2375 ... {node_docker_check}")
-
-        is_node_online = ping_check and node_api_check and node_rpc_check and node_docker_check
-
-        health_check_status = is_node_online
-        if not node_rpc_check:
-            logger.info("Putting all devices to unavailable state because RPC check failed")
-            for dev in snode.nvme_devices:
-                if dev.io_error:
-                    logger.debug(f"Skipping Device action because of io_error {dev.get_id()}")
-                    continue
-                set_device_health_check(cluster_id, dev, False)
-        else:
-            logger.info(f"Node device count: {len(snode.nvme_devices)}")
-            node_devices_check = True
-            node_remote_devices_check = True
-
-            for dev in snode.nvme_devices:
-                if dev.io_error:
-                    logger.debug(f"Skipping Device check because of io_error {dev.get_id()}")
-                    continue
-                ret = health_controller.check_device(dev.get_id())
-                set_device_health_check(cluster_id, dev, ret)
-                if dev.status == dev.STATUS_ONLINE:
-                    node_devices_check &= ret
-
-            logger.info(f"Node remote device: {len(snode.remote_devices)}")
-            rpc_client = RPCClient(
-                snode.mgmt_ip, snode.rpc_port,
-                snode.rpc_username, snode.rpc_password,
-                timeout=5, retry=3)
-            for remote_device in snode.remote_devices:
-                ret = rpc_client.get_bdevs(remote_device.remote_bdev)
-                if ret:
-                    logger.info(f"Checking bdev: {remote_device.remote_bdev} ... ok")
-                else:
-                    logger.info(f"Checking bdev: {remote_device.remote_bdev} ... not found")
-                node_remote_devices_check &= bool(ret)
-
-            health_check_status = is_node_online and node_devices_check and node_remote_devices_check
-        set_node_health_check(snode, health_check_status)
-
-    time.sleep(constants.HEALTH_CHECK_INTERVAL_SEC)
-