170 lines
		
	
	
		
			6.7 KiB
		
	
	
	
		
			Python
		
	
	
	
	
	
			
		
		
	
	
			170 lines
		
	
	
		
			6.7 KiB
		
	
	
	
		
			Python
		
	
	
	
	
	
| #!/usr/bin/env python3
 | |
| 
 | |
| # disk.py - PVC Monitoring example plugin for disk (system + OSD)
 | |
| # Part of the Parallel Virtual Cluster (PVC) system
 | |
| #
 | |
| #    Copyright (C) 2018-2022 Joshua M. Boniface <joshua@boniface.me>
 | |
| #
 | |
| #    This program is free software: you can redistribute it and/or modify
 | |
| #    it under the terms of the GNU General Public License as published by
 | |
| #    the Free Software Foundation, version 3.
 | |
| #
 | |
| #    This program is distributed in the hope that it will be useful,
 | |
| #    but WITHOUT ANY WARRANTY; without even the implied warranty of
 | |
| #    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 | |
| #    GNU General Public License for more details.
 | |
| #
 | |
| #    You should have received a copy of the GNU General Public License
 | |
| #    along with this program.  If not, see <https://www.gnu.org/licenses/>.
 | |
| #
 | |
| ###############################################################################
 | |
| 
 | |
| # This script provides an example of a PVC monitoring plugin script. It will create
 | |
| # a simple plugin to check the system and OSD disks for errors and faults and return
 | |
| # a health delta corresponding to severity.
 | |
| 
 | |
| # This script can thus be used as an example or reference implementation of a
 | |
| # PVC monitoring pluginscript and expanded upon as required.
 | |
| 
 | |
| # A monitoring plugin script must implement the class "MonitoringPluginScript" which
 | |
| # extends "MonitoringPlugin", providing the 3 functions indicated. Detailed explanation
 | |
| # of the role of each function is provided in context of the example; see the other
 | |
| # examples for more potential uses.
 | |
| 
 | |
| # WARNING:
 | |
| #
 | |
| # This script will run in the context of the node daemon keepalives as root.
 | |
| # DO NOT install untrusted, unvetted plugins under any circumstances.
 | |
| 
 | |
| 
 | |
| # This import is always required here, as MonitoringPlugin is used by the
 | |
| # MonitoringPluginScript class
 | |
| from pvcnoded.objects.MonitoringInstance import MonitoringPlugin
 | |
| 
 | |
| 
 | |
| # A monitoring plugin script must always expose its nice name, which must be identical to
 | |
| # the file name
 | |
| PLUGIN_NAME = "disk"
 | |
| 
 | |
| 
 | |
| # The MonitoringPluginScript class must be named as such, and extend MonitoringPlugin.
 | |
| class MonitoringPluginScript(MonitoringPlugin):
 | |
|     def setup(self):
 | |
|         """
 | |
|         setup(): Perform special setup steps during node daemon startup
 | |
| 
 | |
|         This step is optional and should be used sparingly.
 | |
| 
 | |
|         If you wish for the plugin to not load in certain conditions, do any checks here
 | |
|         and return a non-None failure message to indicate the error.
 | |
|         """
 | |
| 
 | |
|         from daemon_lib.common import run_os_command
 | |
|         from json import loads
 | |
| 
 | |
|         _, _all_disks, _ = run_os_command("lsblk --json --paths --include 8,259")
 | |
|         try:
 | |
|             all_disks = loads(_all_disks)
 | |
|         except Exception as e:
 | |
|             return f"Error loading lsblk JSON: {e}"
 | |
| 
 | |
|         disk_details = list()
 | |
| 
 | |
|         def get_smartinfo(disk, extra_opt=""):
 | |
|             _, _smart_info, _ = run_os_command(f"smartctl --info --json {extra_opt} {disk}")
 | |
|             try:
 | |
|                 smart_info = loads(_smart_info)
 | |
|             except Exception as e:
 | |
|                 return None
 | |
| 
 | |
|             return smart_info
 | |
| 
 | |
|         for disk in [disk["name"] for disk in all_disks['blockdevices']]:
 | |
|             extra_opt = ""
 | |
|             smart_info = get_smartinfo(disk)
 | |
|             if smart_info is None or smart_info["smartctl"]["exit_status"] > 1:
 | |
|                 continue
 | |
|             elif smart_info["smartctl"]["exit_status"] == 1:
 | |
|                 if "requires option" in smart_info["smartctl"]["messages"][0]["string"]:
 | |
|                     extra_opt = smart_info["smartctl"]["messages"][0]["string"].split("'")[1].replace('N','0')
 | |
|                     smart_info = get_smartinfo(disk, extra_opt)
 | |
|                     if smart_info is None or smart_info["smartctl"]["exit_status"] > 0:
 | |
|                         continue
 | |
|                 else:
 | |
|                     continue
 | |
| 
 | |
|             disk_type = smart_info["device"]["type"]
 | |
| 
 | |
|             disk_details.append((disk, extra_opt, disk_type))
 | |
| 
 | |
|         self.disk_details = disk_details
 | |
| 
 | |
| 
 | |
|     def run(self, coordinator_state=None):
 | |
|         """
 | |
|         run(): Perform the check actions and return a PluginResult object
 | |
| 
 | |
|         The {coordinator_state} can be used to check if this is a "primary" coordinator, "secondary" coordinator, or "client" (non-coordinator)
 | |
|         """
 | |
| 
 | |
|         # Re-run setup each time to ensure the disk details are current
 | |
|         self.setup()
 | |
| 
 | |
|         # Run any imports first
 | |
|         from daemon_lib.common import run_os_command
 | |
|         from json import loads
 | |
| 
 | |
|         health_delta = 0
 | |
|         messages = list()
 | |
| 
 | |
|         for _disk in self.disk_details:
 | |
|             disk = _disk[0]
 | |
|             extra_opt = _disk[1]
 | |
|             disk_type = _disk[2]
 | |
| 
 | |
|             _, _smart_info, _ = run_os_command(f"smartctl --all --json {extra_opt} {disk}")
 | |
|             try:
 | |
|                 smart_info = loads(_smart_info)
 | |
|             except Exception as e:
 | |
|                 health_delta += 10
 | |
|                 messages.append(f"{disk} failed to load SMART data")
 | |
|                 continue
 | |
| 
 | |
|             if disk_type == 'nvme':
 | |
|                 for attribute in smart_info.get('nvme_smart_health_information_log', {}).items():
 | |
|                     if attribute[0] == "critical_warning" and attribute[1] > 0:
 | |
|                         health_delta += 10
 | |
|                         messages.append(f"{disk} critical warning value {attribute[1]}")
 | |
|                     if attribute[0] == "media_errors" and attribute[1] > 0:
 | |
|                         health_delta += 10
 | |
|                         messages.append(f"{disk} media errors value {attribute[1]}")
 | |
|                     if attribute[0] == "percentage_used" and attribute[1] > 90:
 | |
|                         health_delta += 10
 | |
|                         messages.append(f"{disk} percentage used value {attribute[1]}%")
 | |
|             else:
 | |
|                 for attribute in smart_info.get('ata_smart_attributes', {}).get('table', []):
 | |
|                     if attribute["when_failed"]:
 | |
|                         health_delta += 10
 | |
|                         messages.append(f"{disk} attribute {attribute['name']} value {attribute['raw']['value']}")
 | |
| 
 | |
|         if len(messages) < 1:
 | |
|             messages.append(f"All {len(self.disk_details)} checked disks report OK: {', '.join([disk[0] for disk in self.disk_details])}")
 | |
| 
 | |
|         # Set the health delta in our local PluginResult object
 | |
|         self.plugin_result.set_health_delta(health_delta)
 | |
| 
 | |
|         # Set the message in our local PluginResult object
 | |
|         self.plugin_result.set_message(', '.join(messages))
 | |
| 
 | |
|         # Return our local PluginResult object
 | |
|         return self.plugin_result
 | |
| 
 | |
|     def cleanup(self):
 | |
|         """
 | |
|         cleanup(): Perform special cleanup steps during node daemon termination
 | |
| 
 | |
|         This step is optional and should be used sparingly.
 | |
|         """
 | |
| 
 | |
|         pass
 |