[checkmk-commits] Check kubernetes Pods custom metrics, cpu, memory, fs

Module: check_mk Branch: master Commit: be297ace6a79b4f480d7bfa4c75fcba59f16889e URL: http://git.mathias-kettner.de/git/?p=check_mk.git;a=commit;h=be297ace6a79b4… Author: Óscar Nájera &lt;on(a)mathias-kettner.de&gt; Date: Thu Dec 20 18:04:58 2018 +0100 Check kubernetes Pods custom metrics, cpu, memory, fs CMK-1476 Change-Id: I8d136d94cd405cb6a86e549e676c844cf691c332 --- checkman/k8s_pods_cpu | 13 ++++ checkman/k8s_pods_fs | 13 ++++ checkman/k8s_pods_memory | 13 ++++ checks/k8s_pods_cm | 128 +++++++++++++++++++++++++++++++++ cmk/special_agents/agent_kubernetes.py | 10 +-- 5 files changed, 172 insertions(+), 5 deletions(-) diff --git a/checkman/k8s_pods_cpu b/checkman/k8s_pods_cpu new file mode 100644 index 0000000..684b830 --- /dev/null +++ b/checkman/k8s_pods_cpu @@ -0,0 +1,13 @@ +title: Kubernetes pods CPU resources +agents: linux +catalog: containerization/kubernetes +license: GPL +distribution: check_mk +description: + This check monitors the pods' cpu usage inside a namespace of a Kubernetes node. + +item: + The namespace whose cpu resources are monitored + +inventory: + One service per namespace is created. diff --git a/checkman/k8s_pods_fs b/checkman/k8s_pods_fs new file mode 100644 index 0000000..634ee14 --- /dev/null +++ b/checkman/k8s_pods_fs @@ -0,0 +1,13 @@ +title: Kubernetes pods Filesystem resources +agents: linux +catalog: containerization/kubernetes +license: GPL +distribution: check_mk +description: + This check monitors the pods' filesystem usage inside a namespace of a Kubernetes node. + +item: + The namespace whose filesystem resources are monitored + +inventory: + One service per namespace is created. diff --git a/checkman/k8s_pods_memory b/checkman/k8s_pods_memory new file mode 100644 index 0000000..bd2569d --- /dev/null +++ b/checkman/k8s_pods_memory @@ -0,0 +1,13 @@ +title: Kubernetes pods memory resources +agents: linux +catalog: containerization/kubernetes +license: GPL +distribution: check_mk +description: + This check monitors the pods' memory usage inside a namespace of a Kubernetes node. + +item: + The namespace whose memory resources are monitored + +inventory: + One service per namespace is created. diff --git a/checks/k8s_pods_cm b/checks/k8s_pods_cm new file mode 100644 index 0000000..037969e --- /dev/null +++ b/checks/k8s_pods_cm @@ -0,0 +1,128 @@ +#!/usr/bin/python +# -*- encoding: utf-8; py-indent-offset: 4 -*- +# +------------------------------------------------------------------+ +# | ____ _ _ __ __ _ __ | +# | / ___| |__ ___ ___| | __ | \/ | |/ / | +# | | | | '_ \ / _ \/ __| |/ / | |\/| | ' / | +# | | |___| | | | __/ (__| < | | | | . \ | +# | \____|_| |_|\___|\___|_|\_\___|_| |_|_|\_\ | +# | | +# | Copyright Mathias Kettner 2018 mk(a)mathias-kettner.de | +# +------------------------------------------------------------------+ +# +# This file is part of Check_MK. +# The official homepage is at http://mathias-kettner.de/check_mk. +# +# check_mk 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 in version 2. check_mk is distributed +# in the hope that it will be useful, but WITHOUT ANY WARRANTY; with- +# out even the implied warranty of MERCHANTABILITY or FITNESS FOR A +# PARTICULAR PURPOSE. See the GNU General Public License for more de- +# tails. You should have received a copy of the GNU General Public +# License along with GNU Make; see the file COPYING. If not, write +# to the Free Software Foundation, Inc., 51 Franklin St, Fifth Floor, +# Boston, MA 02110-1301 USA. + +from cmk.special_agents.agent_kubernetes import parse_frac_prefix + + +def parse_mul_prefix(value): + # type: (str) -> int + if value.endswith('k'): + return 1000 * int(value[:-1]) + return int(value) + + +@get_parsed_item_data +def check_k8s_pods_memory(item, _no_params, metrics): + + rss = 0 + swap = 0 + usage_bytes = 0 + max_usage_bytes = 0 + + for pod_data in metrics: + rss += parse_mul_prefix(pod_data['metrics']['memory_rss']) + swap += parse_mul_prefix(pod_data['metrics']['memory_swap']) + usage_bytes += parse_mul_prefix(pod_data['metrics']['memory_usage_bytes']) + max_usage_bytes += parse_mul_prefix(pod_data['metrics']['memory_max_usage_bytes']) + + yield check_levels(rss, 'rss', _no_params, human_readable_func=get_bytes_human_readable) + yield check_levels(swap, 'swap', _no_params, human_readable_func=get_bytes_human_readable) + yield check_levels( + usage_bytes, 'usage_bytes', _no_params, human_readable_func=get_bytes_human_readable) + yield check_levels( + max_usage_bytes, + 'max_usage_bytes', + _no_params, + human_readable_func=get_bytes_human_readable) + + +check_info['k8s_pods_memory'] = { + 'parse_function': parse_k8s, + 'inventory_function': discover(), + 'check_function': check_k8s_pods_memory, + 'service_description': 'Memory usage for Pods in %s namespace', + 'includes': ['k8s.include'], +} + + +@get_parsed_item_data +def check_k8s_pods_cpu(item, _no_params, metrics): + + system = 0 + user = 0 + usage = 0 + + for pod_data in metrics: + system += parse_frac_prefix(pod_data['metrics']['cpu_system']) + user += parse_frac_prefix(pod_data['metrics']['cpu_user']) + usage += parse_frac_prefix(pod_data['metrics']['cpu_usage']) + + yield check_levels(system, 'system', _no_params) + yield check_levels(user, 'user', _no_params) + yield check_levels(usage, 'usage', _no_params) + + +check_info['k8s_pods_cpu'] = { + 'parse_function': parse_k8s, + 'inventory_function': discover(), + 'check_function': check_k8s_pods_cpu, + 'service_description': 'CPU usage for Pods in %s namespace', + 'includes': ['k8s.include'], +} + + +@get_parsed_item_data +def check_k8s_pods_fs(item, _no_params, metrics): + + inodes = 0 + reads = 0 + writes = 0 + limit_bytes = 0 + usage_bytes = 0 + + for pod_data in metrics: + inodes += parse_frac_prefix(pod_data['metrics']['fs_inodes']) + reads += parse_frac_prefix(pod_data['metrics']['fs_reads']) + writes += parse_frac_prefix(pod_data['metrics']['fs_writes']) + limit_bytes += parse_mul_prefix(pod_data['metrics']['fs_limit_bytes']) + usage_bytes += parse_mul_prefix(pod_data['metrics']['fs_usage_bytes']) + + yield check_levels(inodes, 'inodes', _no_params) + yield check_levels(reads, 'reads', _no_params) + yield check_levels(writes, 'writes', _no_params) + yield check_levels( + limit_bytes, 'limit_bytes', _no_params, human_readable_func=get_bytes_human_readable) + yield check_levels( + usage_bytes, 'usage_bytes', _no_params, human_readable_func=get_bytes_human_readable) + + +check_info['k8s_pods_fs'] = { + 'parse_function': parse_k8s, + 'inventory_function': discover(), + 'check_function': check_k8s_pods_fs, + 'service_description': 'FS usage for Pods in %s namespace', + 'includes': ['k8s.include'], +} diff --git a/cmk/special_agents/agent_kubernetes.py b/cmk/special_agents/agent_kubernetes.py index 5efe7a0..03951d3 100644 --- a/cmk/special_agents/agent_kubernetes.py +++ b/cmk/special_agents/agent_kubernetes.py @@ -95,7 +95,7 @@ def setup_logging(verbosity): logging.basicConfig(level=lvl, format=fmt) -def parse_cpu(value): +def parse_frac_prefix(value): # type: (str) -> float if value.endswith('m'): return 0.001 * float(value[:-1]) @@ -199,11 +199,11 @@ class Node(Metadata): return view capacity, allocatable = self._status.capacity, self._status.allocatable if capacity: - view['capacity']['cpu'] += parse_cpu(capacity.get('cpu', '0.0')) + view['capacity']['cpu'] += parse_frac_prefix(capacity.get('cpu', '0.0')) view['capacity']['memory'] += parse_memory(capacity.get('memory', '0.0')) view['capacity']['pods'] += int(capacity.get('pods', '0')) if allocatable: - view['allocatable']['cpu'] += parse_cpu(allocatable.get('cpu', '0.0')) + view['allocatable']['cpu'] += parse_frac_prefix(allocatable.get('cpu', '0.0')) view['allocatable']['memory'] += parse_memory(allocatable.get('memory', '0.0')) view['allocatable']['pods'] += int(allocatable.get('pods', '0')) return view @@ -253,14 +253,14 @@ class Pod(Metadata): continue limits = resources.limits if limits: - view['limits']['cpu'] += parse_cpu(limits.get('cpu', 'inf')) + view['limits']['cpu'] += parse_frac_prefix(limits.get('cpu', 'inf')) view['limits']['memory'] += parse_memory(limits.get('memory', 'inf')) else: view['limits']['cpu'] += float('inf') view['limits']['memory'] += float('inf') requests = resources.requests if requests: - view['requests']['cpu'] += parse_cpu(requests.get('cpu', '0.0')) + view['requests']['cpu'] += parse_frac_prefix(requests.get('cpu', '0.0')) view['requests']['memory'] += parse_memory(requests.get('memory', '0.0')) return view