diff options
Diffstat (limited to 'roles/os_zabbix/vars/template_os_linux.yml')
| -rw-r--r-- | roles/os_zabbix/vars/template_os_linux.yml | 134 |
1 files changed, 112 insertions, 22 deletions
diff --git a/roles/os_zabbix/vars/template_os_linux.yml b/roles/os_zabbix/vars/template_os_linux.yml index 3173c79b2..c6e557f12 100644 --- a/roles/os_zabbix/vars/template_os_linux.yml +++ b/roles/os_zabbix/vars/template_os_linux.yml @@ -10,17 +10,20 @@ g_template_os_linux: - key: kernel.all.cpu.wait.total applications: - Kernel - value_type: int + value_type: float + units: '%' - key: kernel.all.cpu.irq.hard applications: - Kernel - value_type: int + value_type: float + units: '%' - key: kernel.all.cpu.idle applications: - Kernel - value_type: int + value_type: float + units: '%' - key: kernel.uname.distro applications: @@ -35,7 +38,8 @@ g_template_os_linux: - key: kernel.all.cpu.irq.soft applications: - Kernel - value_type: int + value_type: float + units: '%' - key: kernel.all.load.15_minute applications: @@ -45,7 +49,8 @@ g_template_os_linux: - key: kernel.all.cpu.sys applications: - Kernel - value_type: int + value_type: float + units: '%' - key: kernel.all.load.5_minute applications: @@ -55,7 +60,8 @@ g_template_os_linux: - key: kernel.all.cpu.nice applications: - Kernel - value_type: int + value_type: float + units: '%' - key: kernel.all.load.1_minute applications: @@ -75,7 +81,8 @@ g_template_os_linux: - key: kernel.all.cpu.user applications: - Kernel - value_type: int + value_type: float + units: '%' - key: kernel.uname.machine applications: @@ -90,7 +97,8 @@ g_template_os_linux: - key: kernel.all.cpu.steal applications: - Kernel - value_type: int + value_type: float + units: '%' - key: kernel.all.pswitch applications: @@ -180,38 +188,105 @@ g_template_os_linux: multiplier: 1024 units: B - # Disk items - - key: filesys.full.xvda2 + zdiscoveryrules: + - name: disc.filesys + key: disc.filesys + lifetime: 1 + description: "Dynamically register the filesystems" + + - name: disc.disk + key: disc.disk + lifetime: 1 + description: "Dynamically register disks on a node" + + - name: disc.network + key: disc.network + lifetime: 1 + description: "Dynamically register network interfaces on a node" + + zitemprototypes: + - discoveryrule_key: disc.filesys + name: "disc.filesys.full.{#OSO_FILESYS}" + key: "disc.filesys.full[{#OSO_FILESYS}]" + value_type: float + description: "PCP filesys.full option. This is the percent full returned from pcp filesys.full" applications: - Disk + + - discoveryrule_key: disc.filesys + name: "Percentage of used inodes on {#OSO_FILESYS}" + key: "disc.filesys.inodes.pused[{#OSO_FILESYS}]" value_type: float + description: "PCP derived value of percentage of used inodes on a filesystem." + applications: + - Disk - - key: filesys.full.xvda3 + - discoveryrule_key: disc.disk + name: "TPS (IOPS) for disk {#OSO_DISK}" + key: "disc.disk.tps[{#OSO_DISK}]" + value_type: int + description: "PCP disk.dev.totals metric measured over a period of time. This shows how many disk transactions per second the disk is using" applications: - Disk + + - discoveryrule_key: disc.disk + name: "Percent Utilized for disk {#OSO_DISK}" + key: "disc.disk.putil[{#OSO_DISK}]" value_type: float + description: "PCP disk.dev.avactive metric measured over a period of time. This is the '%util' in the iostat command" + applications: + - Disk - ztriggers: - - name: 'Filesystem: / has less than 10% free on {HOST.NAME}' - expression: '{Template OS Linux:filesys.full.xvda2.last()}>90' - url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc' - priority: warn + - discoveryrule_key: disc.network + name: "Bytes per second IN on network interface {#OSO_NET_INTERFACE}" + key: "disc.network.in.bytes[{#OSO_NET_INTERFACE}]" + value_type: int + units: B + delta: 1 + description: "PCP network.interface.in.bytes metric. This is setup as a delta in Zabbix to measure the speed per second" + applications: + - Network - - name: 'Filesystem: / has less than 5% free on {HOST.NAME}' - expression: '{Template OS Linux:filesys.full.xvda2.last()}>95' + - discoveryrule_key: disc.network + name: "Bytes per second OUT on network interface {#OSO_NET_INTERFACE}" + key: "disc.network.out.bytes[{#OSO_NET_INTERFACE}]" + value_type: int + units: B + delta: 1 + description: "PCP network.interface.out.bytes metric. This is setup as a delta in Zabbix to measure the speed per second" + applications: + - Network + + ztriggerprototypes: + - name: 'Filesystem: {#OSO_FILESYS} has less than 10% free disk space on {HOST.NAME}' + expression: '{Template OS Linux:disc.filesys.full[{#OSO_FILESYS}].last()}>90' url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc' priority: high - - name: 'Filesystem: /var has less than 10% free on {HOST.NAME}' - expression: '{Template OS Linux:filesys.full.xvda3.last()}>90' + # This has a dependency on the previous trigger + # Trigger Prototypes do not work in 2.4. They will work in Zabbix 3.0 + - name: 'Filesystem: {#OSO_FILESYS} has less than 15% free disk space on {HOST.NAME}' + expression: '{Template OS Linux:disc.filesys.full[{#OSO_FILESYS}].last()}>85' url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc' priority: warn + dependencies: + - 'Filesystem: {#OSO_FILESYS} has less than 10% free disk space on {HOST.NAME}' - - name: 'Filesystem: /var has less than 5% free on {HOST.NAME}' - expression: '{Template OS Linux:filesys.full.xvda3.last()}>95' + - name: 'Filesystem: {#OSO_FILESYS} has less than 5% free inodes on {HOST.NAME}' + expression: '{Template OS Linux:disc.filesys.inodes.pused[{#OSO_FILESYS}].last()}>95' url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc' priority: high + # This has a dependency on the previous trigger + # Trigger Prototypes do not work in 2.4. They will work in Zabbix 3.0 + - name: 'Filesystem: {#OSO_FILESYS} has less than 10% free inodes on {HOST.NAME}' + expression: '{Template OS Linux:disc.filesys.inodes.pused[{#OSO_FILESYS}].last()}>90' + url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_filesys_full.asciidoc' + priority: warn + dependencies: + - 'Filesystem: {#OSO_FILESYS} has less than 5% free inodes on {HOST.NAME}' + + ztriggers: - name: 'Too many TOTAL processes on {HOST.NAME}' expression: '{Template OS Linux:proc.nprocs.last()}>5000' url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_proc.asciidoc' @@ -222,3 +297,18 @@ g_template_os_linux: url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_memory.asciidoc' priority: warn description: 'Alert on less than 30MegaBytes. This is 30 Million Bytes. 30000 KB x 1024' + + # CPU Utilization # + - name: 'CPU idle less than 5% on {HOST.NAME}' + expression: '{Template OS Linux:kernel.all.cpu.idle.max(#5)}<5' + url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_cpu_idle.asciidoc' + priority: average + description: 'CPU is less than 5% idle' + + - name: 'CPU idle less than 10% on {HOST.NAME}' + expression: '{Template OS Linux:kernel.all.cpu.idle.max(#5)}<10' + url: 'https://github.com/openshift/ops-sop/blob/master/V3/Alerts/check_cpu_idle.asciidoc' + priority: average + description: 'CPU is less than 10% idle' + dependencies: + - 'CPU idle less than 5% on {HOST.NAME}' |