Introduction
Este documento descreve como identificar um Unified Computing System (UCS) e verificar entradas de falhas nele na Cloud Native Deployment Platform (CNDP).
Informações de Apoio
Os alertas relacionados ao hardware são relatados no ambiente de execução comum (CEE) do Gerente de cluster (CM) do Ultra Cloud Core Subscriber Microservices Infrastructure (SMI). Os Kubernetes (K8s), docker, etc., são relatados no CM virtual IP (VIP).
Caution: Consulte Projeto de rede e Questionário de informações do cliente (CIQ) para verificar os IPs.
Problema
O erro "Alarme do equipamento" é relatado em mostrar alertas.
- Efetue login no CM-CEE, execute o comando show alert ative detail e show alert history summary para exibir todos os alertas ativos e históricos.
- Observe o IP do servidor relatado no alerta.
[lab-deployer/labceec01] cee# show alerts active detail
alerts active detail server-alert 9c367ce5ee48
severity major
type "Equipment Alarm"
startsAt 2021-10-27T17:10:37.025Z
source 10.10.10.10
summary "DDR4_P1_C1_ECC: DIMM 5 is inoperable : Check or replace DIMM"
labels [ "alertname: server-alert" "cluster: cr-chr-deployer" "description: DDR4_P1_C1_ECC: DIMM 5 is inoperable : Check or replace DIMM" "fault_id: sys/rack-unit-1/board/memarray-1/mem-5/fault-F0185" "id: 134219020" "monitor: prometheus" "replica: cr-chr-deployer" "server: 10.10.10.10" "severity: major" ]
annotations [ "dn: cr-chr-deployer/10.10.10.10/sys/rack-unit-1/board/memarray-1/mem-5/fault-F0185/134219020" "summary: DDR4_P1_C1_ECC: DIMM 5 is inoperable : Check or replace DIMM" "type: Equipment Alarm" ]
[lab-deployer/labceec01] cee# show alerts history summary
NAME UID SEVERITY STARTS AT DURATION SOURCE SUMMARY
---------------------------------------------------------------------------------------------
vm-alive f6a65030b593 minor 09-02T10:28:28 1m40s 10-192-0-13 labd0123 is alive.
vm-error 3a6d840e3eda major 09-02T10:27:18 1m 10-192-0-13 labd0123 is down.
vm-alive 49b2c1941dc6 minor 09-02T10:25:38 1m40s 10-192-0-14 labd0123 is alive.
Solução
Identifique os serviços (contêineres) e/ou a máquina virtual (VM) ou a máquina virtual baseada em kernel (KVM) hospedada no servidor no SMI, execute o comando show running-config e encontre a configuração para o IP do servidor.
- Faça login no CM VIP (nome de usuário: usuário de nuvem)
- Obtenha o IP do Centro OPS para o namespace smi-cm
- Faça login no Centro OPS e verifique a configuração do cluster
- Identificar nós e VMs que são executados no servidor
cloud-user@lab-deployer-cm-primary:~$ kubectl get svc -n smi-cm
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
cluster-files-offline-smi-cluster-deployer ClusterIP 10.102.200.178 <none> 8080/TCP 98d
iso-host-cluster-files-smi-cluster-deployer ClusterIP 10.102.100.208 192.168.1.102 80/TCP 98d
iso-host-ops-center-smi-cluster-deployer ClusterIP 10.102.200.73 192.168.1.102 3001/TCP 98d
netconf-ops-center-smi-cluster-deployer ClusterIP 10.102.100.207 192.168.184.193 3022/TCP,22/TCP 98d
ops-center-smi-cluster-deployer ClusterIP 10.10.20.20 <none> 8008/TCP,2024/TCP,2022/TCP,7681/TCP,3000/TCP,3001/TCP 98d
squid-proxy-node-port NodePort 10.102.60.114 <none> 3128:32261/TCP 98d
cloud-user@lab-deployer-cm-primary:~$ ssh -p 2024 admin@10.10.20.20
admin@10.10.20.20's password:
Welcome to the Cisco SMI Cluster Deployer on lab-deployer-cm-primary
Copyright © 2016-2020, Cisco Systems, Inc.
All rights reserved.
admin connected from 192.168.1.100 using ssh on ops-center-smi-cluster-deployer-7848c69844-xzdw6
[lab-deployer-cm-primary] SMI Cluster Deployer# show running-config clusters
Exemplo de saída para contêineres
Neste exemplo, o servidor é usado pelo nó primary-1.
[lab-deployer-cm-primary] SMI Cluster Deployer# show running-config clusters lab01-smf nodes primary-1
clusters lab01-smf
nodes primary-1
maintenance false
k8s node-type primary
k8s ssh-ip 10.192.10.22
k8s sshd-bind-to-ssh-ip true
k8s node-ip 10.192.10.22
k8s node-labels smi.cisco.com/node-type oam
exit
k8s node-labels smi.cisco.com/node-type-1 proto
exit
ucs-server cimc user admin
ucs-server cimc ip-address 10.10.10.10
Exemplo de saída para VMs
O servidor pode ser usado para a VM baseada em KVM.
Neste exemplo, o servidor tem UPFs (User Plane Functions, Funções do Plano do Usuário) - upf1 e upf2.
[lab-deployer-cm-primary] SMI Cluster Deployer# show running-config clusters lab01-upf nodes labupf
clusters lab01-upf
nodes labupf
maintenance false
ssh-ip 10.192.30.7
type kvm
vms upf1
upf software lab...
...
type upf
exit
vms upf2
upf software lab...
...
type upf
exit
ucs-server cimc user admin
...
ucs-server cimc ip-address 10.10.10.10
...
exit
SSH para o host UCS
Conecte-se ao host UCS e verifique as entradas de falha com falha de escopo, mostre as entradas de falha e mostre o histórico de falhas.
labucs111-cmp1-11 /fault # show fault-entries
Time Severity Description ------------------------- ------------- ---------------------------------------
2021-03-26T10:10:10 major "DDR4_P1_C1_ECC: DIMM 19 is inoperable : Check or replace DIMM"
LABCP0222-Server22-02 /fault # show fault-history
Time Severity Source Cause Description
------------------- ------------- --------------- ------------------------- ----------------------------------------
2021 Dec 10 02:02:02 UTC info %CIMC EQUIPMENT_INOPERABLE "[F0174][cleared][equipment-inoperable][sys/rack-unit-1/board] IERR: A catastrophic fault has occurred on one of the processors: Cleared "
2021 Dec 1 01:01:01 UTC critical %CIMC EQUIPMENT_INOPERABLE "[F0174][critical][equipment-inoperable][sys/rack-unit-1/board] IERR: A catastrophic fault has occurred on one of the processors: Please check the processor's status. "
Feedback