Basic Kubectl Commands

Some of the basic kubectl commands apply to all k8s objects

Namespaces

• used to group objects in the cluster
• each namespace acts like a folder holding a set of objects
• kubectl works with the default namespace by default
• —namespace can be passed to specify a different one

Context

• can be used to change the default namespace, manage different clusters or different users for authenticating to them
• To change the namespace to abc, run: $ kubectl config set-context my-context –namespace=abc
• This command just creates a context. To run it, use: $ kubectl config use-context my-context
• This records the change in the kubectl config file while is usually located at $HOME/.kube/config. This file also stores information related to finding and authenticating to the cluster.

Viewing Kubernetes API Objects

• Everything is represented by a RESTful resource, called objects.
• Each object has a unique HTTP endpoint scoped using namespaces.
• kubectl uses these endpoints to fetch the objects
• Viewing resources in the current namespace: kubectl get
• Viewing a specific resource: kubectl get  Use describe instead of `get“ for more detailed information about the object
• To get more info than normally displayed, use -o wide flag
• To get raw JSON or YAML objects from the API server, use -o json, -o yaml.

Creating, Updating, and Destroying Kubernetes Objects

• k8s objects are represented using YAML or JSON files (each object has its own JSON/YAML file?)
• used to manipulate objects on the server
• To create an object described using obj.yaml: kubectl apply -f obj.yaml K8s automatically infers the type, it doesn’t need to be specified.

Labeling and Annotating Objects

• used to tag objects
• kubectl label/annotate pods bar color=red adds the color=red label to a pod called bar.
• To remove a label: kubectl label pods bar -color

Debugging Commands

To see logs for a container: kubectl logs

• Use -c to choose a particular container
• Use -f to stream logs to terminal
• kubectl exec -it — bash provides an interactive shell within the container.

Troubleshoot your Sitecore environment using these Application Insightsquery

Sitecore sends what you would expect (errors, warnings, info logs) to Application Insights from a log perspective. In Application Insights, you can identify error and warning messages, identify messages from specific roles and Sitecore instances, and search for specific log messages.

For further information related to Sitecore logs, you can refer this Sitecore document: Analyze Sitecore logs with Application Insights and Sitecore KB Access logs and diagnostics data in Sitecore XP on Azure Web Apps

Below are the necessary Application Insights queries which might help you during the troubleshooting sessions of your Sitecore application:

Availability Results:

availabilityResults
| where timestamp > ago(7d)
| summarize avg(toint(success)) * 100 by bin(timestamp, 1h), name
| order by timestamp asc
| render timechart

Get unique users per day past 90 days:

union pageViews,customEvents
| where timestamp > ago(90d)
| summarize Users=dcount(user_Id) by bin(timestamp, 1d)
| render timechart

Get 500 errors in dependencies line graph past 90 days:

dependencies
| where timestamp > ago(90d)
| where resultCode in (\"500\", \"502\")
| summarize count() by bin(timestamp, 1d)
| render timechart

Get 404 errors in requests line graph past 90 days:

requests
| where timestamp > ago(90d)
| where client_Type != \"Browser\"
| where resultCode == 404
| summarize failedCount=sumif(itemCount, success == false) by bin(timestamp, 1d), resultCode
| extend request=\'Overall\'
| render timechart

Get server exceptions line graph past 90 days:

exceptions
| where timestamp > ago(90d)
| where client_Type != \"Browser\"
| summarize count() by bin(timestamp, 1d)
| render timechart

Get all exceptions bar chart past 90 days:

exceptions
| where timestamp > ago(90d)
| summarize count = sum(itemCount) by bin(timestamp, 4h), problemId, tostring(customDimensions.InstanceName)
| order by timestamp desc
| render barchart

Clientside and react errors line graph past 90 days:

exceptions
| where timestamp > ago(90d)
| where client_Type == \"Browser\"
| summarize count() by bin(timestamp, 1d)
| render timechart

Get WARN total count line graph past 90 days:

traces
| where timestamp > ago(90d)
| where severityLevel == 2
| parse message with * \"WARN\" TrimmedMessage
| summarize count() by bin(timestamp, 1d)
| render timechart

Get Trace ERROR total count line graph past 90 days:

traces
| where timestamp > ago(90d)
| where severityLevel == 3
| parse message with * \"ERROR\" TrimmedMessage
| summarize count() by bin(timestamp, 1d)
| render timechart

Get pageviews line graph past 90 days:

pageViews
| where timestamp > ago(90d)
| summarize count() by bin(timestamp, 1d)
| render timechart

Check your CD(s) for errors:

traces
| where timestamp > ago(1d)
| where message contains \"ERROR\"
| where (customDimensions).Role contains \"CD\"
| project timestamp, message, (customDimensions).InstanceName
| order by timestamp desc

To create a better overview on ERRORS you could extend the timespan and render the query using, for example, a timechart.

traces
| where timestamp > ago(144h)
| where message contains \"ERROR\"
| where (customDimensions).Role contains :CD\"
| extend localTime = timestamp + 1h
| summarize Amount_Of_Errors=dcount(message) by bin(localTime, 1h)
| order by localTime asc
| render timechart

Check your CM(s) for errors:

traces
| where timestamp > ago(1d)
| where message contains \"ERROR\"
| where (customDimensions).Role contains \"CM\"
| project timestamp, message, (customDimensions).InstanceName
| order by timestamp desc

Check all instances for errors:

traces
| where timestamp > ago(1d)
| extend localTime = timestamp + 1h
| where message contains \"ERROR\"
| where message !contains \"INFO\" and message !contains \"WARN\"
| project localTime, message, (customDimensions).InstanceName
| order by localTime desc

Check your environment on exceptions:

exceptions
| where timestamp > ago(14d)
| extend localTime = timestamp + 1h
| project localTime, (customDimensions).MachineName, assembly, outerMessage

Check your environment on server side exceptions:

exceptions
| where timestamp > ago(7d)
| where client_Type != \"Browser\"
| summarize Total = count() by problemId
| order by Total
| project Total, problemId
| take 10

Get server exceptions total count:

exceptions
| where timestamp > ago(7d)
| where client_Type != \"Browser\"
| summarize count()

Get client side exceptions:

exceptions
| where timestamp > ago(7d)
| where client_Type == \"Browser\"
| summarize Total = count() by problemId
| order by Total
| project Total, problemId
| take 10

Get client side exceptions total count:

exceptions
| where timestamp > ago(7d)
| where client_Type == \"Browser\"
| summarize count()

Get WARN count by occurrence:

traces
| where timestamp > ago(7d)
| where severityLevel == 2
| parse message with * \"WARN\" TrimmedMessage
| project TrimmedMessage, customDimensions.Role
| summarize Total = count() by TrimmedMessage, tostring(customDimensions_Role)
| order by Total
| project Total, customDimensions_Role, TrimmedMessage
| take 10

Get WARN total count:

traces
| where timestamp > ago(7d)
| where severityLevel == 2
| parse message with * \"WARN\" TrimmedMessage
| summarize count()

Get Trace ERROR count by occurrence:

traces
| where timestamp > ago(7d)
| where severityLevel == 3
| parse message with * \"ERROR\" TrimmedMessage
| project TrimmedMessage, customDimensions.Role
| summarize Total = count() by TrimmedMessage, tostring(customDimensions_Role)
| order by Total
| project Total, customDimensions_Role, TrimmedMessage
| take 10

Get TRACE ERROR total count:

traces
| where timestamp > ago(7d)
| where severityLevel == 3
| parse message with * \"ERROR\" TrimmedMessage
| summarize count()

Check your environment on log entry’s with severity > 3:

traces |
where severityLevel >= 3

What are the most hit Urls:

requests
| where timestamp > ago(14d)
| summarize count() by url
| order by count_ desc | project url, count_

Give me all 404 status codes:

requests
| where timestamp > ago(14d)
| extend localTime = timestamp + 1h
| where resultCode == \"404\"
| order by localTime desc

and summarize the Urls including a count

requests
| where timestamp > ago(14d)
| where resultCode == \"404\"
| summarize count() by url
| order by count_ desc | project url, count_

Check Sitecore license misconfigurations

traces
| where timestamp > ago(7d)
| where message contains \"license\" and severityLevel > 1

How many unique sessions hit my environment:

pageViews
| where timestamp > ago(7d)
| summarize TotalUniqueSessions = dcount(session_Id) by bin(timestamp, 1d)
| order by timestamp asc
| render barchart kind=default

Top 10 – Most requested pageviews:

pageViews
| where timestamp > ago(7d)
| summarize count() by url
| order by count_ desc | project url, count_
| take 10

Top 10 – Most requested pageviews (combined UA and City):

pageViews
| where timestamp > ago(7d)
| summarize count() by url, client_Browser, client_City, client_CountryOrRegion
| order by count_ desc | project url, count_, client_Browser, client_City, client_CountryOrRegion
| take 10

Top 10 – 404 resultCodes:

requests
| where resultCode == 404
| where timestamp > ago(7d)
| summarize count() by url
| order by count_ desc
| take 10

Top 10 – 500 Errors in dependencies:

dependencies
| where resultCode in (\'500\', \'502\')
| where timestamp > ago(7d)
| summarize count() by name, resultCode, target, type
| project name, resultCode, target, type, count_
| order by count_ desc
| take 10

CPU Usage:

performanceCounters
| where timestamp > ago(@{variables(\'timestamp\')})
| where category == \'Processor\'
| where counter == \'% Processor Time\'
| where instance == \'_Total\'
| summarize avg(value) by bin(timestamp, 1h), cloud_RoleInstance
| order by timestamp asc
| render timechart