This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(70.4, 37%, 16%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EAM%3C/text%3E%3C/svg%3E)
I have 3 Azure Container Apps connected to one Log Analytics Workspace and 1 Container App Jobs. Yesterday I've stopped to see logs from 2 of 3 these container apps. I see only logs from 1 app and job.
Those 2 container apps (by 1 replica each) where redeployed via Bicep without any other infra changes. Azure Portal for these apps shows full integration with Log Analytics Workspace even with "System" logs from them. But application's "Console" logs are empty.
One of these apps had "degraded replica" for couple of hours due to long db migrations and during this time, logs from new revision are available in Log Analytics Workspace, but once old revision pod was removed the logs disappeared as well.
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(118.4, 87%, 21%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EVB%3C/text%3E%3C/svg%3E)
Hello Aleksandr Makarov,
Since you've confirmed Log Analytics integration is intact and the system logs are visible, but application (stdout/stderr) logs are missing, and this coincides with a Bicep-based redeployment, it’s likely that diagnostic settings for console logs might have been unintentionally reset or omitted during redeployment.
After redeploying via Bicep, please verify if the logAnalyticsConfiguration.destination and logAnalyticsConfiguration.customerId are still present under properties.configuration.diagnostics for the affected Container Apps using the Azure Resource Explorer or ARM:
configuration: {
diagnostics: {
logAnalyticsConfiguration: {
customerId: '<your-log-analytics-workspace-id>'
sharedKey: '<your-workspace-shared-key>'
}
}
}
Refer:
https://learn.microsoft.com/en-us/azure/container-apps/monitor?tabs=azure-cli#enable-logging
Using the powershell command make sure that the affected apps have log streaming explicitly enabled
az containerapp show --name <app-name> --resource-group <rg-name> --query "properties.configuration.diagnostics.logAnalyticsConfiguration"
If the new app revision introduced code changes, validate that logs are indeed written to stdout or stderr. In managed environments like Azure Container Apps, only console output gets picked up by Log Analytics.
Refer:
https://learn.microsoft.com/en-us/azure/container-apps/logging?tabs=azure-cli#log-types
Please let us know if it was helpful, and feel free to reach out if you have any further queries.
Vinay - thank you for suggestion! But I don't see diagnostics section in ARM/Azure Resource Explorer/JSON view both for the affected Container Apps and for those which logs are still visible in Log Analytics. All my /subscriptions/<uuid>/resourceGroups/<rg-name>providers/Microsoft.App/containerApps/<app-name> are looking similar to:
{
...
type: Microsoft.App/containerApps
properties {
provisioningState ...
...
latestRevisionName ...
configuration: {
secrets ...
activeRevisionsMode ...
targetLabel ...
registries ...
identitySettings ...
dapr ...
runtime ...
maxInactiveRevisions ...
service: null
}
template ...
Therefore az containerapp show --query returns nothing.
Are you sure that such section exists in general? Because I can't find such in https://learn.microsoft.com/en-us/azure/templates/microsoft.app/containerapps?pivots=deployment-language-bicep
While Portal shows that they are all connected to required Log Analytics Workspace - see screenshot.
(forgotten for post above screenshot)
Hello Aleksandr Makarov,
It appears the log analytics configuration is missing from your Bicep deployment. Since you confirmed that the Log Analytics Workspace is visible in the Azure Portal, try ensuring that the Bicep deployment contains the correct diagnostics configuration. The Bicep code you are using may need an explicit inclusion for the logAnalyticsConfiguration under the properties.configuration.diagnostics section, which might have been overlooked in previous deployments.
If you're not seeing the logs in the Azure Portal, manually re-enable console logging for the Container Apps:
Also try using Azure CLI to force a refresh of the diagnostic settings:
az containerapp update --name <app-name> --resource-group <rg-name> --set properties.configuration.diagnostics
Sometimes, logging issues can occur due to hitting resource quotas or log limits in Log Analytics. Review the Log Analytics workspace’s ingestion limits, especially if this issue arose suddenly.
Ensure that the log ingestion isn’t restricted or exceeded. If so, consider increasing the limits or managing log ingestion.
Please let us know if it was helpful, please upvote if it helps you anyway to reach to address the issue and feel free to reach out if you have any further queries.
It was Microsoft/Azure bug which was fixed for us without any actions from our side (no deploys or any edits in Portal/CLI). I.e. after 2+ days logs just appeared in "Logs" page of Analytics Workspace (they even haven't been lost).