Troubleshoot

Logging

MailScheduler keeps track of errors in log files. These log files can be viewed in the application and are also visible in the file storage on the machine.

• In the frontend, Click on About in the menu and click the pink System logs button. At the left there is a list of log files. The left part of the page shows the lines in the log files. Clicking on a line will open the details of the log item. There is also a download logs button that will compress all log files into a zip file for easy sending.

• On the filesystem in the directory where MailScheduler.exe is located, there is a directory .app-runtime/storage/logs which contains all log files.

When you encounter an error in the application, it is always helpful to send the logs with the support ticket so we can directly start investigating.

Application startup

When the application does not start, the cause can probably be found in the log files. they are located in <installation_path>/.app-runtime/storage/logs/*.log. When opening a support ticket, please provide these files.

Could not startup server, Contact Apps For Tableau support when the problem persists

The app server is unable to start. This can have multiple causes like:

• Application is configured as administrator but is currently not running with administrator privileges.

• Check if there is no other MailScheduler process running in the background. Open Task manager and select the Details tab. You should have only one process called MailScheduler.

• Check if there is no other process running on the port that is configured in MailScheduler. You can do that with this command in cmd:

netstat -aon

• Open power shell and navigate to the directory where MailScheduler.exe is located. run this command to manually trigger the application server:

./MailScheduler.exe serve

This command might give more details about what went wrong.

listen tcp :443: bind: Only one usage of each socket address (protocol/network address/port) is hormally permitted

When this error appears in your go-application.log it is not able to spin the server because the port you have configured is already in use by another program.

MailScheduler is slow or unresponsive

A typical reason of a slow experience is the connection to tableau. MailScheduler uses the REST API to fetch the visuals from Tableau. Each unique combination of filters will cause an extra request to tableau for fetching this information. On large mailing lists this can cause delay.

When the frontend for example is slow you can take a look at the application performance. Open Powershell and navigate to the directory where MailScheduler.exe is located. Run this command to see the statistics of the internal workers:

./MailScheduler.exe workers --interactive

This will generate an interactive statistics table like so:

Allowed memory size of xxx bytes exhausted

Big tasks can involve large amounts of memory. This error indicates that MailScheduler is trying to use more memory than the memory limit is allowing. To resolve this issue, increase the memory limit by altering the app.config.yaml file by adding the memory limit, and setting it to a higher value, like so:

A generally 'safe' memory limit is 1G, or 1 Gigabyte. Note that the indentation in app.config.yaml matters! Always use two spaces.

After setting the new limit, make sure you restart MailScheduler to apply it.

Single Sign-On

Common error messages:

A valid SubjectConfirmation was not found on this Response

This could indicate there is a mismatch between the Recipient and Destination URLs. Check in the SAML response XML if Recipient and Destination URL's match. When it does not match it should be configured in your SAML provider.

Unknown AssertionConsumerServiceURL

The URL in the configuration file is probaby incorrect. Check your app.config.yaml file and look at the app.url variable. That should match the domain that is used to access MailScheduler

Authentication method by which the user authenticated with the service doesn't match requested authentication method

This problem occures/happends, because of the way how the session authentication method (SAML AuthnRequest) is configured on the other SSO app. MailScheduler by defaults use “Password, ProtectedTransport” as request authentication method.

Solution

We have added a configuration variable to allow all options, but allow any cross request authentication method. Within the config.yaml file set the following variable to not strictly check on the cross request authentication method:

  env:
    portal:
      saml:
        requested_auth: false