Steps to do before contacting support regarding build failures
What to do and which information about errors to gather before contacting support team to ensure faster resolution of the issue.
1st step for any build error:
-Check error logs in UI. Carefully read the error message. It will help you to have an idea of what is wrong: is there not enough memory to build a cube or something with a connection to one of the data sources? Usually, the table that caused the issue is specific.
In case you understand the error message and suggested actions (which can be like adjusting several simultaneous builds, creating a duplicated cube and trying to perform a sample build, adjusting specific timeouts, etc) - please proceed with suggested adjustments and see if it fixes the issue. Note all changes that you make, so they can be reverted if necessary.
In case, you have never encountered errors specified in an error message, and do not know how to perform suggested changes or no changes are suggested, please gather the information specified below to share it with the support team in a tick you will create.
[ALT Text: A screenshot of a user interface showing a dropdown menu with options like "Schedule Build," "Model Statistics," "Data Security," and others. Below, a build process indicator shows "0.67 Tasks Completed" along with a message detailing a "Table query schema error" and a timestamp.]
Make screenshots of all the information listed below. Screenshots should be done in a way, so the same information as in the example is visible:
- Screenshot of error from the UI. You can see it when choosing the 'View latest logs' option in the cube (check the screenshots from the section above).
- Check Grafana to see how much memory is consumed. Here is the link to an article on how to check specific PO consumption in Grafana: Check these pods:
query
build
ec-bld (affected cube and all cube
ec-qry (affected cube and all cube - Check Data Group settings. Specifically ‘Stop when idle' options and memory limits
[ALT text: A screenshot of a software interface showing the "Admin" section. The left sidebar includes options like "Home," "User Management," "App Configuration," and "Data Groups." The main area displays a table titled "Data Groups" with columns for "Group Name," "Built Role," "Query Details," "Number of Instances," and "Email Jobs." The row contains an example entry labeled "Default Checklist."]
[ALT text: A screenshot of an "Edit Data Group" interface. It displays fields for "Group Name," "Build Node," "Query Nodes," "ExistCubes," and "Instances," along with a slider labeled "Stop when idle" and a dropdown for "Idle Time (Minutes)." The values "Default," "node1," "node1," "Sample ECommerce, Sample Healthcare, Sample Leads," and "1" are filled in certain fields.]
[ALT Text: A screenshot displaying configuration settings for an ElastCube. It includes two sections: "Query Nodes" with fields for Reserved Cores, Max Cores, Reserved RAM, and Max RAM, all listed as -1 for cores and RAM. The "Build Node" section shows Reserved Cores as -1, Max Cores as -1, Reserved RAM as 512 MB, and Max RAM as 8192 MB. There is also a field labeled "User label" at the bottom.]
4. Check memory limits of build and query pods in Configuration
[ALT Text: A screenshot of the "Admin" panel from a software interface. The left sidebar shows various menu options including "Search," "User Management," "App Configuration," and "System Management." The main section displays the "Configuration" tab under "System Management," listing attributes like "Server," "Application Node," "Cluster Node," "Build Node," "Status," and "Description" with a specific server named "node1" showing a status of "Up." The interface has a clean, modern design with a white background and light accents.]
Make 5 fast clicks on the Sisense logo
[ALT Text: Screenshot of the Sisense service configuration interface, showing fields for 'MaxResultCachePrecision', 'MaxNumberOfExecutionBlocks', 'Message_init', and 'Memory Limit' within a 'Build' section. The interface has a light background and a navigation menu at the top.]
[ALT Text: A screenshot of the Sisense application displaying the "Service Configuration" menu, specifically focusing on the "Query" section. The settings shown include "LiveResultsCacheTimeout," "MaxStreamingThreads," "Memory_limit," "Memory_area," and "NumberOfReadersForQueryCube," with some values highlighted in yellow.]
- Run the following commands from the Sisense server (Only for on-premise customers:):
kubectl -n sisense get deployments.apps build -o yaml | grep -A2 limits
kubectl -n sisense get deployments.apps query -o yaml | grep -A2 limits
5. Are there any EC2EC data sources in any of the affected cubes? Specify this when creating a ticket
6. If in UI error any table name is mentioned use this table. If so, choose any table from the cube. Try to re-connect to this table using the ‘Change Connection’ or ‘Switch Connection’ option (depending on which one you have)
[ALT text: Screenshot of a data visualization software interface showing a menu on the left with CSV data sources listed, including "DimEmployees." The right side displays a diagram with yellow nodes and connections. A context menu is open, highlighting options like "Change Provider" and "Change Connection."]
[ALT text: "Screenshot of a data management software interface displaying a menu on the left for 'Sample ECommerce' with options for 'Brand,' 'Category,' 'Commerce,' and 'Directories.' A highlighted menu item shows 'Connection Settings' with options including 'Preview,' 'Duplicate,' 'Delete,' and 'Switch Connection,' among others. A yellow circle labeled 'Commerce' is shown on the right."]
7. Go through all steps of changing a connection and check if the review of the table is available. In case you receive any error during this process, repeat the process while recording the .har file (here is the link to the article on how to record the .har file). Also, take a screenshot of the error you receive. Let us know in a ticket if an error appears during the step.
8. In case any error appears in step 7, check your database and make sure the user you are using in Sisense has proper credentials and permissions, your database can accept connections.
9. Login the to Sisense server and download the log files specified below. Make sure to proceed with this step right a performing all the steps above. This way, if you record a .har file the information about it will be included in logs, as well the information about the errors will be still present (Only for on-premise customers).
- build.log
- query.log
- combined.log
- ec-bld.log (of affected cub
You may check how to download and identify necessary logs here.
10. Include information about any recent changes you have made to the cube or environment. Like adding/removing tables, more data is uploaded from the data base, new relations are created between tables, etc.
