I've never seen data server be down like that. What release of Tableau are you using? Do you have enough network bandwidth, cpu, memory, disk? Are you running VM or physical? Have you looked in the server tabadmin log for more details?
Its 9.3 and had enough resources already in place and checked them.
3 of 3 people found this helpful
I believe you are aware about the all processes of Tableau Server components. As Backgrounder take maximum of processor and Tableau Data Server holds all connection information related to Extract , Workbook etc... Since your both backgrounder processors are busy and Data Server is not getting enough memory to work upon the request received that's why it is getting down.
If I am not wrong you have to change the setting of all the nodes (as both nodes are busy in serving backgrounder processor need) to maintain the performance of the server could you please help me with below details so that I can help you with best configuration of all the nodes along with Load balancer in between the node to distribute the request to less load server.
# of Cores Node1 & Node2
RAM on Node1 & Node2
What is the hardware capacity of the Primary Node?
it's a good point about the backgrounders being active and consuming all the CPU, I didn't initially see this upon my first review...
If this is for a Production system, not a bad idea to open a support case if you haven't already: Support Case | Tableau Software
Though as some folks have pointed out on this thread it could be resources are becoming constrained. Tableau has a great tool, LogShark which you can run your Tableau Server log files through and it will create a workbook that gives some great insight into your system. Tableau Server includes a service resource monitor that periodically checks both memory/CPU usage. Part of the workbook includes this and its easy to setup. The only caveat is since LogShark is a Tableau tool, it's not officially supported. That is, if you have issues or need additional insight you'll want to use the forums:
I agree with Mark....the answer is in your server logs so it would be worth sending them to Tableau Support.
Seems unlikely to me that all 4 of the Data Server processes across multiple nodes went down due to a lack of resources, while all the other app tier components (most of which are more resource intensive than a Data Server process) are unaffected.