We are receiving below error. Why it is coming.
Error occurred during SQL script execution
SQL Error  [HY000]: Leaf Error (10.100.0.18:3306): The operating system failed to allocate memory (MemSQL memory use 25576.12 Mb). The request was not processed. See Memory Errors · SingleStore Documentation for additional information
That error is usually due to a linux miss configuration. It means MemSQL sees lots of memory available on the machine but linux is refusing memory allocation requests.
you may want to try running this to see if it can find the miss configuration:
otherwise, send a cluster report to firstname.lastname@example.org and we can take a look at what the issue may be.
what about k8s instalments? does memsql know how to ignore the underlying node’s ram?
currently the studio node reports are all off.
MemSQL is looking at the maximum_memory system variable to know if there is memory available for it to use (We have safe guards to prevents maximum_memory higher then the physical memory on the machine - but if your running multiple memsql on the same host those safe guards don’t work). So its up to the code setting up memsql on k8s to set maximum_memory appropriately to split up your underlying host (which our k8s operator will do for you).
I face a similar error on my docker SingleStore setup on an EC2 instance (t2.xlarge)
Anyway to fix this or avoid this?
Ran into this issue with the docker install of Singlestore but could not find an answer. I was able to solve this by setting my docker host’s vm settings to the Singlestore requirements:
sudo sysctl -w vm.max_map_count=1000000000
sudo sysctl -w vm.min_free_kbytes=4194304
and then restrarting the docker container. Docker shares these values from the host so they cannot be set from within the running container. Hope this helps others with the issue.
Note: forum won’t let me include links in my post, so search for this in the Singlestore documentation: " Configure Linux vm Settings"