If both the loading of resources from a gateway and the AAGM interface are very slow, the problem could be that the Gateway Manager's database buffer is too low.
We recommend a buffer of at least 1 GB. If you have more than eight gateways, 1.5 GB may also be necessary.
You can also take a step-by-step approach:
Increase the buffer
innodb_buffer_pool_size
is a read_only variable. You have to shut down the database first.SHOW VARIABLES LIKE 'innodb_buffer_pool_size';
--innodb_buffer_pool_size=1G
See all recommended settings here: https://apiida.atlassian.net/wiki/spaces/AAGM/pages/7760183321/Installation#Start-the-Containers-Using-Docker-Compose
Restart the database
Perform some operations in the Gateway Manager
Something with a lot of database access, for example: Reloading Resources or Discover APIs
Check the buffer utilisation
SELECT FLOOR((SELECT variable_value FROM INFORMATION_SCHEMA.GLOBAL_STATUS WHERE variable_name = 'Innodb_buffer_pool_bytes_data') * 100 / (SELECT variable_value FROM INFORMATION_SCHEMA.GLOBAL_VARIABLES WHERE variable_name = 'innodb_buffer_pool_size')) AS BufferPoolUsageInPercent
Start again at step 1 if the utilisation of the buffer is higher than 80 percent.