This increase is much slower than the above issue, happening slowly over days and weeks:
The drop in memory is when the ECS was restarted due to us pushing a new image. We have not let it run for long enough without any restarts to see it crash at 100%, this is a preemptive report to the team before it happens.
Mage version
v0.9.68
Describe the bug
We have deployed the Mage image into ECS using terraform as described in the docs: https://github.com/mage-ai/mage-ai-terraform-templates
There was an earlier issue with memory rising quickly that was reported by us: https://github.com/mage-ai/mage-ai/issues/3244 which was fixed by an update.
This increase is much slower than the above issue, happening slowly over days and weeks:
The drop in memory is when the ECS was restarted due to us pushing a new image. We have not let it run for long enough without any restarts to see it crash at 100%, this is a preemptive report to the team before it happens.
To reproduce
No response
Expected behavior
No response
Screenshots
No response
Operating system
No response
Additional context
No response