You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, crawler pods have a fixed resource usage. With our custom operator and metrics api, we can dynamically give crawler pods more memory, and/or restart them before they are killed by the system.
Context
Crawling some sites will inevitable require more memory than others. With ability to resize the pods (with a restart for now, eventually without), crawling can ideally be more smooth and avoid data corruption due to restarts.
The idea is to gracefully stop the crawler pods and increase memory if >90% of memory is used.
We can also set the memory limit > memory request, and less gracefully restart the pod if >100% requested memory is used (soft OOM), but before the pod is killed by OOM Killer
The text was updated successfully, but these errors were encountered:
- set memory limit to 1.2x memory request to provide extra padding and
avoid OOM
- attempt to resize crawler pods by 1.2x when exceeding 90% of available
memory
- do a 'soft OOM' (send extra SIGTERM) to pod when reaching 100% of
requested memory, resulting in faster graceful restart, but avoiding a
system-instant OOM Kill
- Fixes#1632
---------
Co-authored-by: Tessa Walsh <tessa@bitarchivist.net>
What change would you like to see?
Currently, crawler pods have a fixed resource usage. With our custom operator and metrics api, we can dynamically give crawler pods more memory, and/or restart them before they are killed by the system.
Context
Crawling some sites will inevitable require more memory than others. With ability to resize the pods (with a restart for now, eventually without), crawling can ideally be more smooth and avoid data corruption due to restarts.
The idea is to gracefully stop the crawler pods and increase memory if >90% of memory is used.
We can also set the memory limit > memory request, and less gracefully restart the pod if >100% requested memory is used (soft OOM), but before the pod is killed by OOM Killer
The text was updated successfully, but these errors were encountered: