# AWX EE poor performance **URL:** \ **Category:** AWX Project\ **Tags:** awx, ee, kubernetes\ **Created:** [April 18, 2022, 8:24pm UTC](https://forum.ansible.com/t/awx-ee-poor-performance/36537 "2022-04-18T20:24:48Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![adpavlov](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/adpavlov/32/10205_2.png) [@adpavlov](https://forum.ansible.com/u/adpavlov)\ **Post date:** [April 18, 2022, 8:24pm UTC](https://forum.ansible.com/t/awx-ee-poor-performance/36537/1 "2022-04-18T20:24:48Z")
Hi everyone, Finally trying to migrate my docker based AWX 15 to kubernetes based AWX 20. I have to use custom EE to have ansible 2.9. In EE ideology each playbook execution means separate pod/container creation, which actually takes a lot of time to bring infrastructure up before actual execution starts. So I have few questions that I could not google an answer: 1. Can I make custom EE pod always ready and running like it’s done for control plane EE? 2. Can I make e.g. 4 containers always up and ready to start execution immediately? Thanks in advance! ---
**Author:** ![Michael\_Nhan](https://avatars.discourse-cdn.com/v4/letter/m/58f4c7/32.png) [@Michael\_Nhan](https://forum.ansible.com/u/Michael_Nhan)\ **Post date:** [April 21, 2022, 4:51pm UTC](https://forum.ansible.com/t/awx-ee-poor-performance/36537/2 "2022-04-21T16:51:22Z")
I don’t have any answer but just adding myself to the convo cause I’m interested in such things as well. ---
**Author:** ![Rebeccah\_Hunter](https://avatars.discourse-cdn.com/v4/letter/r/4da419/32.png) [@Rebeccah\_Hunter](https://forum.ansible.com/u/Rebeccah_Hunter)\ **Post date:** [May 4, 2022, 7:42pm UTC](https://forum.ansible.com/t/awx-ee-poor-performance/36537/3 "2022-05-04T19:42:15Z")
Hey folks, Unfortunately AWX doesn’t support having standing EE pods that can be set up and be always up and ready to start. There are no plans at this time to do this but you are more than welcome to open an enhancement request in the AWX repository. [https://github.com/ansible/awx/issues/new/choose](https://github.com/ansible/awx/issues/new/choose) will take you to our Github where you can choose what kind of issue you want to open. Cheers, rebeccahhh + the AWX team ---
**Author:** ![Christian\_Adams](https://avatars.discourse-cdn.com/v4/letter/c/ba8739/32.png) [@Christian\_Adams](https://forum.ansible.com/u/Christian_Adams)\ **Post date:** [May 4, 2022, 7:45pm UTC](https://forum.ansible.com/t/awx-ee-poor-performance/36537/4 "2022-05-04T19:45:51Z")
What takes the most time is pulling the EE image for the first time. so after the initial run, the image is cached on the awx node and subsequent job runs are actually quite speedy. Thanks, Christian ---
**Author:** ![adpavlov](https://sea1.discourse-cdn.com/ansible/user_avatar/forum.ansible.com/adpavlov/32/10205_2.png) [@adpavlov](https://forum.ansible.com/u/adpavlov)\ **Post date:** [May 25, 2022, 2:58pm UTC](https://forum.ansible.com/t/awx-ee-poor-performance/36537/5 "2022-05-25T14:58:26Z")
It’s real degradation in performance comparing to AWX 15 and docker architecture. Experiencing very slow behavior. In my case few parallel jobs are running quite fast and fine. However, if I need to launch more than 10 parallel jobs it will lead to 10 Pods/containers creation, which takes more than 2 minutes to create Pods and containers. At the same time awx\_task container memory usage goes to the sky up to 8Gb, server is lagging. Sometimes jobs are going in Error state with different reasons like: - Error creating pod: rpc error: code = Unknown desc = database is locked - No output found for this job. - Error opening pod stream: Get “[https://ansible:10250/containerLogs/awx/automation-job-503-9j922/worker?follow=true](https://ansible:10250/containerLogs/awx/automation-job-503-9j922/worker?follow=true)”: write tcp 10.50.78.37:6443-\>10.50.78.37:41436: i/o timeout среда, 4 мая 2022 г. в 22:45:52 UTC+3, [cha...@redhat.com](mailto:cha...@redhat.com):