It was so close, question for Local Quickstart:Back-off pulling image "gcr.io/kubebuilder/kube-rbac-proxy:v0.8.0"

[OOAAHH]

Back-off pulling image "gcr.io/kubebuilder/kube-rbac-proxy:v0.8.0"

When observing the deployment of KF-TOOLS--TENSORBOARDS--TENSORBOARD-CONTROLLER, I found that one of my pods reported an error: Back-off pulling image "gcr.io/kubebuilder/kube-rbac-proxy:v0.8.0". This is very strange because when I check with docker images, I can see that the image gcr.io/kubebuilder/kube-rbac-proxy:v0.8.0 is already saved on my machine. However, as shown in the picture below, my pod still encounters issues during the image pulling process. As a measure, I directly changed the image source in its corresponding Live Manifest file, and then I saw my pod working normally. But when I ran bash ./sync_argocd_apps.sh to sync all my applications, I found that my Live Manifest reverted to its original state before my changes. Perhaps I should fork a branch of deployKF to my repository and then change the image? Requesting assistance on which file in the repository I should change to affect the behavior of bash ./sync_argocd_apps.sh?

当我在观察KF-TOOLS--TENSORBOARDS--TENSORBOARD-CONTROLLER的部署的时候发现，我有一个pods报错Back-off pulling image "gcr.io/kubebuilder/kube-rbac-proxy:v0.8.0"，这非常奇怪，我通过docker images是可以看到我的gcr.io/kubebuilder/kube-rbac-proxy:v0.8.0是已经保存在我的计算机中了，然而就像下图中所展示的那样，我的pods依然在拉取镜像的过程中遇到了问题。

于是我采取了措施，我直接在它对应的Live Manifest文件进行了更改，更换了镜像的来源，随后我看到我的pods正常工作了。但是当我进行bash ./sync_argocd_apps.sh我的所有应用以后，我发现我的Live Manifest又回到了我更改前的样子，或许我应该fork一个deployKF的分支到我的存储库，然后进行image的更改？
请求帮助，我应该到更改存储库中的哪个文件来影响bash ./sync_argocd_apps.sh的行为？