ArgoCD

Slow Sync Operations from Kubectl Exec Bottleneck

warning
latencyUpdated Feb 2, 2026

ArgoCD's default kubectl parallelism limit throttles sync operations when applying multiple resources, causing syncs to take 10+ minutes for applications with many Kubernetes objects.

How to detect:

Monitor argocd_app_controller_kubectl_exec_pending showing consistent backlog and argocd_app_controller_app_sync_time_seconds exceeding 5 minutes for apps with 50+ resources. High sync duration correlated with many resources indicates kubectl throttling. Check argocd-application-controller logs for 'waiting for kubectl' messages.

Recommended action:

Increase --kubectl-parallelism-limit from default 20 to 30-50 via controller.extraArgs in Helm values. Balance with available memory (each kubectl process consumes ~50MB). For very large applications, split into multiple smaller Applications using sync waves or dependencies. Monitor for OOMKills after increasing parallelism and adjust memory limits accordingly.