Update fluentd-gcp configuration

cluster/addons/fluentd-gcp/fluentd-gcp-image/Makefile

@@ -26,25 +26,14 @@
 # 6. When PR is approved make the gcr.io version of the image: make build push
 # 7. Revert the referendes to kubernetes/fluentd-gcp:$(TAG) to gcr.io/google_containers/fluentd-gcp:$(TAG)
 
-.PHONY:	kbuild kpush
+.PHONY:	build push
 
-TAG = 1.25
-
-# Rules for building the test image for deployment to Dockerhub with user kubernetes.
-
-kbuild:
-	docker build -t kubernetes/fluentd-gcp:$(TAG) .
-
-
-kpush:
-	docker push kubernetes/fluentd-gcp:$(TAG)
-
-
-# Rules for building the real image for deployment to gcr.io
+PREFIX=gcr.io/google_containers
+TAG = 1.26
 
 build:
-	docker build -t gcr.io/google_containers/fluentd-gcp:$(TAG) .
+	docker build -t $(PREFIX)/fluentd-gcp:$(TAG) .
 
 
 push:
-	gcloud docker -- push gcr.io/google_containers/fluentd-gcp:$(TAG)
+	gcloud docker -- push $(PREFIX)/fluentd-gcp:$(TAG)

cluster/addons/fluentd-gcp/fluentd-gcp-image/google-fluentd-journal.conf

@@ -222,12 +222,18 @@
 # them separate since most users don't care about the node logs.
 <match kubernetes.**>
   type google_cloud
+  # Set the buffer type to file to improve the reliability and reduce the memory consumption
+  buffer_type file
+  buffer_path /var/log/fluentd-buffers/kubernetes.containers.buffer
+  # Set queue_full action to block because we want to pause gracefully
+  # in case of the off-the-limits load instead of throwing an exception
+  buffer_queue_full_action block
   # Set the chunk limit conservatively to avoid exceeding the GCL limit
   # of 10MiB per write request.
   buffer_chunk_limit 2M
   # Cap the combined memory usage of this buffer and the one below to
-  # 2MiB/chunk * (24 + 8) chunks = 64 MiB
-  buffer_queue_limit 24
+  # 2MiB/chunk * (6 + 2) chunks = 16 MiB
+  buffer_queue_limit 6
   # Never wait more than 5 seconds before flushing logs in the non-error case.
   flush_interval 5s
   # Never wait longer than 30 seconds between retries.
@@ -235,7 +241,7 @@
   # Disable the limit on the number of retries (retry forever).
   disable_retry_limit
   # Use multiple threads for processing.
-  num_threads 8
+  num_threads 2
 </match>
 
 # Keep a smaller buffer here since these logs are less important than the user's
@@ -243,10 +249,13 @@
 <match **>
   type google_cloud
   detect_subservice false
+  buffer_type file
+  buffer_path /var/log/fluentd-buffers/kubernetes.system.buffer
+  buffer_queue_full_action block
   buffer_chunk_limit 2M
-  buffer_queue_limit 8
+  buffer_queue_limit 2
   flush_interval 5s
   max_retry_wait 30
   disable_retry_limit
-  num_threads 8
+  num_threads 2
 </match>

cluster/addons/fluentd-gcp/fluentd-gcp-image/google-fluentd.conf

@@ -206,12 +206,18 @@
 # them separate since most users don't care about the node logs.
 <match kubernetes.**>
   type google_cloud
+  # Set the buffer type to file to improve the reliability and reduce the memory consumption
+  buffer_type file
+  buffer_path /var/log/fluentd-buffers/kubernetes.containers.buffer
+  # Set queue_full action to block because we want to pause gracefully
+  # in case of the off-the-limits load instead of throwing an exception
+  buffer_queue_full_action block
   # Set the chunk limit conservatively to avoid exceeding the GCL limit
   # of 10MiB per write request.
   buffer_chunk_limit 2M
   # Cap the combined memory usage of this buffer and the one below to
-  # 2MiB/chunk * (24 + 8) chunks = 64 MiB
-  buffer_queue_limit 24
+  # 2MiB/chunk * (6 + 2) chunks = 16 MiB
+  buffer_queue_limit 6
   # Never wait more than 5 seconds before flushing logs in the non-error case.
   flush_interval 5s
   # Never wait longer than 30 seconds between retries.
@@ -219,7 +225,7 @@
   # Disable the limit on the number of retries (retry forever).
   disable_retry_limit
   # Use multiple threads for processing.
-  num_threads 8
+  num_threads 2
 </match>
 
 # Keep a smaller buffer here since these logs are less important than the user's
@@ -227,10 +233,13 @@
 <match **>
   type google_cloud
   detect_subservice false
+  buffer_type file
+  buffer_path /var/log/fluentd-buffers/kubernetes.system.buffer
+  buffer_queue_full_action block
   buffer_chunk_limit 2M
-  buffer_queue_limit 8
+  buffer_queue_limit 2
   flush_interval 5s
   max_retry_wait 30
   disable_retry_limit
-  num_threads 8
+  num_threads 2
 </match>