Github user neykov commented on a diff in the pull request:
https://github.com/apache/brooklyn-server/pull/177#discussion_r66234819
--- Diff:
core/src/main/java/org/apache/brooklyn/core/location/access/PublicNetworkFaceEnricher.java
---
@@ -0,0 +1,373 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+package org.apache.brooklyn.core.location.access;
+
+import java.net.URI;
+import java.net.URISyntaxException;
+import java.net.URL;
+import java.util.Collection;
+
+import org.apache.brooklyn.api.entity.Entity;
+import org.apache.brooklyn.api.entity.EntityLocal;
+import org.apache.brooklyn.api.location.Location;
+import org.apache.brooklyn.api.location.MachineLocation;
+import org.apache.brooklyn.api.sensor.AttributeSensor;
+import org.apache.brooklyn.api.sensor.SensorEvent;
+import org.apache.brooklyn.api.sensor.SensorEventListener;
+import org.apache.brooklyn.config.ConfigKey;
+import org.apache.brooklyn.core.config.ConfigKeys;
+import org.apache.brooklyn.core.enricher.AbstractEnricher;
+import org.apache.brooklyn.core.entity.AbstractEntity;
+import org.apache.brooklyn.core.location.Machines;
+import org.apache.brooklyn.core.sensor.Sensors;
+import org.apache.brooklyn.util.core.flags.TypeCoercions;
+import org.apache.brooklyn.util.exceptions.Exceptions;
+import org.apache.brooklyn.util.guava.Maybe;
+import org.apache.brooklyn.util.net.Networking;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import com.google.common.annotations.Beta;
+import com.google.common.base.Predicates;
+import com.google.common.collect.Lists;
+import com.google.common.net.HostAndPort;
+import com.google.common.reflect.TypeToken;
+
+/**
+ * Can be added to an entity so that it advertises its mapped ports
(according to the port-mappings
+ * recorded in the PortForwardManager). This can be used with sensors of
type URI, HostAndPort
+ * or plain integer port values. The port-mappings is retrieved by looking
up the entity's machine
+ * and the private port, in the PortForwardManager's recorded
port-mappings.
+ *
+ * For example, to configure each Tomcat node to publish its mapped uri,
and to use that sensor
+ * in Nginx for the target servers:
+ * <pre>
+ * {@code
+ * services:
+ * - type: cluster
+ * id: cluster
+ * brooklyn.config:
+ * memberSpec:
+ * $brooklyn:entitySpec:
+ * type: org.apache.brooklyn.entity.webapp.tomcat.TomcatServer
+ * brooklyn.enrichers:
+ * - type:
org.apache.brooklyn.core.location.access.PublicNetworkFaceEnricher
+ * brooklyn.config:
+ * sensor: main.uri
+ * - type: org.apache.brooklyn.entity.proxy.nginx.NginxController
+ * brooklyn.config:
+ * member.sensor.hostandport:
$brooklyn:sensor("main.uri.mapped.public")
+ * serverPool: cluster
+ * }
+ * </pre>
+ */
+@Beta
+public class PublicNetworkFaceEnricher extends AbstractEnricher {
+
+ // TODO Is this the best package for the enricher?
+ //
+ // TODO Need more logging, particularly for when the value has *not*
been transformed.
+ //
+ // TODO What if the sensor has an unrelated hostname - we will
currently still transform this!
+ // That seems acceptable: if the user configures it to look at the
sensor, then we can make
+ // assumptions that the sensor's value will need translated.
+ //
+ // TODO If there is no port-mapping, should we advertise the original
sensor value?
+ // That would allow the enricher to be used for an entity in a private
network, and for
+ // it to be a no-op in a public cloud (so the same blueprint can be
used in both).
+ // However I don't think we should publish the original value: it
could be the association
+ // just hasn't been created yet. If we publish the wrong (i.e.
untransformed) value, that
+ // will cause other entity's using attributeWhenReady to immediately
trigger.
+
+ private static final Logger LOG =
LoggerFactory.getLogger(PublicNetworkFaceEnricher.class);
+
+ @SuppressWarnings("serial")
+ public static final ConfigKey<AttributeSensor<?>> SENSOR =
ConfigKeys.newConfigKey(
+ new TypeToken<AttributeSensor<?>>() {},
+ "sensor",
+ "The sensor whose mapped value is to be re-published (with
suffix \"mapped.public\"); "
+ + "either 'sensor' or 'sensors' should be specified");
+
+ @SuppressWarnings("serial")
+ public static ConfigKey<Collection<? extends AttributeSensor<?>>>
SENSORS = ConfigKeys.newConfigKey(
+ new TypeToken<Collection<? extends AttributeSensor<?>>>() {},
+ "sensors",
+ "The multiple sensors whose mapped values are to be
re-published (with suffix \"mapped.public\"); "
+ + "either 'sensor' or 'sensors' should be specified");
+
+ public static final ConfigKey<PortForwardManager> PORT_FORWARD_MANAGER
= ConfigKeys.newConfigKey(
+ PortForwardManager.class,
+ "portForwardManager",
+ "The PortForwardManager storing the port-mappings; if null,
the global instance will be used");
--- End diff --
Looking at `BrooklynAccessUtils.getBrooklynAccessibleAddress` it's looking
at the entity for a PFM. How would things work with a non-default PFM - does it
need to be set both at the entity and the enricher?
Are we deprecating this method in light of the "Working with multiple
networks proposal?
Looking at the big picture - the port forwarding happens on the location,
using the ports of the entity. Doesn't this mean that a PFM is per location
(but for practical reasons entity)?
I'm trying to figure out how using a non-default PFM would look like -
haven't used one before.
I like the idea of always using the default one - once we have a better
understanding of using non-default PFMs and a concrete use case then we can
extend the enricher for it.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
