openshift · adellape · Jul 11, 2017 · Jun 30, 2017 · Jul 5, 2017 · Jul 7, 2017
diff --git a/install_config/install/disconnected_install.adoc b/install_config/install/disconnected_install.adoc
@@ -2,7 +2,7 @@
 = Disconnected Installation
 {product-author}
 {product-version}
-:latest-tag: v3.4.1.24
+:latest-tag: v3.4.1.44
 :latest-int-tag: 3.4.1
 :latest-registry-console-tag: 3.4
 :data-uri:

diff --git a/install_config/upgrading/automated_upgrades.adoc b/install_config/upgrading/automated_upgrades.adoc
@@ -2,7 +2,7 @@
 = Performing Automated In-place Cluster Upgrades
 {product-author}
 {product-version}
-:latest-tag: v3.4.1.24
+:latest-tag: v3.4.1.44
 :data-uri:
 :icons:
 :experimental:

diff --git a/install_config/upgrading/manual_upgrades.adoc b/install_config/upgrading/manual_upgrades.adoc
@@ -2,7 +2,7 @@
 = Performing Manual In-place Cluster Upgrades
 {product-author}
 {product-version}
-:latest-tag: v3.4.1.24
+:latest-tag: v3.4.1.44
 :data-uri:
 :icons:
 :experimental:
@@ -1499,6 +1499,27 @@ upgrade process.
 There are no additional manual steps for the upgrade to
 xref:../../release_notes/ocp_3_4_release_notes.adoc#ocp-3-4-1-24[{product-title} 3.4.1.24] that are not already mentioned inline during the standard manual
 upgrade process.
+
+[[manual-step-ocp-3-4-1-33]]
+=== {product-title} 3.4.1.33
+
+There are no additional manual steps for the upgrade to
+xref:../../release_notes/ocp_3_4_release_notes.adoc#ocp-3-4-1-33[{product-title} 3.4.1.33] that are not already mentioned inline during the standard manual
+upgrade process.
+
+[[manual-step-ocp-3-4-1-37]]
+=== {product-title} 3.4.1.37
+
+There are no additional manual steps for the upgrade to
+xref:../../release_notes/ocp_3_4_release_notes.adoc#ocp-3-4-1-37[{product-title} 3.4.1.37] that are not already mentioned inline during the standard manual
+upgrade process.
+
+[[manual-step-ocp-3-4-1-44]]
+=== {product-title} 3.4.1.44
+
+There are no additional manual steps for the upgrade to
+xref:../../release_notes/ocp_3_4_release_notes.adoc#ocp-3-4-1-44[{product-title} 3.4.1.44] that are not already mentioned inline during the standard manual
+upgrade process.
 endif::[]
 
 ifdef::openshift-origin[]

diff --git a/release_notes/images/redcircle-1.png b/release_notes/images/redcircle-1.png
diff --git a/release_notes/images/redcircle-2.png b/release_notes/images/redcircle-2.png
diff --git a/release_notes/images/redcircle-3.png b/release_notes/images/redcircle-3.png
diff --git a/release_notes/index.adoc b/release_notes/index.adoc
@@ -25,36 +25,42 @@ endif::[]
 == Versioning Policy
 
 {product-title} provides strict backwards compatibility guarantees for all
-supported APIs (excluding alpha APIs, which may be changed without notice, and
-beta APIs, which may occasionally be changed in a non-backwards compatible
-manner). {product-title} will continue to support older clients against newer
-servers. For example, a 3.1 client will work against 3.0, 3.1, and 3.2 servers.
+supported APIs, excluding alpha APIs (which may be changed without notice) and
+beta APIs (which may occasionally be changed in a non-backwards compatible
+manner).
+
+The {product-title} version must match between master and node hosts, excluding
+temporary mismatches during cluster upgrades. For example, in a 3.4 cluster, all
+masters must be 3.4 and all nodes must be 3.4. However, {product-title} will
+continue to support older `oc` clients against newer servers. For example, a 3.3
+`oc` will work against 3.2, 3.3, and 3.4 servers.
 
 Changes of APIs for non-security related reasons will involve, at minimum, two
-minor releases (3.1 to 3.2 to 3.3, for example) to allow older clients to
-update. Using new capabilities may require newer clients. A 3.2 server may have
-additional capabilities that a 3.1 client cannot use and a 3.2 client may have
-additional capabilities that are not supported by a 3.1 server.
+minor releases (3.1 to 3.2 to 3.3, for example) to allow older `oc` to update.
+Using new capabilities may require newer `oc`. A 3.2 server may have additional
+capabilities that a 3.1 `oc` cannot use and a 3.2 `oc` may have additional
+capabilities that are not supported by a 3.1 server.
 
 .Compatibility Matrix
 
 [cols="1,1,1"]
 |===
 
 |
-|*X.Y* (Client)
-|*X.Y+N* footnoteref:[versionpolicyn,Where *N* is a number greater than 1.] (Client)
+|*X.Y* (`oc` Client)
+|*X.Y+N* footnoteref:[versionpolicyn,Where *N* is a number greater than 1.] (`oc` Client)
 
 |*X.Y* (Server)
-|1
-|3
+|image:redcircle-1.png[]
+|image:redcircle-3.png[]
 
 |*X.Y+N* footnoteref:[versionpolicyn] (Server)
-|2
-|1
+|image:redcircle-2.png[]
+|image:redcircle-1.png[]
 
 |===
+image:redcircle-1.png[] Fully compatible.
+
+image:redcircle-2.png[] `oc` client may not be able to access server features.
 
-<1> Fully compatible.
-<2> Client may not be able to access server features.
-<3> Client may provide options and features that may not be compatible with the accessed server.
+image:redcircle-3.png[] `oc` client may provide options and features that may not be compatible with the accessed server.
diff --git a/release_notes/ocp_3_4_release_notes.adoc b/release_notes/ocp_3_4_release_notes.adoc
@@ -1498,3 +1498,44 @@ initiate Ansible-based installation or upgrade procedures:
 * Previously, it was only possible to redeploy the etcd CA certificate by also redeploying the OpenShift CA certificate, which was unnecessary maintenance. With this enhancement, the etcd CA certificate may now be replaced independent of the OpenShift CA certificate using the etcd CA certificate redeployment playbook (*_playbooks/byo/openshift-cluster/redeploy-etcd-ca.yml_*). Note that the OpenShift CA redeployment playbook (*_playbooks/byo/openshift-cluster/redeploy-openshift-ca.yml_*) now only replaces the OpenShift CA certificate. Similarly, the etcd CA redeployment playbook only redeploys the etcd CA certificate. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1463773[*BZ#1463773*])
 
 * Each {product-title} version works properly with specific range of versions of packages. Thus, the package versions must be limited and the ranges enforced. This enhancement extends the installation and upgrade playbooks to install the **-excluder* packages that protects RPMs against upgrading to undesired versions. As a result, the range of versions of packages for each {product-title} version (since 3.3) is now protected. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1436343[*BZ#1436343*])
+
+[[ocp-3-4-1-44]]
+=== RHBA-2017:1640 - {product-title} 3.4.1.44 Bug Fix Update
+
+Issued: 2017-07-11
+
+{product-title} release 3.4.1.44 is now available. The packages and bug fixes
+included in the update are documented in the
+link:https://access.redhat.com/errata/RHBA-2017:1640[RHBA-2017:1640] advisory.
+The container images included in the update are documented in the
+link:https://access.redhat.com/errata/RHBA-2017:1646[RHBA-2017:1646] advisory.
+
+Space precluded documenting all of the bug fixes for this release in the
+advisory. See the following sections for notes on upgrading and details on the
+bug fixes included in this release.
+
+[[ocp-3-4-1-44-upgrading]]
+==== Upgrading
+
+To upgrade an existing {product-title} 3.3 or 3.4 cluster to this latest release, use the
+automated upgrade playbook. See
+xref:../install_config/upgrading/automated_upgrades.adoc#running-the-upgrade-playbook-directly[Performing Automated In-place Cluster Upgrades] for instructions.
+
+[[ocp-3-4-1-44-bug-fixes]]
+==== Bug Fixes
+
+* When doing an incremental build, the S2I builder pulls its builder image before calling the *_save-artifacts_* script and does not ensure that the builder image is still there when it calls *_assemble_*. This leaves a gap of time between the start of the build and the calling of the *_assemble_* script in which the image can be removed. If the image is removed, the build fails. This bug fix adds a call to ensure that the builder image exists right before calling the *_assemble_* script. As a result, the chance of the *_assemble_* script running and not finding an available builder image is greatly reduced. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1446925[*BZ#1446925*])
+
+* When Elasticsearch logging is not configured with console logging, the method that determines whether the cluster is available is not written to the logs returned by the `oc logs` command. This causes the *_runs.sh_* script to time out and exits looking for the log message. This bug fix evaluates the logging configuration to determine where to look for the `cluster.service` message. As a result, the *_run.sh_* script finds the desired message and continues to start the cluster. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1461294[*BZ#1461294*])
+
+* The Elasticsearch (ES) default value for sharing storage between ES instances was wrong. The incorrect default value allowed an ES pod starting up (when another ES pod was shutting down, e.g., during deployment configuration redeployments) to create a new location on the persistent volume (PV) for managing the storage volume. This duplicated data, and in some instances, potentially caused data loss. With this bug fix, all ES pods now run with `node.max_local_storage_nodes` set to `1`. As a result, the ES pods starting up or shutting down will no longer share the same storage, preventing data duplication and data loss. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1462277[*BZ#1462277*])
+
+* The version of Netty that is part of Cassandra 3.0.9 had a memory leak. This bug fix updates Cassandra to 3.0.13, which has a version of Netty that has a fix for the memory leak. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1457499[*BZ#1457499*])
+
+* When an IP address was re-used, it would be generated with a random MAC address that would be different from the previous one. Any node with an ARP cache that still held the old entry for the IP would not be able to communicate with the node. This bug fix generates the MAC address deterministically from the IP address. As a result, a re-used IP address will always have the same MAC address, so the ARP cache no longer gets out of sync, allowing traffic to flow. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1462952[*BZ#1462952*])
+
+* Due to a coding error, `Pop()` operations could panic and cause the router to stop. This bug fix updates this logic and as a result panics no longer occur. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1464567[*BZ#1464567*])
+
+* When master controller routines watch for persistent volume Recycle success events, they may never receive one, but still keep trying indefinitely. This caused the potential for high CPU usage by the master controller as it leaks routines. This bug fix updates the routines to stop watching for these Recycle success events when they will never be received. As a result, the chance of high CPU usage by the master controller is reduced. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1438741[*BZ#1438741*])
+
+* Due to older Kubernetes libraries requiring directory renames instead of deletion, certain `emptyDir` situations could cause high CPU usage. This bug updates these processes to delete `emptyDir` instead of rename, and as a result the high CPU is avoided. (link:https://bugzilla.redhat.com/show_bug.cgi?id=1460260[*BZ#1460260*])
diff --git a/release_notes/revhistory_release_notes.adoc b/release_notes/revhistory_release_notes.adoc
@@ -8,6 +8,26 @@
 
 // do-release: revhist-tables
 
+== Tue Jul 11 2017
+
+// tag::release_notes_tue_jul_11_2017[]
+[cols="1,3",options="header"]
+|===
+
+|Affected Topic |Description of Change
+//Tue Jul 11 2017
+
+|xref:../release_notes/index.adoc#release-notes-index[Overview]
+|Clarified that "client" referred to the `oc` client.
+
+|xref:../release_notes/ocp_3_4_release_notes.adoc#release-notes-ocp-3-4-release-notes[{product-title} 3.4 Release Notes]
+|Added release notes for
+xref:../release_notes/ocp_3_4_release_notes.adoc#ocp-3-4-1-44[RHBA-2017:1640 - OpenShift Container Platform 3.4.1.44 Bug Fix Update].
+
+|===
+
+// end::release_notes_tue_jul_11_2017[]
+
 == Thu Jun 29 2017
 
 // tag::release_notes_thu_jun_29_2017[]

diff --git a/welcome/revhistory_full.adoc b/welcome/revhistory_full.adoc
@@ -11,6 +11,11 @@ date.
 
 // do-release: revhist-tables
 
+== Tue Jul 11 2017
+
+.Release Notes
+include::release_notes/revhistory_release_notes.adoc[tag=release_notes_tue_jul_11_2017]
+
 == Thu Jul 06 2017
 .Installation and Configuration
 include::install_config/revhistory_install_config.adoc[tag=install_config_thu_jul_06_2017]