Changelog

CHANGELOG.md

@@ -1,3 +1,19 @@
+## 2.6.1
+
+- Update CONTRIBUTORS
+- Prevent gathering facts for the same servers in loop. (thanks Pavel Zinchuk)
+- Update consul\_systemd.service.j2 (thanks Han Sooloo)
+- Allow "Connect" for any and all nodes (#401) (thanks adawalli)
+- Increment to 1.7.3 (#395) (thanks Stuart Low)
+- fix: Stop sending logs to syslog through systemd (#393) (thanks Samuel Mutel)
+- Don't create TLS folder if consul\_tls\_copy\_keys is false (#369) (thanks Samuel Mutel)
+- Setup the role to be used in check mode (thanks Louis Paret)
+- Use consul\_node\_name as empty by default to use hostname (#373) (thanks Louis Paret)
+- Corrected version Fedora dropped libselinux-python. (#385) (thanks jebas)
+- add leave\_on\_terminate (thanks Le Minh Duc)
+- Allow consul connect on bootstrap nodes (thanks Robert Edström)
+- Fix unarchive consul package to run_once (thanks schaltiemi)
+
 ## 2.6.0
 
 - Consul v1.7.0
@@ -120,14 +136,14 @@
 - fixed multiarch deployment race condition (thanks @lanefu)
 - Switched from systemctl command to systemd module [lint]
 - Update for E504 use 'delegate_to: localhost' [lint]
-  - asserts
-  - install
-  - encrypt_gossip
+ - asserts
+ - install
+ - encrypt_gossip
 - Update for E104 in with_fileglob for install_remote [lint]
 - Update for E601 in syslog [lint]
 - Update for E602 in tasks [lint]
-  - acl
-  - main
+ - acl
+ - main
 - Update example site playbook roles format
 - Support install on Debian Testing (thanks @gfeun)
 - Fix consul_bind_address (thanks @danielkucera)
@@ -163,8 +179,8 @@
 ## v2.3.0
 
 - The role no longer attempts to install the unzip binary locally onto
-  the Ansible control host; it is now a hard dependency and role execution
-  will fail if unzip is not in the PATH on the control host.
+ the Ansible control host; it is now a hard dependency and role execution
+ will fail if unzip is not in the PATH on the control host.
 - Snapshot agent installation and configuration (thanks @drewmullen)
 - Delegate Consul datacenter DNS domain to Consul (thanks @teralype)
 - Allow DNSmasq binding to particular interfaces (thanks @teralype)
@@ -294,13 +310,13 @@
 ## v1.30.1
 
 - Add performance tuning configuration (thanks @t0k4rt)
-  - Set raft multiplier to 1
+ - Set raft multiplier to 1
 - Conditionally install Python dependency baed on virtualenv or --user
-  Addresses https://github.com/brianshumate/ansible-consul/issues/129#issuecomment-356095611
+ Addresses https://github.com/brianshumate/ansible-consul/issues/129#issuecomment-356095611
 - Update includes to import_tasks and include_tasks
 - Remove invalid consul_version key from configuration
 - Update Vagrantfile
-  - Set client address to 0.0.0.0 so Vagrant based deploy checks now pass
+ - Set client address to 0.0.0.0 so Vagrant based deploy checks now pass
 - Update documentation
 
 ## v1.30.0
@@ -494,11 +510,11 @@ Do not enable a default set of ACL policies
 ## v1.18.0
 
 - Add new vars
-  - `consul_run_path` for the PID file
+ - `consul_run_path` for the PID file
 - Add bootstrap-expect toggle option (thanks @groggemans)
 - Use directory variables in dirs tasks
 - Do not attempt to install Consul binary if already found on consul_bin_path
-  - Fixes #60
+ - Fixes #60
 - Rename intermediate `boostrap_marker` var
 - Formatting on CONTRIBUTING
 - Update CONTRIBUTORS
@@ -514,8 +530,8 @@ Do not enable a default set of ACL policies
 ## v1.17.3
 
 - Revert local_action tasks
-  - Ansible generally spazzes out with  "no action detected in task"
-    for any variation of local_task I tried
+ - Ansible generally spazzes out with "no action detected in task"
+  for any variation of local_task I tried
 
 ## v1.17.2
 
@@ -529,8 +545,8 @@ Do not enable a default set of ACL policies
 ## v1.17.0
 
 - Updated configuration directory structure (thanks @groggemans)
-  - Updated `consul_config_path` to point to `/etc/consul`
-  - Added `consul_configd_path` defaulting to `/etc/consul.d`
+ - Updated `consul_config_path` to point to `/etc/consul`
+ - Added `consul_configd_path` defaulting to `/etc/consul.d`
 - Added `consul_debug` variable - defaults to *no* (thanks @groggemans)
 - Moved all config related tasks to `tasks/config.yml` (thanks @groggemans)
 - Added ACL and TLS parameters to the main `config.json` (thanks @groggemans)
@@ -547,13 +563,13 @@ Do not enable a default set of ACL policies
 ## v1.16.2
 
 - Standing corrected - put node_role back into defaults as it will still be
-  overridden by host vars (sorry @groggemans)
+ overridden by host vars (sorry @groggemans)
 - Update documentation
 
 ## v1.16.1
 
 - Revert node_role addition to default vars so clusters will still properly
-  come up since we basically lost access the bootstrap role
+ come up since we basically lost access the bootstrap role
 
 ## v1.16.0
 
@@ -575,24 +591,24 @@ Do not enable a default set of ACL policies
 
 - Add option to download binaries directly to remotes (thanks @jonhatalla)
 - Add environment variable overrides for the following default variables:
-  - `consul_bind_address`
-  - `consul_datacenter`
-  - `consul_domain`
-  - `consul_group_name`
-  - `consul_log_level`
-  - `consul_syslog_enable`
-  - `consul_acl_default_policy`
-  - `consul_acl_down_policy`
-  - Rename `consul_src_files` variable
-  - Rename `consul_copy_keys` variable
-  - Rename `consul_ca_crt` variable
-  - Rename `consul_server_crt` variable
-  - Rename `consul_tls_server_key` variable
-  - Rename `consul_verify_outgoing` variable
-  - Rename `consul_verify_server_hostname` variable
-  - Move `consul_iface` default to value of `hostvars.consul_iface`
-    - Override with elsewhere or with `CONSUL_IFACE` environment variable
-    - Closes #40
+ - `consul_bind_address`
+ - `consul_datacenter`
+ - `consul_domain`
+ - `consul_group_name`
+ - `consul_log_level`
+ - `consul_syslog_enable`
+ - `consul_acl_default_policy`
+ - `consul_acl_down_policy`
+ - Rename `consul_src_files` variable
+ - Rename `consul_copy_keys` variable
+ - Rename `consul_ca_crt` variable
+ - Rename `consul_server_crt` variable
+ - Rename `consul_tls_server_key` variable
+ - Rename `consul_verify_outgoing` variable
+ - Rename `consul_verify_server_hostname` variable
+ - Move `consul_iface` default to value of `hostvars.consul_iface`
+  - Override with elsewhere or with `CONSUL_IFACE` environment variable
+  - Closes #40
 - Update documentation
 
 ## v1.14.0
@@ -641,22 +657,22 @@ Do not enable a default set of ACL policies
 
 - File permission updates (thanks @arledesma)
 - Explicit consul_user/consul_group ownership of configurations
-  (thanks @arledesma)
+ (thanks @arledesma)
 - Use consul_bin_path throughout (thanks @arledesma)
 
 
 ## v1.10.5
 
 - Additional fixes to debian init
 - Add consul_config_custom for role users to specify new or overwrite
-  existing configuration (thanks @arledesma)
+ existing configuration (thanks @arledesma)
 
 ## v1.10.4
 
 - Corrections to config_debianint.j2 for #34
 - Update main task to prefer open Consul HTTP API port over PID file
 - Update package cache before installing OS packages
-  (watch for and refuse reversion of this as it's occurred once now)
+ (watch for and refuse reversion of this as it's occurred once now)
 
 ## v1.10.3
 
@@ -680,7 +696,7 @@ Do not enable a default set of ACL policies
 - Initial FreeBSD support
 - Vagrantfile updated for FreeBSD
 - Added checks for interface addresses for differences (obj vs. literal list)
-  in ipv4 addresses as returned by Linux vs. BSD/SmartOS
+ in ipv4 addresses as returned by Linux vs. BSD/SmartOS
 - New `consul_os` var gets operating system name as lowercase string
 - Add AMD64 pass-through/kludge to consul_architecture_map configuration
 - Update Vagrantfile
@@ -745,9 +761,9 @@ Do not enable a default set of ACL policies
 
 - Consul 0.7.5
 - BREAKING CHANGE: Deprecate read/write of ACL tokens from file system
-  functionality and prefer setting tokens from existing cluster nodes with
-  `CONSUL_ACL_MASTER_TOKEN` and `CONSUL_ACL_REPLICATION_TOKEN` environment
-  variables instead
+ functionality and prefer setting tokens from existing cluster nodes with
+ `CONSUL_ACL_MASTER_TOKEN` and `CONSUL_ACL_REPLICATION_TOKEN` environment
+ variables instead
 - Update documentation
 
 ## v1.7.4
@@ -777,7 +793,7 @@ Do not enable a default set of ACL policies
 ## v1.6.2
 
 - Stop reconfiguring bootstrap node as it's not really necessary and
-  spurious races cause failure to re-establish cluster quorum when doing so
+ spurious races cause failure to re-establish cluster quorum when doing so
 - CONSUL_VERSION environment variable
 - Deprecated default variables cleanup
 
@@ -793,7 +809,7 @@ Do not enable a default set of ACL policies
 - Update documentation
 - Fixing bug with deleting file. Better regex. Formatting. (Thanks @violuke)
 - Remember ACL master/replication tokens between runs.
-  Actually set replication token. (Thanks @violuke)
+ Actually set replication token. (Thanks @violuke)
 - Typo fix (Thanks @violuke)
 - Allowing re-running to add new nodes. More HA too. (Thanks @violuke)
 
@@ -828,7 +844,7 @@ Do not enable a default set of ACL policies
 ## v1.5.1
 
 - Fail when ethernet interface specified by consul_iface not found on
-  the system (addresses #13)
+ the system (addresses #13)
 
 ## v1.5.0
 
@@ -949,7 +965,7 @@ Do not enable a default set of ACL policies
 
 - Updated README
 - Undo 125bd4bb369bb85f58a09b5dc81839e2779bd29f as dots in node_name breaks
-  DNS API (without recursor option) and also breaks dnsmasq option
+ DNS API (without recursor option) and also breaks dnsmasq option
 
 ## v1.2.3
 
@@ -972,11 +988,11 @@ Do not enable a default set of ACL policies
 ## v1.1.0
 
 - Bare role now installs and bootstraps cluster; included site.yml will also
-  reconfigure bootstrap node as server and optionally enable dnsmasq
-  forwarding for all cluster agents
+ reconfigure bootstrap node as server and optionally enable dnsmasq
+ forwarding for all cluster agents
 - Remove bad client_addr bind in favor of default (localhost)
-  Some weirdness was occurring whereby the client APIs were listening on
-  TCP6/UDP6 sockets but not TCP4/UDP4 when client_addr set to 0.0.0.0
+ Some weirdness was occurring whereby the client APIs were listening on
+ TCP6/UDP6 sockets but not TCP4/UDP4 when client_addr set to 0.0.0.0
 - Adjust timeouts for cluster UI check
 - Default configurable domain to "consul" so that examples from docs work, etc.
 - Combine all OS vars into main (addresses undefined var warnings)
@@ -1000,7 +1016,7 @@ Do not enable a default set of ACL policies
 
 - Add initial dnsmasq front end bits
 - Reconfigure bootstrap node for normal operation (remove bootstrap-expect)
-  after initial cluster formation and restart bootstrap node
+ after initial cluster formation and restart bootstrap node
 
 ## v1.0.12
 
@@ -1042,7 +1058,7 @@ Do not enable a default set of ACL policies
 
 - Updated defaults and Consul version (thanks @bscott)
 - Made cluster bootable and switch to become_user + other Ansibel best
-  practices (thanks @Rodjers)
+ practices (thanks @Rodjers)
 - Updated minimum Ansible version required in meta
 
 ## v1.0.4