CHANGELOG

TO BE RELEASED

v5.1.2 - 6/24/2024

pinned public_suffix gem to a version that still supports ruby 2.x
fix queued jobs metric generator script to sent operation types to new endpoint

v5.1.1 - 6/14/2024

queued job count metric generator updated to use new OC endpoint
some other unneeded OC metric generators removed as their endpoints have gone away and were not recreated

v5.1.0 - 6/12/2024

Opencast 15 configuration changes:

Remove activemq
Remove unused configuration (ldap, ca sync service)
Update templates to latest
Change ES index name (just so that we can rollback if needed)
Change default captions flavor

v5.0.7 - 3/7/2024

Adding tags to the crowdstrike config
Bump activemq version to 5.15.16

v5.0.6 - 2/2/2024

Add cold archive configuration

v5.0.5 - 12/11/2023

/search/lucene* endpoint to be accessed only by admins

v5.0.4 - 10/26/2023

opencast.service file should use path to the current deployment (vs hardcoded release path)

v5.0.3 - 10/04/2023

Clean pip cache just like we do for yum
Removed the bits of the ua-harvester that were pulling zoom analytics

v5.0.2 - 09/12/2023

Upping the engage nginx large_client_header_buffers setting (from 4 * 8k to 4 * 16k) to allow for big cookies and avoid HTTP 431/400 errors

v5.0.1 - 09/08/2023

Fixing again the MemoryUsed metric; first try didn't end up with the correct modification
mysql backup cronjob wasn't sourcing anything that enabled the msyql software collection
add missing SASL packages needed by postfix

v5.0.0 - 08/21/2023

DEPLOY-86: update recipes to be compatible with RHEL7

Siginificant changes include:

java 11 openjdk installed (instead of amazon corretto)
installation and enabling of rhel7 "software collections" for python 3.8 and mysql 5.7
switch to using systemd service units for activemq and opencast
update chef service blocks to explicitly use systemd
setting selinux to permissive, meaning it will log warnings but not block anything
creating a common function for pip installing python packages
fixed issue where karaf pid/lock directories weren't getting created prior to OC start
crowdstrike recipe needed to use yum localinstall to prevent errors when package was already installed (eg as part of the AMI)
provide an updated java.security file to allow TLSv1.1 which OC needs for the jdbc connection handshake
disable ssl on the jdbc connection from OC to MySQL: The RDS instance is on a private subnet and all communication between OC and the RDS resources takes place within the VPC, so disabling SSL is OK
fixing a bug in the ganglia monitoring service definition file causing hung opsworks configure lifecycle events

v4.4.6

fix for missing engage id not being passed to install_multitenancy_config from the update-host-based-configurations recipe and causing configure lifecycle event failures

v4.4.5

OPC-682: paella7 paths
Add engage host id to use in IC video urls
Self-service standalone video editor role configuration

v4.4.4

Maven 3.8.8

v4.4.3

OPC-848: Immersive classroom url added to organization props.

v4.4.2

IC: OPC-851 *.dcex.harvard.edu cors support from Prod OC Engage

v4.4.1

IC: *.dcex.harvard.edu cors support, requires nginx update on engage
IC: OC admin ability to reference a specific IC server (for mpId trim updates)

v4.4.0

Add stack name to org props in order to display it on the admin ui
Fix for admin-ui series properties endpoint access

v4.3.1

Hot-fix for maven 3.8.7

v4.3.0

Add configuration for HarvardKey login.

v4.2.0

Remove legacy auth configuration
Add thread id to log

v4.1.1

Remove uneccesary template

v4.1.a

Enable logging aws metrics
Revert cleanup config for collection not used anymore (asset_aws_s3)

v4.0.0

Changes for Opencast 11.
Bump version of Java and Maven.
Admin node can use https. Needs a change in the default cluster config (Admin layer).
Opencast configuration changes (elastic search, lti, active mq, watson, others)

v3.2.0 - 05/26/2022

DEPLOY-77: updated buildspec.yml, new artifact handling & notifications
Fixing/adding a couple of metadata fields that foodcritic was complianing about
oops. fixing the fix. Don't mess with those metadata version values, i guess.

v3.1.1 - 04/29/2022

Fix typo in deploy-engage recipe (:presentation vs :admin)

v3.1.0 - 04/28/2022

DEPLOY-76: allow use of prebuilt oc artifacts during deployment

v3.0.6 - 04/27/2022

Fix download url for maven

v3.0.5 - 03/22/2022

DEPLOY-75: fix exec-dist-upgrade recipe to use yum

v3.0.4

OPC-668: New cluster config param: production_management_email

v3.0.3 - 02/07/2022

OPC-649: New retract recordingAction endpoint GET mh_org access

v3.0.2 - 02/04/2022

DEPLOY-64: update path to public root cert bundle; ensure we're using postfix
DEPLOY-64: ensure tls enabled at all points

v3.0.1 - 01/20/2022

DEPLOY-63: fixed path for elasticsearch options file

V3.0.0 - 01/10/2022

DEPLOY-20: transition opencast opsworks deployment to amazon linux This commit rewrites a substantial number of our custom chef recipes to be compatible with Amazon Linux 2018.03, along with a few other changes here and there.

At a high-level, the changes include:
- standardizing on python3 and defining a common method of creating virtualenvs
- removing all of the unused capture-agent-manager stuff
- updating the package repos and keys for elasticsearch, et al
- removing cloudwatch logs agent setup (we're using the opsworks baked-in functionality now)
- removed newrelic installation and configuration
- removed ubuntu-specific recipes
- installing our own run-one script
- properly implemented crowdstrike install recipes
- removed a bunch of configuration references to 1x stuff
- fixed bug in dont_start_opencast_automatically flag behavior
- added an override mechanism for dont_start_opencast_automatically

v2.27.1 - 11/01/2021

OPC-637 Metasynch configurable distribution. Admin node.

v2.26.0 - 09/29/2021

REQUIRES MANUAL RECIPE RUN REQUIRES EDITS TO CLUSTER CONFIG MI-196: update syntax for esm registration command The ubuntu_advantage_esm block of the custom json must be updated from this:

"ubuntu_advantage_esm": {
  "user": "...",
  "password": "..."
},

to this:

"ubuntu_advantage_esm": {
  "token": "...",
},

For an existing cluster you must then execute the modified recipe:

./bin/rake stack:commands:execute_recipes_on_layers layers="Admin,Engage,Utility,Storage,Analytics,Workers" recipes="oc-opsworks-recipes::enable-ubuntu-advantage-esm"

v2.25.0

OPC-554 Configuration for porta auth and porta migration services.

v2.24.0

OPC-554: Add configuration for porta series metadata update. (#244)

v2.23.0

Changed notification email

v2.22.0

OPC-596 Increase activemq memory (affects stage and prod).

v2.21.0

OPC-581 Configuration for the video export tool (send email with s3 bucket presigned urls).

v2.20.1

OPC-580: redirect http requests to engage to https when host is *.harvard.edu

v2.19.0 - 12/17/2020

MI-193: wipe /root/.cache/pip/wheels directory during cloudwatch log agent install
OPC-577 remember-me bean for OC upstream security patches (companion OC patch)

v2.18.0 - 11/02/2020

OPC-568 Watson url from cluster config. Decrease check interval after expected job duration. (#235)

v2.17.0 - 06/15/2020

OPC-543 admin-ng series property rights for producers
MI-187 don't let missing old ffmpeg fail the provisioning

v2.16.3 - 04/27/2020

OPC-498 update ffmpeg to 4.2

v2.16.2 - 04/14/2020

MI-185 use the aurora cluster's reader endpoint for the mysqldump

v2.16.1 - 04/09/2020

OPC-510 otherpubs helix value change email notification configurable on-off

v2.15.0 - 3/20/2020

OPC-496 On-demand zoom ingest: add configuration for zoom ingester endpoint.

v2.14.0 - 1/24/2020

OPC-446 Helix Googlesheets config for Otherservice Requires new cluster config params and update recipe

  helix_googlesheets, {
    enabled: <if false, the helix google sheets managed config is not created>,
    defaultduration_min: <no longer used, it's default dur of a helix pub>,
    token: <the service api, ask the dev team>,
    cred: <the old oauth2, ask the dev team>,
    helix_sheet_id: <the Helix Googledoc spreadsheet id of choice>
  }

v2.13.0 - 1/21/2020

OPC-357 HLS-VOD ffmpeg version and logging
Requires ffmpeg push with ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::install-ffmpeg"

v2.12.0 - 11/22/2019

PSUPP-3145 playeroldRedirect auth for special test-local redirect page

v2.11.0 - 10/10/2019

OPC-381-lamda-timeout-conf Search transcripts config

v2.10.1 - 9/26/2019

OPC-381 search-transcripts endpoint access update
MI-126 adjust nfs availability cron and metric alarm to only trigger if two failures over two 60s periods to avoid false positives

TO BE RELEASED

OPC-357 HLS-VOD ffmpeg version and logging

v2.10 - 8/22/2019

OPC-359 search-transcripts
OPC-139 LTI Oauth config change from Upstream
OPC-334 many-embedded Added configuration to auth service
MI-164: set nginx logrotate to only keep 30
MI-171: nginx config reload needs to watch for changes to both the ssl key and cert

v2.9 - 7/19/2019

OPC-371 report a problem template updates

v2.8 - 7/05/2019

OPC-363 bug report feature
OPC-344 Configuration for CA notification via CATracker (start/stop now) I. Requires adding this to cluster configs custom_json: "capture_agent_sync": { "url": "https://:<low level username's pasword>@catracker.<dev or prod catracker, preferably prod>.edu/tracker/notify/#{name}", "threshold": "180" },

II. Required manually run recipe for OPC-344 Run: ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin" recipes="mh-opsworks-recipes::configure-activemq" To push template change: templates/default/activemq.xml.erb

             <queue physicalName="SCHEDULER.CaptureAgentSync" />
             <queue physicalName="SCHEDULER.Liveschedule" />
           </forwardTo>

v2.7.1 - 04/26/2019

Fix for MI-162: ubuntu advantage recipe now included by a recipe that runs on all layer types

v2.7.0 - 04/26/2019

MI-162: new recipe to enable ubuntu advantage esm

v2.6.0 - 04/23/2019

OPC-340: producer permission for DCE stop-start scheduler endpoints

v2.5.0 - 04/18/2019

OPC-314: editor shortcut display config
MI-160: analytics harvester install, don't create virtualenv if it exists

v2.4.2 - 04/02/2019 (hotfix)

increase execution timeout for the awslogs-agent-setup.py script to 300s

v2.4.1 - 03/06/2019

MI-159: fix /etc/filename hostname value

v2.4.0 - 02/27/2019

MI-156: Update engage nginx to allow requests from localhost for OpencastAvailable metric
MI-157: Re-disable proxy buffering for the admin node's /assets path
OPC-247: ibm watson creditials changes

v2.3.0 - 02/13/2019

MI-153: re-enable proxy buffering of upstream responses; ignore client abort

v2.2.0

MI-152: make instances failed and rds alarm names less misleading
OPC-259 publisher and instructor name missing in email message body

v2.1.0 - 01/25/2019

OPC-226: restrict engage admin access to whitelisted hosts

v2.0.0 - 01/11/2019

MI-97: update mysql alarms for rds aurora cluster
OPC-226: redirect http -> https in engage nginx configuration
MI-125: rename maven cache archive to distinguish 5x vs 1x
OPC-219: remove creation of etc/opencast.conf and allow enabling java debugging via bin/setenv
MATT-2324: In 5x, it's Ok remove LTI process filter proxy workaround. Has companion OC 5x change.
MI-117: recipe for installing the linux-aws kernel package
MI-133: remove recipe for building enhanced networking kernel module
updating some references to the "dont_start_matterhorn_automatically" flag
MI-138: generate a cloudwatch metric based on Opencast's job load
MI-140: cluster reset scripts now use correct solr indices path
MI-143: fix for MYSQLServerAvailable metric generator
MI-141: new workflow and job_load metrics

notes merged from 5.x development

11/02/2018

ensure dist-upgrade runs with updated package repo and non-interactive mode
mo-scaler now uses virtualenv to better manage/isolate python dependencies
install additional network/io performance testing tools

10/12/2018

MI-114: cloudwatch metric for number of online workers

09/27/2018

OPC-155 Editor-shortcuts Waveform bigger encoding change
OPC MATT-2406 only admin performs hearbeat checks config change
OPC-59-metasynch Added ROLE_DCE_METASYNCH to metasynch endpoints

09/10/2018

MI-106: improvements to activemq install

09/07/2018

enable activemq install for all-in-one local clusters

08/22/2018

upgrade pip during moscaler install to handle improved requirements.txt dependency format

08/10/2018

activemq & nginx service fixes:
- update to latest, stable activemq
- fix for MI-86: activemq/opencast communication borked on first cluster start
- reworking of nginx recipes to ensure config reloads happen on e.g. ssl cert updates

08/02/2018

cherry-picking a fix for the ua-harvester dependency installation note: this commit included some unrelated CHANGELOG updates

07/23/2018

fixing the enabling logic for the earlier newrelic changes

07/18/2018

only run the squid install recipe if external storage (i.e. zadara) is being used

07/12/2018

REQUIRES MANUAL RECIPE RUN install latest stable nginx & disable proxy_request_buffering for admin nginx to allow direct streaming of uploads. For an existing cluster you can either reboot the cluster, or, if that's not desired, the various nginx recipes must be run:

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin" recipes="oc-opsworks-recipes::configure-nginx-proxy"

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Engage" recipes="oc-opsworks-recipes::configure-engage-nginx-proxy"

If you have an analytics node:

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Analytics" recipes="oc-opsworks-recipes::configure-elk-nginx-proxy"

07/11/2018

REQUIRES MANUAL RECIPE RUN REQUIRES EDITS TO CLUSTER CONFIG OPC-97: install newrelic via recipe. To enable newrelic in a cluster the custom json must include a block like:
```
  "newrelic": {
    "agent_version": "4.2.0",
    "admin": {
      "key": "xxxxx"
    },
    "engage": {
      "key": "xxxxx"
    },
  },
```
For an existing cluster you must then execute the install recipe, either manually (command below) or by adding it to the layer's "setup" recipe run list in the cluster config and restarting the cluster instances:
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin,Engage" recipes="oc-opsworks-recipes::install-newrelic"
```
Then you must also execute a deploy so that the opencast init scripts get updated as well: ./bin/rake deployment:redeploy_app opencast:restart

(end 5.x dev notes)

v1.34.1 - 11/02/2018

ensure the dist-upgrade runs with updated package repo && in non-interactive mode

v1.34.0 - 10/29/2018

OPC-149 Otherpubs config to combine Opencast pubs on pub listing

v1.33.0 - 10/15/2018

MI-114: cloudwatch metric for number of online workers

v1.32.1 - 09/20/2018

OPC-149, Otherpubs config to combine Opencast pubs on pub listing
REQUIRES MANUAL RECIPE RUN set rds write iops alert threshold based on allocated storage & bump cpu usage alert to a more reasonable threshold
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Ganglia" recipes="mh-opsworks-recipes::create-mysql-alarms"
```

v1.32.0 - 08/18/2018

LDAP integration (OPC-66).
fix ua-harvester dependencies installation. No action required; only affects new clusters.
REQUIRES MANUAL_RECIPE RUN - disable deprecated/insecure ssl protocols in nginx config For existing dev clusters manual recipe runs are not required; a cluster reboot will suffice as the nginx configs get updated during the opsworks "setup" phase. For clusters where it is not desirable to reboot, the following recipes must be run:
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin" recipes="mh-opsworks-recipes::configure-nginx-proxy"

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Engage" recipes="mh-opsworks-recipes::configure-engage-nginx-proxy"
```
If the cluster has an Analytics node run the following as well:
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Analytics" recipes="mh-opsworks-recipes::configure-elk-nginx-proxy"
```

v1.31.1 - 04/19/2018

REQUIRES MANUAL _RECIPE RUN updated loggly TLS certificate

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin,Workers,Engage" recipes="mh-opsworks-recipes::rsyslog-to-loggly"

v1.31.0 - 3/05/2018

ruby gem version revert because CI requires a different version that we use to build in production (security issue not relevant because it's for Travis CI)
MATT-2406 limit service checking to just be from the service dispatcher to reduce unnecessary work and network traffic from the workers and engage nodes

v1.30.0 - 1/12/2018

'yajl-ruby' gem version update to address security vulnerability
added buildspec.yml file to allow automated CodeBuild builds. See the harvard-dce/mh-opsworks-builder project.
pinned 'mixlib-archive' gem to previous version because of broken release

v1.29.0 - 10/26/2017

REQUIRES MANUAL RECIPE RUN Update useraction harvester & elasticsearch recipes for new, combined zoom + useraction analytics harvester
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Analytics" recipes="mh-opsworks-recipes::install-ua-harvester"
```

v1.28.0 - 10/06/2017

MATT-2464 nginx error log to INFO level for fileupload debugging

v1.27.0 - 09/25/2017

create cloudwatch log group for utility node's squid logs

REQUIRES MANUAL RECIPE RUNS MI-74: nginx config performance improvements

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin,Workers" recipes="mh-opsworks-recipes::configure-nginx-proxy"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Engage" recipes="mh-opsworks-recipes::configure-engage-nginx-proxy"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Analytics" recipes="mh-opsworks-recipes::configure-elk-nginx-proxy"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Utility" recipes="mh-opsworks-recipes::configure-capture-agent-manager-nginx-proxy"

REQUIRES MANUAL RECIPE RUN MI-73: recipe and script to pull capture agent logs w/ rsync, push to cloudwatch

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Utility" recipes="mh-opsworks-recipes::configure-capture-agent-cwlogs"

v1.26.0 - 08/24/2017

Allow exit status of '255' on cloudwatch log group creation to get rid of errors due to ResourceAlreadyExistsException
cloudwatch logs agent install requires python-dev package
refactor of moscaler install to allow separate recipes for creating/removing cron_d resources. this is to facilitate moscaler pause/resume rake tasks in mh-opsworks.

v1.25.0 - 08/17/2017

copy dce-config encodings engage-image.properties

v1.24.0 - 08/10/2017

add new recipe for executing apt-get dist-upgrade to facilitate kernel package updates
update to latest recommended version of ixgbevf enhanced networking driver
MATT-2355: added deployment config property for enabling/disabling workspace cleanup
MATT-2388: spring config changes related to social annotations reports
don't install yourkit by default on any nodes

v1.23.1 - 07/25/2017

don't install yourkit on non-mh nodes
REQUIRES MANUAL RECIPE RUN bump evaluation period from 5m to 10m on RDS cpu and queue depth alarms. These metrics are occasionally experiencing short (<= 5m) spikes which appear to be benign but are triggering the alarms.
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Ganglia" recipes="mh-opsworks-recipes::create-mysql-alarms"
```

v1.23.0 - 05/18/2017

redirect output from MH start-on-boot cron entry to syslog to prevent unnecessary emails
Allow enabling of G1 Garbage Collection method and YourKit profiler agent. Both are off by default. To enable, add either/both of the following attributes respectively to the cluster config's custom json:

    ...
    "enable_G1GC": true,
    ...
    "enable_yourkit_agent": true,
    ...

set JVM -Xxms value based on ratio to-Xxmx depending on node type
fetch awslogs-agent-setup.py cloudwatch logs setup script from shared assets bucket to ensure working/tested version

v1.22.0 - 04/21/2017

MI-63: remove nodejs install recipe as it is no longer needed
REQUIRES EDITS TO CLUSTER CONFIG MI-62: create cron entry to sync ibm watson transcript results to s3. Prior to deployment, the custom_json block in cluster config should be updated with the name of the target s3 bucket:
```
  "ibm_watson_transcript_sync_bucket_name": "dce-ibm-watson-transcripts"
```

v1.21.2-hotfix - 03/13/2017

hotfix to address nginx proxy config mixup. The previous manual recipe run instructions should have been:

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin" recipes="mh-opsworks-recipes::configure-nginx-proxy"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Engage" recipes="mh-opsworks-recipes::configure-engage-nginx-proxy"

v1.21.1 - 03/10/2017

REQUIRES MANUAL RECIPE RUN Add $request_time value to nginx access log events Manual recipe run should come after deploy and an update_chef_recipes.
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin,Engage" recipes="mh-opsworks-recipes::configure-nginx-proxy"
```

v1.21.0 - 02/13/2017

MATT-2231 login for annots summary path

v1.20.0 - 02/06/2017

MATT-1929 Create inbox for republish with re-trimming.

v1.19.0 - 01/30/2017

MATT-2245 Enable DCE annot property endpoint
REQUIRES EDITS TO CLUSTER CONFIG include auth key when deploying mh auth properties file. auth_key value must be added to prod cluster config prior to deployment.

v1.18.0 - 01/19/2017

REQUIRES EDITS TO EXISTING CLUSTER CONFIG
MATT-2046 Enable the vanilla Opencast LTI endpoint, requires new lit_oauth custom_json
replace with the iCommons token from the private file (or create a new one from the iCommons APi webpage)
replace with the consumerkey and shared secret from the private file "stack": { "chef": { "custom_json": { "icommons_api_token": "replace-with-an-icommons-api-token-or-manually-create-series-mappings", "lti_oauth": { "consumerkey": "dce_lti_oauth_consumer_key", "sharedsecret": "dce_lti_oauth_sharedsecret" }, ...

v1.17.0 - 01/10/2017

update CA hack: 53chur-l01 now points to new prod akami stream id (same as byerly)
skip cloudwatch log setup for local clusters; fix rsyslog restart (MATT-2238)
REQUIRES MANUAL RECIPE RUN Crowdstrike falcon host installation. The following command is an example of what would be run on the prod cluster. For dev clusters, adjust the layers list according to what nodes are present.
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin,Engage,Utility,Storage,Analytics" recipes="oc-opsworks-recipes::install-crowdstrike"
```

v1.16.0 - 12/8/2016

REQUIRES EDITS TO EXISTING CLUSTER CONFIG Update the prod DNS certificate, see Confluence release notes
added additional packages needed to compile mo-scaler required python packages
Force reinstallation of elasticsearch plugins to avoid version mismatches.

1.15.0 - 11/18/2016

REQUIRES EDITS TO CLUSTER CONFIG Configuration for the new Matterhorn ibm watson transcription service (service credentials).
Changed location of temporary zip files to Zadara to avoid cross-device link errors. Zip operations are executed when republishing and failing a workflow.
MATT-2215 add threshold config params to log connection durations in hudce-auth and leg-otherpubs (sys web msg)
REQUIRES MANUAL RECIPE RUN Enable dynamic scripting in elasticsearch.yml

1.14.1 - 11/03/2016

add retries to setting of cloudwatch log retention to avoid race condition when previous log group creation is still processing

1.14.0 - 10/28/2016

REQUIRES EDITS TO CLUSTER CONFIG Refactoring of analytics node recipes:
- move install-ua-harvester to setup list (after install-elasticsearch)
- remove configure-logstash-kibana from configure list
- add configure-ua-harvester to configure list

REQUIRES MANUAL RECIPE RUN Adding redis service to analytics node for useraction harvest caching of episode data

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Analytics" recipes="oc-opsworks-recipes::install-ua-harvester"

1.13.0 - 10/20/2016

explicitly set the $HOME env variable when pip-installing the ca manager dependencies.
remove submodule init during Matterhorn code deploy (MATT-2173)
update CA hack: byerly-013 now points to new prod akami stream id

REQUIRES EDITS TO CLUSTER CONFIG REQUIRES MANUAL CHEF RECIPE RUN New recipe, install-cwlogs, that installs and configures the AWS Cloudwatch Log Agent. mh-opsworks-recipe::install-cwlogs should be added to the setup list of the Admin, Engage and Workers (and Analytics/Utility, if present) layers.

Manual recipe runs should come after any MH release/deploy and an update_chef_recipes.

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin,Engage,Workers" recipes="oc-opsworks-recipes::install-cwlogs"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Analytics" recipes="oc-opsworks-recipes::install-cwlogs,oc-opsworks-recipes::install-elasticsearch,oc-opsworks-recipes::install-kibana"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Utility" recipes="oc-opsworks-recipes::install-cwlogs,oc-opsworks-recipes::install-capture-agent-manager,oc-opsworks-recipes::configure-capture-agent-manager-nginx-proxy"

1.12.0 - 10/7/2016

Akamai account limit hack. See MATT-2182.

REQUIRES MANUAL CHEF RECIPE RUN Add new custom metric script for feeding jvm stats to cloudwatch.

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin, Workers, Engage" recipes="oc-opsworks-recipes::install-custom-metrics"

REQUIRES MANUAL CHEF RECIPE RUNS Updates to ELK pipeline components and apt pinning of package major versions. Config support and new cron entry for mh-user-action-harvester load_episodes command.

Manual recipe run should come after any MH release/deploy and an update_chef_recipes.
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Analytics" recipes="oc-opsworks-recipes:install-elasticsearch,oc-opsworks-recipes::install-logstash-kibana,oc-opsworks-recipes::install-ua-harvester"
```

1.11.0 - 9/8/2016

moscaler_release now defaults to "master"

REQUIRES MANUAL CHEF RECIPE RUN Additional helpful utils to be installed by default

  ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::install-utils"

prevent autofs restart only if existing mount matches the storage hostname. this enables use of the nfs-client recipe for switching storage nodes or zadara vpsas
REQURES EDITS TO THE CLUSTER CONFIG REQUIRES MANUAL CHEF RECIPE RUNS: Capture agent time drift metrics generation via utility node. Update the cluster config to include the ca_private_ssh_key value in the caputure_agent_manager custom json block, add the oc-opsworks-recipes::install-capture-agent-timedrift-metric recipe to the utility node's setup phase, then manually run:
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Utility" recipes="oc-opsworks-recipes::install-ca-timedrift-metric"
```

1.10.0 - 8/12/2016

REQURES EDITS TO THE CLUSTER CONFIG REQUIRES MANUAL CHEF RECIPE RUNS: Changes to moscaler configuration deployment. Updated config will be detailed in the MH release notes. After an update_chef_recipes do a manual recipe exec:
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Ganglia" recipes="oc-opsworks-recipes::install-moscaler"
```
REQURES EDITS TO THE CLUSTER CONFIG REQUIRES MANUAL CHEF RECIPE RUNS: Fixes to the recipe/script that provides the MatterhornJobsQueue metric. Cluster config should be updated to include the install-job-queued-metrics recipe. See MH release notes for more deets. After an update_chef_recipes do a manual recipe exec:
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Ganglia" recipes="oc-opsworks-recipes::install-job-queued-metrics"
```

1.9.0 - 7/21/2016

REQUIRES EDITS TO THE CLUSTER CONFIG REQUIRES MANUAL CHEF RECIPE RUNS: Renames install-ec2-scaling-manager to install-moscaler and updates the recipe to allow control of the scaling strategy (time vs autoscale) as well as disabling entirely. See README.horizontal_scaling.md in the mh-opsworks repo for details.

1.8.0 - 7/13/2016

Fixed notification email
Added live_monitor_url config for live akamai stream monitoring "stack": { "chef": { "custom_json": { "live_monitor_url": rtmp:akamai_server:port/../#{caName}/...

1.7.0 - 7/5/2016

Added stack short name to config.properties so that we can show the cluster name when sending notification emails.

1.6.0 - 6/24/2016

Configuration for the newrelic is now different for each layer (admin,work,engage) "stack": { "chef": { "custom_json": { "newrelic": { "admin": { "key": "admin_newrelic_key" }, "workers": { "key": "worker_newrelic_key" } } ... If the newrelic key or a layer name is omitted, new relic will not be run for the stack or the layer. all-in-one deployment will use admin key.

1.5.0 - 6/10/2016

Configuration for the new Matterhorn aws s3 file archive service (s3 bucket name).
[OPTIONAL] Changes to the install-ua-harvester recipe. Changed naming path of the s3 bucket used to store the useraction harvester's last action timestamp value. Only relevant for analytics node and only affects new clusters. Also bumped harvest batch size from the default 1000/per

1.4.0 - 6/8/2016

[OPTIONAL] Recipes to install and configure the utility node to host the capture-agent-manager flask-gunicorn webapp. The utility node is optional and only for DCE production cluster.

[OPTIONAL] EDITS TO THE CLUSTER CONFIG REQUIRES MANUAL CHEF RECIPE RUNS: recipes, etc to install and configure the utility node to host the capture-agent manager flask-gunicorn webapp. See README.capture-agent-manager.md in mh-opsworks for setup instructions.

The manual recipe run can be executed any time since they affect only the utility node

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Utility"
  recipes="oc-opsworks-recipes::create-capture-agent-manager-user,oc-opsworks-recipes::create-capture-agent-manager-directories,oc-opsworks-recipes:install-capture-agent-manager-packages,oc-opsworks-recipes:install-capture-agent-manager,oc-opsworks-recipes:configure-capture-agent-manager-gunicorn,oc-opsworks-recipes:configure-capture-agent-manager-nginx-proxy,oc-opsworks-recipes:configure-capture-agent-manager-supervisor"

1.3.3 - 5/19/2016

Fix oc-opsworks-recipes::install-ec2-scaling-manager to correctly update the cron jobs when run. Implement basic serverspec tests to ensure the default attributes and a run with customized attributes works. Nothing to run.
Fix all-in-one node provisioning, primarily for local vagrant development. Nothing to run.
Only update apt repo if it's more than 30 minutes old. Nothing to be run.
Recipe level docs. There are no functional changes in this feature. Nothing to run.
Implement foodcritic linting via travis. Nothing to run.
Use the service resource to idiomatically start services. Nothing to run, just make sure that matterhorn starts correctly after a deploy.
Fix and improve test-kitchen bats tests. Nothing to run.
Better docs.

1.3.2 - 4/22/2016

Local opsworks cluster support, in concert with changes made to mh-opsworks proper. Modify some chef recipes to only run functions when on AWS. Nothing is required to be run outside of a normal deploy. See README.local-opsworks.md in mh-opsworks for full info on how this feature works.

1.3.1 - 4/14/2016

Start matterhorn at boot via cron's @reboot functionality. Runs automatically as part of a deploy.
REQUIRES MANUAL CHEF RECIPE RUNS * MUST BE RUN BEFORE THE DEPLOY* Change the dont_start_matterhorn_after_deploy stack custom_json override to dont_start_matterhorn_automatically because it has a larger scope now. When set to true, this allows you to start ec2 instances without starting the matterhorn daemon. This hooks into the init script (no other way, really) and will prevent matterhorn from booting until you unset it or set it to false. That's a lot of scary words for something that'll only impact you if you explictly use it. If you want to start matterhorn after disabling it with the variable, you: 1) remove or edit the variable in your cluster config via "cluster:edit", and 2) run "matterhorn:restart" via the mh-opsworks tooling. Using the mh-opsworks tooling will propagate the custom_json change and ensure matterhorn can start properly.
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin, Workers, Engage" recipes="oc-opsworks-recipes::install-mh-base-packages"
```

1.3.0 - 4/7/2016

REQUIRES MANUAL CHEF RECIPES RUNS MUST BE RUN BEFORE THE DEPLOY Update the version of node on the cluster to work around an npm / gzipping bug.

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin, Workers, Engage" recipes="oc-opsworks-recipes::install-mh-base-packages"

REQUIRES MANUAL CHEF RECIPE RUNS Hang more alarms on database metrics to catch problems relating to disk IO. Lower the CPU threshold. This can be done any time before or after a deploy.
```
  # This is currently on the "Ganglia" layer in prod,
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Ganglia" recipes="oc-opsworks-recipes::create-mysql-alarms"
```

REQUIRES MANUAL CHEF RECIPE RUNS Add a metric and alarm to check if instances fail to start up. This can be done any time before or after a deploy.

  ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::install-custom-metrics,oc-opsworks-recipes::create-alerts-from-opsworks-metrics"

OPTIONAL EDITS TO THE CLUSTER CONFIG REQUIRES MANUAL CHEF RECIPE RUNS: Recipes, etc for setting up Analytics layer/node. See README.analytics.md in mh-opsworks for setup instructions. The manual recipe run can be executed any time after a successful deploy.
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Engage" recipes="oc-opsworks-recipes::configure-engage-nginx-proxy"
```

1.2.0 - 3/23/2016

Allow for cluster state to be turned into a "seed file" and applied to another cluster. See README.cluster_seed_files.txt in https://github.com/harvard-dce/mh-opsworks/ for more information on how this feature works. Does not require specific chef recipes be run for a deployment.
Fix the squid3 proxy ACL to correct rules for s3 endpoints - the zadara docs were very, very wrong. This does not require an explicit chef recipe run.

REQUIRES MANUAL CHEF RECIPE RUNS: Updated loggly TLS certificate. Manual recipe run can be executed any time after a successful deploy.

  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin, Workers, Engage" recipes="oc-opsworks-recipes::rsyslog-to-loggly"

1.1.6 - 3/3/2016

REQUIRES EDITS TO THE CLUSTER CONFIG REQUIRES MANUAL CHEF RECIPE RUNS: Before doing a deploy and after the morning scaled instances have started, you must edit the production cluster config custom_json to set:
```
  "matterhorn_repo_root": "/opt/matterhorn/deploy",
  "nginx_log_root_dir": "/opt/matterhorn",
```
matterhorn_repo_root already exists and should be updated - nginx_log_root_dir is new and should be created. Proceed to update chef recipes and do your deploy normally after that.

This new feature clarifies the name of the deployed app and allows it to easily be moved into a separate subdirectory under /opt/matterhorn. The manual recipe run below can be run at any time after a successful deploy.
```
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin, Workers, Engage" recipes="oc-opsworks-recipes::remove-legacy-deploy-dir"
```

REQUIRES MANUAL CHEF RECIPE RUNS: Open up nginx logs to be world-readable by default. This happens post daily log rotation

  # After the recipes have been updated. . .
  ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::configure-nginx-proxy" layers="Admin, Workers"
  ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::configure-engage-nginx-proxy" layers="Engage"

1.1.5 - 2/25/2016

Create a squid3 proxy to be used by the zadara storage array to get to s3 for object store backups. See mh-opsworks for more information about how to use this recipe. This recipe is independent of deployments.
REQUIRES MANUAL CHEF RECIPE RUNS: Set the bash prompt to be in color and put the cluster name in there. When a cluster includes "production" in the name, make it red so that users know they're working in production. This is not tied to deployment and can be run at any time
```
  ./bin/rake stack:commands:execute_recipes_on_instances recipes="oc-opsworks-recipes::set-bash-as-default-shell"
```
REQUIRES MANUAL CHEF RECIPE RUNS: This recipe can be run at any time and is deploy independent. Switch to daily mysql backups as a way of having a historical record of database state. RDS gives you the ability to restore to any point in time back to the beginning of your retention period (30 days for production), obviating the value and pretty major resource drain hourly MySQL dumps impose on production. Be sure to confirm which layer you're running your mysql backups on before running the command below:
```
  # This is currently on the "Ganglia" layer in prod,
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Ganglia" recipes="oc-opsworks-recipes::install-mysql-backups"
```
Ignore 401s in newrelic logging.

1.1.4 - 2/18/2016

REQUIRES MANUAL CHEF RECIPE RUNS: Copy and symlink the nginx log dir to /opt/matterhorn/nginx. This should be registered into the engage layer's setup recipes and run whenever. It does not have to be run during a deploy.
```
  ./bin/rake stack:commands:execute_recipes_on_instances recipes="oc-opsworks-recipes::symlink-nginx-log-dir" hostnames="engage1"
```
Remove the /var/log/nginx_old directory after you've manually confirmed logs are going to the right place.
Allow java unit tests to be run during a (re)deploy. They do not run by default. If you want to run them by default for all deploys set skip_java_unit_tests to false in your cluster config's custom_json config. You can run unit tests manually via the deployment:redeploy_app_with_unit_tests rake task.

REQUIRES MANUAL CHEF RECIPE RUNS: Use maven3 instead of maven2, prepares us for matterhorn 1.6.x

Right before the deploy. . .

  ./bin/rake stack:commands:update_chef_recipes stack:commands:execute_recipes_on_layers layers="Admin, Workers, Engage" recipes="oc-opsworks-recipes::install-mh-base-packages"

REQUIRES MANUAL CHEF RECIPE RUNS Install a cloudwatch metric and alarm to track nfs mount space available. Does not require anything to be done during a deployment.

  ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::install-custom-metrics,oc-opsworks-recipes::create-alerts-from-opsworks-metrics"

Install awscli directly, without the awscli recipe. Simplifies dependencies significantly. Does not require manual chef recipe runs, it'll run automatically when needed by other recipes.

1.1.3 - 2/17/2016

HOTFIX - define dependencies better to ensure we stay with chef 11 compatible cookbooks. Fixes awscli cookbook which depends on the broken build-essential cookbook.

1.1.2 - 2/10/2016

Allow the "create-file-uploader-user" recipe to fail, to work around mysterious "usermod: no options" chef bug. Does not require anything to be done during a deployment. The recipe always works the first time it's run, it's just the second time that we sometimes see this failure.

1.1.1 - 2/3/2016

Allow java debug to be set to true or false via the java_debug_enabled stack custom_json parameter. It currently defaults to "true" (so, enabled) because most clusters are going to be development or testing clusters.
Install the newrelic agent when '{"newrelic": {"key": "your key"}' is included in the stack's custom_json. See the main mh-opsworks README for info.
REQUIRES MANUAL CHEF RECIPE RUNS Make the mysql dumps more bandwidth efficient by using "--compress" on the mysqldump client and an inline gzip when saving to shared storage. This causes bandwidth during mysql dumps to drop by 4 or 5 times without an appreciable performance penalty. This recipe can by run at any time, be sure to confirm the instance your dumps are happening on by finding this recipe in your active cluster config.
```
  # This is currently on the "Ganglia" layer in prod,
  # the value below is for dev clusters.
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin" recipes="oc-opsworks-recipes::install-mysql-backups"
```
REQUIRES OPTIONAL CHEF RECIPE RUNS Installs the file_uploader user onto the instance of your choice. This allows you to use an rsync backchannel for uploads directly to the matterhorn inbox in combination with a couple cron jobs to copy and maintain these uploads. If you want to use this feature, add it to the setup lifecycle in the appropriate layer in your cluster config and then run this recipe. This recipe does not need to be run at any particular time relative to a deployment.
```
  # The current value for the admin node
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin" recipes="oc-opsworks-recipes::create-file-uploader-user"
```

1.1.0 - 1/28/2016

All clusters should now include a private_assets_bucket_name in their cluster config, which can be the same as the cluster_config_bucket_name as that's a private bucket too and includes some shared assets.

REQUIRES CHEF RECIPE RUNS Rotate nginx logs more aggressively than a week, but still keep a year. All this change does is install a new logrotate file, overriding the default. The recipes can be run whenever.

  # After the recipes have been updated. . .
  ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::configure-nginx-proxy" layers="Admin, Workers"
  ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::configure-engage-nginx-proxy" layers="Engage"

REQUIRES MANUAL CHEF RECIPE RUNS Remove the "ok-action" alarms - they are too chatty and not useful. These recipe runs can happen post-deploy.

  ./bin/rake stack:commands:execute_recipes_on_layers recipes="oc-opsworks-recipes::create-alerts-from-opsworks-metrics"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Ganglia" recipes="oc-opsworks-recipes::create-mysql-alarms"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin" recipes="oc-opsworks-recipes::install-mysql-backups"
  ./bin/rake stack:commands:execute_recipes_on_layers layers="Admin,Workers,Engage,Utility,Asset Server" recipes="oc-opsworks-recipes::nfs-client"

Be sure to double-check all layers and recipe combinations.

1.0.7 - 1/21/2016

Clean up s3 distribution configuration. Remove asset server hooks.
REQUIRES MANUAL CHEF RECIPE RUNS engage nginx uid cookie & usertracking session id + uid log for identifying shopping period viewers
REQUIRES MANUAL CHEF RECIPE RUNS Use mo-scaler --scale-available option

1.0.6 - 1/18/2016

Configuration for the new Matterhorn S3 distribution service.

1.0.5 - 1/14/2016

Hot fix for Otherpubs sevice config

1.0.4 - 1/14/2016

REQUIRES MANUAL CHEF RECIPE RUNS Send X-Forwarded-For header from the nginx proxies so that user-tracking gets correct client ip
Get rid of the no longer necessary ffmpeg package removal
Modify mh_default and otherpubs config for MATT-1822 auto populate admin UI upload

1.0.3 - 1/8/2016

Fix find's mtime invocation to ensure we're compressing and reaping files at the correct times. Requires that the MySQL backup installation chef recipe be run on the admin node.
Relax the matterhorn_availability alert to require errors over two 2 minute periods instead of just one.
Modify the matterhorn init script to check that the daemon started properly, working around the race condition related to bundle loading. See the relevant commit for details.

1.0.2 - 1/6/2016

Only set the org.opencastproject.server.maxload property on worker nodes - all other nodes can have an unlimited number of jobs.

1.0.1 - 1/5/2016

Allow the default "start matterhorn if it's not running after deploy" to be overridden via the :dont_start_matterhorn_after_deploy chef custom_json attribute.

1.0.0 - 1/4/2016

Initial release

Files

CHANGELOG.md

Latest commit

History

CHANGELOG.md

File metadata and controls

CHANGELOG

TO BE RELEASED

v5.1.2 - 6/24/2024

v5.1.1 - 6/14/2024

v5.1.0 - 6/12/2024

v5.0.7 - 3/7/2024

v5.0.6 - 2/2/2024

v5.0.5 - 12/11/2023

v5.0.4 - 10/26/2023

v5.0.3 - 10/04/2023

v5.0.2 - 09/12/2023

v5.0.1 - 09/08/2023

v5.0.0 - 08/21/2023

v4.4.6

v4.4.5

v4.4.4

v4.4.3

v4.4.2

v4.4.1

v4.4.0

v4.3.1

v4.3.0

v4.2.0

v4.1.1

v4.1.a

v4.0.0

v3.2.0 - 05/26/2022

v3.1.1 - 04/29/2022

v3.1.0 - 04/28/2022

v3.0.6 - 04/27/2022

v3.0.5 - 03/22/2022

v3.0.4

v3.0.3 - 02/07/2022

v3.0.2 - 02/04/2022

v3.0.1 - 01/20/2022

V3.0.0 - 01/10/2022

v2.27.1 - 11/01/2021

v2.26.0 - 09/29/2021

v2.25.0

v2.24.0

v2.23.0

v2.22.0

v2.21.0

v2.20.1

v2.19.0 - 12/17/2020

v2.18.0 - 11/02/2020

v2.17.0 - 06/15/2020

v2.16.3 - 04/27/2020

v2.16.2 - 04/14/2020

v2.16.1 - 04/09/2020

v2.15.0 - 3/20/2020

v2.14.0 - 1/24/2020

v2.13.0 - 1/21/2020

v2.12.0 - 11/22/2019

v2.11.0 - 10/10/2019

v2.10.1 - 9/26/2019

TO BE RELEASED

v2.10 - 8/22/2019

v2.9 - 7/19/2019

v2.8 - 7/05/2019

v2.7.1 - 04/26/2019

v2.7.0 - 04/26/2019

v2.6.0 - 04/23/2019

v2.5.0 - 04/18/2019

v2.4.2 - 04/02/2019 (hotfix)

v2.4.1 - 03/06/2019

v2.4.0 - 02/27/2019

v2.3.0 - 02/13/2019

v2.2.0

v2.1.0 - 01/25/2019

v2.0.0 - 01/11/2019

notes merged from 5.x development

11/02/2018

10/12/2018