Marcelo Ochoa

Since August 2017 We are moved to Medium, some of the reason are great described into this blog post "3 reasons we moved our startup blog to Medium". You are invited to see the new channel, greetings.

There is great post on Applying the first RU for Oracle Database 12.2.0.1 written by my friend Mike.
But this post is about on how to apply this important patch on Docker environment using official Oracle Docker images.
First, some time DBAs have fear on applying patches for two important thing:

• What happen if something goes wrong, how to rollback this process quickly on production
• Which is my downtime, also apply on production systems where minutes means money

Well, on Docker environment this two points are quickly answered:

• You could rollback the patch on binaries by simply stop your Docker container started with the RU image and go back using the official image, remember that on Docker environment bin/libs are part of the layered file-system where your container start and it acts like an SVN branch, if you want to go back with the changes made on the binary files modified by a patch you can rollback this change stopping your container and starting again using the previous used image.
• The downtime as I will show below is only the time involving by shutdown your database, and starting it again with the new image, the time for patching your binaries are NOT part of your downtime, when the database startup for the first time with a new patched image there is only a fixed time of datapatch utility which is called once.

Let see in action, first check that you have the two Oracle Database Release 2 images ready at your local repository:
If you have a running container with the official R2 base image oracle/database:12.2.0.1-ee, the start script may be look like:
note that your datafiles reside outside the container as usual to survive the stop/start/rm container life-cycle which obviously make sense for an production database.
Now I will patch this database with first RU patchset, for doing that a run command will point to the image patched with 26123830 patchset, here an screenshot:
here the process for applying a patch:
database downtime is about:

• 0m4.869s (docker rm -f demo)
• 0m8.503s (docker run with new image, ./run-demo-patch.sh file)
• 1m20s (time taken by SQL Patching tool starting time 12:43:06, ending time 12:44:26)

Not bad for my modest laptop.
But what are under the hood on the new patched imaged with the RDBMS RU included?
First there is an official script provided by Oracle to apply generic patches to the official image, here the link at GitHub, but this script doesn't take into account the post-processing task during first startup time, so I decided to provide mine using this Dockerfile and modified startDB.sh script.
Here the explanation of Dockerfile, first the Docker image tagged oracle/database:12.2.0.1-ee-26123830 is built on top of oracle/database:12.2.0.1-ee.
FROM oracle/database:12.2.0.1-eeFor building a new image with the patch applied on binary/libs file is simple as calling docker with:
docker build -t "oracle/database:12.2.0.1-ee-26123830" .Note that the process described above could be executed by an experimented DBA or a DevOps operator, and the process of stopping a production database and starting again with the patch installed could be executed by a novel DBA or any other operator.

Image: oracletechnocampus blogContinuing with my previous post about Doing full hot backups with RMan now is time for Data Pump.
The example is how to backup a full schema from a RDBMS running as Docker container, let see, a DB started using:
$ docker run --name test \
-p 1521:1521 -p 5500:5500 \
-e ORACLE_SID=TEST \
-e ORACLE_PDB=PDB1 \
-e ORACLE_PWD=Oracle2017\! \
-v /home/data/db/test:/opt/oracle/oradata \
oracle/database:12.2.0.1-eeOnce the DB is ready to use and assuming that there is an SCOTT schema with one table for testing the steps for doing a full schema backup using DataPump are:

• Create a RDBMS directory object to allows backups on the container external directory

$ docker exec -ti test sqlplus system/Oracle2017\!@pdb1
SQL> !mkdir -p /opt/oracle/oradata/backup
SQL> create directory bdir as '/opt/oracle/oradata/backup';
SQL> exit

• Start DataPump doing a full schema backup:

$ docker exec test expdp system/Oracle2017\!@pdb1 dumpfile=scott.dmp directory=bdir logfile=scott.log schemas=SCOTT
....
Total estimation using BLOCKS method: 13 MB
....
. . exported "SCOTT"."TEST_SOURCE_BIG"                   10.92 MB  135795 rows
Dump file set for SYSTEM.SYS_EXPORT_SCHEMA_01 is:
  /opt/oracle/oradata/backup/prod/scott.dmp
Job "SYSTEM"."SYS_EXPORT_SCHEMA_01" successfully completed at Wed Jun 7 19:13:18 2017 elapsed 0 00:01:03

Two weeks ago I presented in the ArOUG Cloud Day at Buenos Aires an introduction to Oracle Container Cloud Services (OCCS) and Oracle IaaS Compute management using console and API.
For this presentation I showed how to implement your Container Cloud Services (CCS) directly on top of IaaS compute.
Let's compare OCCS and my own CCS, here how they look like:
OCCS welcome pageMy CCS welcome pagethey look similar :), my own CCS is implemented using Portainer.io project.
Going deeper with both implementations I can resume the pros/cons as:
OCCS:

• Pros
• Easy manage
• Pre-configured templates
• Fast jump-start, 1-click deploy
• Support for using official repository of Oracle images
• Cons:
• Host OS not configurable, ej. disk, semaphores, etc
• Old Docker version 1.12
• Not Swarm support
• Basic orchestration features

My CCS:

• Pros
• Latest Docker/Swarm version 17.04.0-ce
• Full Swarm support, scaling, upgrade, load-balancing
• One-click deploy
• Public templates (Portainer.io and LinuxServer.io)
• Graphical scaling Swarm services
• Console  Log/Stats for services
• Full custom software/hardware selection
• Cons
• Oracle official repositories only available from command line
• Registry with login not supported in graphical interface, only command line
• A little complex deployment (scripts)

In my opinion the main problem of the OCCS is that you can't touch low level details of the implementation, so for example you can't change the host OS file /etc/sysctl.conf parameter vm.max_map_count=262144, then you will never get up and running a cluster of ElasticSearch 5.x version, more on this, supported version of Docker (1.12) is a bit old compared with latest features included in 17.04-0-ce.
On the other side my CCS best feature is that supports Swarm and command line operation if you connect to the host OS using ssh, with Swarm support you have a serious implementation for Docker data center solution, specially compared with other orchestration solutions like Kubernetes, see this great post about Docker Swarm exceeds Kubernetes performance at scale.
If you want to test by your self my CCS I extended the scripts of my previous post Managing Oracle Cloud IaaS - The API way to include a Portainer console up and running after deployment of the ES cluster and Swarm Visualizer.
The scripts are available at my GitHub repository, basically it deploys Portainer console at the end of the script deploy-cluster.sh using this command line:
docker run -d --name portainer -p 9000:9000 -v /var/run/docker.sock:/var/run/docker.sock portainer/portainer -H unix:///var/run/docker.sockSwarm nodesSwarm Services Scale up/down op.

I think many databases are going in production now using Docker environment, specially with the official support provides by Oracle when allows pulling Docker images from official Docker store.
If you are using a custom build image using Oracle's official scripts you can do a hot full backup using RMAN as is described in this post.
We will test using a container started as:
[mochoa@localhost ols-official]$ docker run -d --privileged=true --name test --hostname test --shm-size=1g -p 1521:1521 -p 5500:5500 -e ORACLE_SID=TEST -e ORACLE_PDB=PDB1 -v /etc/timezone:/etc/timezone:ro -e TZ="America/Argentina/Buenos_Aires" -v /home/data/db/test:/opt/oracle/oradata oracle/database:12.1.0.2-ee[mochoa@localhost ols-official]$ docker exec -ti test bash [oracle@test ~]$ sqlplus "/ as sysdba"First checks if your database is running in archive log mode:
SQL> SELECT LOG_MODE FROM SYS.V$DATABASE;
LOG_MODE------------ NOARCHIVELOG/opt/oracle/oradata/fast_recovery_area/${ORACLE_SID}/archivelog/yyyy_mm_dd/[mochoa@localhos ols-official]$ cat /etc/cron.daily/backup-db.sh #!/bin/bash docker exec test /opt/oracle/product/12.1.0.2/dbhome_1/bin/rman target=/ cmdfile='/opt/oracle/oradata/backup_full_compressed.sh' log='/opt/oracle/oradata/backup_archive.log'/home/data/db/test/backup_archive.log/opt/oracle/oradata/backup_archive.log

My last post shows how to deploy, for example, a Docker Swarm cluster at Oracle Cloud IaaS services.
This time I'll show you how to do the same without using the Web interface, I am an old fashioned sysadmin and I loves scripting and console tools ;)
The idea is to create a set of 5 nodes of Docker swarm cluster in five steps:

• Deploy storage (Disks)
• Deploy instances (VMs)
• Deploy Docker
• Deploy Docker Machine
• Deploy Swarm

Oracle provides a complete set of URL end points to manage all the infrastructure of your needs, here an example of the functionalities:

• Accounts
• ACLs
• Images
• IP Address
• Orchestrations
• Security
• Snapshots
• Storage
• Virtual NICs
• VPN

as you can see it virtually cover all your needs for managing the cloud infrastructure.
Before start you need some basic information to manage the cloud through the API.

• Username/Password cloud access
• Indentity domain

above information usually is included at the welcome mail when you register a cloud account, here an screenshot:
here a full explanation on how to deal with above information.
Also using the compute UI I uploaded a public ssh key named ubnt.pub, imports an Ubuntu 16.10 compute image from the Cloud Market place:

and allowing ssh access to my compute instances.

Once you have user/password/identity/API endpoint information you are able to log using CURl Linux command line required to test my shell examples, you can see on-line at my GitHub account.
Step one - Deploy storage create disk VMsdeploy-storage.sh script basically creates a boot disk for your VM and data disk to store Docker repository (aufs backed storage on top of ext4 file-system)
Here an example calling the scripts with proper arguments:
[mochoa@localhost es]$ export API_URL="https://api-z999.compute.us0.oraclecloud.com/"
[mochoa@localhost es]$ export COMPUTE_COOKIE="Set-Cookie: nimbula=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; Path=/; Max-Age=1800"
[mochoa@localhost es]$ ./deploy-storage.sh "$COMPUTE_COOKIE" "$API_URL"Step two - Create instances (VMs)[mochoa@localhost es]$ ./deploy-nodes.sh "$COMPUTE_COOKIE" "$API_URL"
-----------------
Creating Nodes...
-----------------
creating oc1 node...Step three - Deploy Docker softwareStep four - Deploy Docker Machine software (cloud side and local side)Step five - Init Docker SwarmTesting with Elastic Search images

Following with last post about how to deploy an Elastic Search (ES) cluster at Oracle Cloud facilities this post will play with scaling up and down a Swarm service.
We have an ES cluster deployed including a master node, four data nodes and one ingest node:
$ eval $(docker-machine env oc5)
[oc5]$ docker service ls
ID            NAME       MODE        REPLICAS  IMAGE
a32wxoob53hv  es_ingest  replicated  1/1       elasticsearch/swarm:5.0.0
crjhzsunyj12  es_master  replicated  1/1       elasticsearch/swarm:5.0.0
w7abwyokpp6d  es_data    replicated  4/4       elasticsearch/swarm:5.0.0Then using en example provides by the book Elastic Search 5.x Cookbook by Packt Publishing chapter 8 I uploaded at ingest node 1000 documents:
$ eval (docker-machine env oc1)
[oc1] docker ps
CONTAINER ID        IMAGE                       COMMAND                  CREATED             STATUS              PORTS                NAMES
98dd1e6882f4        elasticsearch/swarm:5.0.0   "/docker-entrypoin..."   About an hour ago   Up About an hour    9200/tcp, 9300/tcp   es_ingest.1.svszgk01qrupuk9o7uloakprp
[oc1] docker exec -ti es_ingest.1.svszgk01qrupuk9o7uloakprp bash
root@98dd1e6882f4:/usr/share/elasticsearch# curl -XHEAD 'http://192.168.0.8:9200/test-index?pretty=true'
root@98dd1e6882f4:/usr/share/elasticsearch# curl -XDELETE 'http://192.168.0.8:9200/test-index?pretty=true'
root@98dd1e6882f4:/usr/share/elasticsearch# curl -XHEAD 'http://192.168.0.8:9200/test-index?pretty=true'
root@98dd1e6882f4:/usr/share/elasticsearch# curl -H "Expect:" -XPUT 'http://192.168.0.8:9200/test-index?pretty=true' -d '{"mappings": {"test-type": {"properties": {"name": {"term_vector": "with_positions_offsets", "boost": 1.0, "store": "yes", "type": "text"}, "title": {"term_vector": "with_positions_offsets", "boost": 1.0, "store": "yes", "type": "text"}, "parsedtext": {"index": "analyzed", "term_vector": "with_positions_offsets", "boost": 1.0, "store": "yes", "type": "text"}, "tag": {"type": "keyword", "store": "yes"}, "date": {"type": "date", "store": "yes"}, "position": {"type": "geo_point", "store": "yes"}, "uuid": {"store": "yes", "type": "keyword"}}}}, "settings": {"index.number_of_replicas": 1, "index.number_of_shards": 5}}'
root@98dd1e6882f4:/usr/share/elasticsearch# curl -XPOST 'http://192.168.0.8:9200/test-index/_refresh?pretty=true'
root@98dd1e6882f4:/usr/share/elasticsearch# curl -XGET 'http://192.168.0.8:9200/_cluster/health?wait_for_status=green&timeout=0s&pretty=true'
root@98dd1e6882f4:/usr/share/elasticsearch# curl -H "Expect:" -XPUT 'http://192.168.0.8:9200/test-index/test-type/1?pretty=true' -d '{"in_stock": false, "tag": ["perspiciatis", "ullam", "excepturi", "ex"], "name": "Valkyrie", "date": "2014-07-28T16:46:01.668683", "position": {"lat": -17.9940459163244, "lon": -15.110538312702941}, "age": 49, "metadata": [{"num": 5, "name": "Korrek", "value": "3"}, {"num": 5, "name": "Namora", "value": "5"}, {"num": 26, "name": "Nighthawk", "value": "3"}], "price": 19.62106010941592, "description": "ducimus nobis harum doloribus voluptatibus libero nisi omnis officiis exercitationem amet odio odit dolor perspiciatis minima quae voluptas dignissimos facere ullam tempore temporibus laboriosam ad doloremque blanditiis numquam placeat accusantium at maxime consectetur esse earum velit officia dolorum corporis nemo consequatur perferendis cupiditate eum illum facilis sunt saepe"}'
....[oc5]$ docker service scale es_data=6
es_data scaled to 6

Continuing with my previous post about how to deploy a Docker Swarm Cluster at Oracle Cloud the idea now is how to deploy as example an Elastic Search cluster with these characteristics:

• One node working as master
• Two node working as data nodes
• One node working as ingest node

a complete set of scripts used in this post is at GitHub, click here for more details.
Swarm nodes preparationFirst We will tag our cluster of five nodes with some tags to control our ES cluster allocation.
To do that using docker-machine to submit commands to the Swarm cluster do, remember oc4 and oc5 are Swarm master capable nodes:
$ eval $(docker-machine env oc5)
[oc5] $ docker node update --label-add type=es_master oc5
[oc5] $ docker node update --label-add type=es_master oc4
[oc5] $ docker node update --label-add type=es_data oc3
[oc5] $ docker node update --label-add type=es_data oc2
[oc5] $ docker node update --label-add type=es_ingest oc1in my test official Elastic Search 5.0.0 image do not work well to recognize the primary host name on Swarm nodes, so I decided to use a custom ES image, We can build this image on each node with:
$ eval $(docker-machine env oc5)
[oc5]$ git clone https://github.com/marcelo-ochoa/docker.git
[oc5]$ cd docker/es
[oc5]$ docker build -t "elasticsearch/swarm:5.0.0" .
Step 1/6 : FROM elasticsearch:5.0.0
5.0.0: Pulling from library/elasticsearch
386a066cd84a: Pull complete
.....
Step 6/6 : CMD elasticsearch
 ---> Running in a6dd5fb17cf8
 ---> 2f1c7bfe6c67
Removing intermediate container a6dd5fb17cf8
Successfully built 2f1c7bfe6c67
[oc5]$ eval $(docker-machine env oc4)
[oc4]$ docker build -t "elasticsearch/swarm:5.0.0" .
.....
[oc1]$ docker build -t "elasticsearch/swarm:5.0.0" .[oc1]$ eval $(docker-machine env oc5)
[oc5]$ docker network create -d overlay --attachable --subnet=192.168.0.0/24 es_cluster[oc5]$ eval $(docker-machine env oc4)
[oc4]$ docker run -d \
--name viz \
-p 8080:8080 \
--net es_cluster \
-v /var/run/docker.sock:/var/run/docker.sock manomarks/visualizer:latestssh -i /home/mochoa/Documents/Scotas/ubnt -L8080:localhost:8080 ubuntu@oc4[oc4]$ eval $(docker-machine env oc5) [oc5]$ docker service create --network es_cluster --name es_master --constraint 'node.labels.type == es_master' --replicas=1 --publish 9200:9200/tcp --env ES_JAVA_OPTS="-Xms1g -Xmx1g"  elasticsearch/swarm:5.0.0 -E cluster.name="ESCookBook" -E node.master=true -E node.data=false -E discovery.zen.ping.unicast.hosts=es_master [oc5]$ docker service ls ID            NAME       MODE        REPLICAS  IMAGEo4x15kklu520  es_master  replicated  1/1       elasticsearch/swarm:5.0.0[oc5]$ docker service ps es_masterID            NAME         IMAGE                      NODE  DESIRED STATE  CURRENT STATE       ERROR  PORTSn4a8tibaqxpw  es_master.1  elasticsearch/swarm:5.0.0  oc5   Running        Running 17 seconds ago         because es_master was starting at OC5 node we can see the log output with:
[oc5]$ eval $(docker-machine env oc4)
[oc4]$ docker run -d \
  -p 9000:9000 \
  --net es_cluster \
  --env JAVA_OPTS="-Djava.net.preferIPv4Stack=true" \
  --name cerebro yannart/cerebro:latest[oc4]$ eval $(docker-machine env oc5)
[oc5]$ docker service scale es_data=4
es_data scaled to 4
[oc5]$ docker service scale es_ingest=2
es_ingest scaled to 2
[oc5]$ docker service ls
ID            NAME       MODE        REPLICAS  IMAGE
o4x15kklu520  es_master  replicated  1/1       elasticsearch/swarm:5.0.0
ue10tw8c64be  es_ingest  replicated  2/2       elasticsearch/swarm:5.0.0
w9cpobrxvay1  es_data    replicated  4/4       elasticsearch/swarm:5.0.0

As I mentioned in a previous post the Oracle Container Service seem to be little for my cloud expectativa.
But the Cloud Compute service is enough to implement my Docker Swarm implementation.
Why Docker Swarm?

• Is standard and supported by other cloud providers
• Is operated through command line using standard docker and docker-machine operations

When I tested Oracle Cloud Services I found three major problems:

• You can't modify base OS parameters
• Access to persistent storage didn't work for me
• The Docker version supported is a bit outdated

All these things can be avoided if you deploy your own Swarm cluster using Oracle Compute Cloud Services, here detailed instructions about how to do that, ADVICE: is a long post :)
Configuring the networkBefore starting creating a compute node (swarm node) is necessary to create a network definition to access using ssh and docker commands, here the step by step instructions using compute console:
create a security list  by clicking:

1. Network menu up
2. Security list left menu
3. Create Security list right up side
4. Choose a name (DockerNode in this example)
5. Choose Inbound Policy as Drop Packet/Outbound Policy Allow
6. Set a meaningful text description 

next defines a Security Application for Docker remote operations

1. Click on left menu Security Applications
2. Create Security Application
3. Name it as docker
4. Choose port type as TCP
5. Define a port staring range 2376 (docker node daemon)
6. Define a port end range 2377 (docker swarm cluster management communications)
7. Set a meaningful text description

next create a Security Rule allowing SSH and Docker ports

1. Click on Security Rules left menu
2. Click on Create Security Rule right up button
3. Name it (DockerAdminPort/SSHAdminPort in our case)
4. Choose a Security Application (ssh built in defined, docker defined previously)
5. Source will be Public Internet
6. Destination will be our previously defined Security List named DockerNode
7. Set a meaningful text description

1. Click on left side menu IP Network -> IP Neworks
2. Click on right up Create IP Network button
3. Name it in that case LocalInterconnect
4. Define a private IP range We choose 10.0.0.0/24
5. IP Exchange Not Set
6. Set a meaningful text description

1. On Network top menu
2. Click left side menu IP Reservations
3. Click on right up button Create IP Reservations
4. Named it, in our test case SwarmNode1..to..SwarmNode5
5. By now For Instance None

Launching five Swarm nodes

1. At the Compute Console choose Instance upper main menu
2. At the beginning there isn't instance running, click right button Create Instance

1. The wizard start showing pre-build Oracle Image, Choose Private images if you already use Ubuntu 16.10 image or Marketplace to find Ubuntu.16.10
2. Click on Selected button
3. Next choose go to Shape definition

1. Because we want a large cluster of cheap hardware (horizontal scaling) Shape is
2. General Purpose oc3 (1 OCPU/7.5 Gb memory)
3. Go to the Instance options

1. Clicking on Instance tab
2. Define a name for your instance
3. Define some meaningful tags
4. Choose a private SSH key (ubnt) or upload a new one using Add SSH Public Key button
5. Go to next step Network definition

1. On network preferences
2. Select Persistente Public IP Reservation option
3. Chose one of five public IP reservations defined previously
4. Chose our LocalInterconnect network previously defined (10.0.0.0/24)
5. Select our DockerNode Security List (ssh and docker enabled ports from outside)
6. Next go to Storage tab

1. On the Storage tab by default Ubuntu instance have pre-defined 10Gb boot disk
2. Click on button Add New Volume
3. Named it as Repo1 (other instance choose Repo2, Repo3 and so on)
4. Select 30 GB storage, this storage is for Docker local repository instance, some instance could be large, for example 12cR2 instance is 14 GB size
5. Set a meaningful text description

$ ssh -i /home/mochoa/Documents/Scotas/ubnt ubuntu@oc1ubuntu@oc1:~$ wget https://raw.githubusercontent.com/marcelo-ochoa/docker/master/es/oracle-cloud-node-conf.sh
--2017-03-08 14:53:11--  https://raw.githubusercontent.com/marcelo-ochoa/docker/master/es/oracle-cloud-node-conf.sh
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.48.133
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.48.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 757 [text/plain]
Saving to: 'oracle-cloud-node-conf.sh'
oracle-cloud-node-conf.sh          100%[===============================================================>]     757  --.-KB/s    in 0s    
2017-03-08 14:53:12 (201 MB/s) - 'oracle-cloud-node-conf.sh' saved [757/757]
ubuntu@oc1:~$ chmod +x oracle-cloud-node-conf.sh
ubuntu@oc1:~$ ./oracle-cloud-node-conf.sh ubuntu@oc1:~$ df -h
Filesystem      Size  Used Avail Use% Mounted on
udev            3.7G     0  3.7G   0% /dev
tmpfs           749M   17M  732M   3% /run
/dev/xvdb1      9.6G  2.6G  7.0G  27% /
tmpfs           3.7G  128K  3.7G   1% /dev/shm
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs           3.7G     0  3.7G   0% /sys/fs/cgroup
/dev/xvdc        30G  408M   28G   2% /var/lib/docker
/dev/xvdb15     105M  3.6M  101M   4% /boot/efi
tmpfs           749M     0  749M   0% /run/user/1000
ubuntu@oc1:~$ docker version
Client:
 Version:      1.13.1
 API version:  1.26
 Go version:   go1.7.5
 Git commit:   092cba3
 Built:        Wed Feb  8 06:57:21 2017
 OS/Arch:      linux/amd64
Server:
 Version:      1.13.1
 API version:  1.26 (minimum version 1.12)
 Go version:   go1.7.5
 Git commit:   092cba3
 Built:        Wed Feb  8 06:57:21 2017
 OS/Arch:      linux/amd64
 Experimental: true$ docker-machine create \
  --driver generic \
  --generic-ip-address=129.144.12.62 \
  --generic-ssh-key /home/mochoa/Documents/Scotas/ubnt \
  --generic-ssh-user ubuntu \
  oc1after adding our five node docker-machine repository is:
$ docker-machine ls
NAME        ACTIVE   DRIVER    STATE     URL                         SWARM   DOCKER    ERRORS
oc1         -        generic   Running   tcp://129.144.12.62:2376            v1.13.1  
oc2         -        generic   Running   tcp://129.144.12.36:2376            v1.13.1  
oc3         -        generic   Running   tcp://129.144.12.145:2376           v1.13.1  
oc4         -        generic   Running   tcp://129.144.12.14:2376            v1.13.1  
oc5         -        generic   Running   tcp://129.144.12.235:2376           v1.13.1   Note that remote operations are secure managed using ssh encryption and TLS certificates
Latest step is to prepare Swarm cluster:
Define OC5 as Swarm master:$ docker-machine ssh oc5
ubuntu@oc5:~$ docker swarm init --advertise-addr 10.0.0.6
Swarm initialized: current node (j54j1bricmhl3buten3qletxy) is now a manager.
To add a worker to this swarm, run the following command:
    docker swarm join \
    --token SWMTKN-1-1897yn8y1d931tnzfq99vek5fjebznsdp8po03zy15gk7kh9vc-9j50mbq7nkzrwcuvr6zn50svy \
    10.0.0.6:2377
To add a manager to this swarm, run 'docker swarm join-token manager' and follow the instructions.Define OC4 as Swarm second master ubuntu@oc5:~$ docker swarm join-token manager
To add a manager to this swarm, run the following command:
    docker swarm join \
    --token SWMTKN-1-1897yn8y1d931tnzfq99vek5fjebznsdp8po03zy15gk7kh9vc-0hk823fjc3mdqz5cvl88etxm8 \
    10.0.0.6:2377
ubuntu@oc5:~$ exit
$ eval $(docker-machine oc4)
[oc4]]$ docker swarm join \
>     --token SWMTKN-1-1897yn8y1d931tnzfq99vek5fjebznsdp8po03zy15gk7kh9vc-0hk823fjc3mdqz5cvl88etxm8 \
>     10.0.0.6:2377
This node joined a swarm as a manager.Finally add latest three nodes as workers

Elasticsearch 5.x Cookbook - Third Edition is just published, here the link at Amazon.
Again Packt Publishing trust in me as technical reviewer, is great for me be part of technical reviewer staff because it push me to read careful all the content, test the example and stay in touch with new technologies.
For a long time I been in contact with Apache Solr and ElasticSearch as a consequence of Scotas Products but its not the only contact with ElasticSearch.
Now ElasticSearch is part of the functionality for free text searching at Oracle NoSQL solution and is one of the solution stacks available at Oracle Cloud.
If you are starting a BigData project with or without Oracle NoSQL may be you will be in touch with ElasticSearch, in that case this cookbook will be a great starting point to start using ElasticSearch guided through a complete list of recipes covering all the functionality from installing on Windows/Linux or latest Docker distributions to BigData integrations.
Guides not only are designed for beginners users there are a lot of examples for power users/installations.
Happy reading and searching...

Some weeks ago Oracle releases Oracle Container Cloud Service, Oracle's support for Docker containers.
It basically provides several Linux machines to run Docker containers, one of them is designated as controller of the cluster services, here a simple screen shot of the service view:


by clicking on link Container Console you will access to the manager instance

the admin password for this console was defined during the service creation, here the look & feel the container console:
Basically the steps for using your containers are:

• Define a service
• Deploy a service
• Monitor your running containers

Define a service is basically a web representation of docker run command, supported version is Docker 1.10, several services are defined as examples, it includes Apache/PHP web server, Jenkins, Maria DB among others.
I tested this functionality adding a service for running Oracle 12cR1, the image for running an Oracle 12c is not in at a public repository, so you have to define a private registry for Docker images, remember that you must not push Oracle binary in public repository because you violate the license term.
So if  you follow this guide you can build your own registry server, but this is not enough because the registry server should be enabled using https for security reason, so follow this guide you could put NGinx reverse proxy and SSL certificate signed by LetsEncrypt, but to get a free SSL certificate this registry server should be register in a public DNS server.
If you get a registry server up, running and accesible through Internet over https this server could be added at the section Registries->New Registry, mandatory entries are:
Email: user@domain
URL: server.domain.com
UserName: user_name
Password: your_password
Description: A description textPort 443 of the SSL traffic is not required, the URL will be translated for example to https://server.domain.com:443/v2/, registry server will ask for HTTP authenticated user and the Cloud Service will provide the UserName and Password values.
Once you have a registry server up, running and registered at the Cloud Service you can define your Docker test, the Service builder page look like:
the Builder pane is a graphical representation of the docker run command, the image name (in black) must includes 443 port, for example server.domain.com:443/oracle/database:12.1.0.2-ee.
The process to build above image in your local machine is:
$ cd dockerfiles
$ ./buildDockerImage.sh -v 12.1.0.2 -e
$ docker login -u user_name -p user_pwd server.domain.com:443
$ docker tag oracle/database:12.1.0.2-ee server.domain.com:443/oracle/database:12.1.0.2-ee
$ docker push server.domain.com:443/oracle/database:12.1.0.2-eethen when you Deploy your service the Cloud Service will pull above image from your private registry.
The idea of building an Oracle 12cR1 EE image and test it using the Oracle Container Cloud Service is for comparing the performance against the DBAAS and IAAS testing. The result is:
OCCS
Max IOPS = 1387654
Max MBPS = 21313
Latency  = 0not so bad, under the hood this Oracle RDBMS is running on Oracle Linux Server release 6.6.89/Docker 1.10,  4 x Intel(R) Xeon(R) CPU - 16Gb RAM. the file system seem to be XFS.

Drawbacks:- By not proving a private registry where you can build/pull/push your custom images the usage of the service could be complicated for most of non experimented Docker users.
- I can't find a way to define a shared filesystem for a given volume, for example, above deployment puts Oracle Datafiles into an internal container volume, if you stop your deployment all the data is lost, the only possibility is pause/unpause the service if you want not to loose your data. Note: at the Service Editor (Tips & Trick button) there is an example defining an external volume as /NFS/{{.EnvironmentID}}/{{.DeploymentID}}/{{.ServiceID}}/{{.ServiceSlot}}:/mnt/data, but it didn't work for me.
- You can't modify Hosts OS parameters, so for example if you want to deploy an ElasticSearch cluster is necessary to change at /etc/sysctl.conf file vm.max_map_count=262144, so is limited environment also for a simple test case.
- Docker version is 1.10.3, which means, if you want to deploy Oracle XE it doesn't work because --shm-size=1g is not supported
- Some time the Container Cloud Console kill my Chrome browser, Linux or Windows version, here the screenshot, seem to be a JavaScript problem:
Final thoughtsThe Containers Cloud Service console is a good abstraction (graphical interface) of typical Docker command line services, for example:
Services -> docker run command
Stacks -> docker-compose command, docker-compose.yml (graphical interface)
Hosts -> Bare metal/VM servers
In my personal opinion if I have to deploy a docker complex installation I'll deploy a set Oracle Compute Cloud Service running Oracle Linux/Ubuntu installations with latest Docker release and docker swarm native service, why?

• It run faster, see my previous post.
• I have control of the host parameter (sysctl among others, see host setup section)
• I can define shared filesystem in ext4 or NFS partitions
• I can build my own images by using Dockerfile commands
• I can deploy/manage the infrastructure (nodes/network) using docker swarm command.

Two day after publish my post on Building a Docker image with Oracle 11gXE and latest Apex distribution Oracle releases latest production version 5.1.
So I just committed on GitHub a project update with the instruction to build a new Docker image using Oracle 11g official Docker image script and Apex 5.1 installation.
Here an screen shot of the first run when open the URL with http://localhost:8080/apex/apex_admin:

admin password is by default sys/system password output during first boot of the Docker image, see Docker logs for more details, for example:
ORACLE AUTO GENERATED PASSWORD FOR SYS AND SYSTEM: 80c55053b63dc749

This year during the Oracle Development Tour at Argentina I saw that the labs have the problem that the default installation of Oracle 11g XE doesn't include an updated version of Oracle Apex.
So I decide to build a new Oracle 11g XE Docker image including latest Apex version available for download at OTN with two mainly purposes:

• A clean installation for demos and courses
• A simple distribution of latest Oracle Apex based Docker

Everybody knows that Docker simplify the development/testing/production lines by providing a clean/simple image distribution to run basically any Linux/Windows program on any platform,  so the steps below could be replicated on any Linux flavors (Ubuntu/RedHat/Debian), Windows environment or Mac, with the only requirement of having Docker 1.10+ installed and running.
So here the simple steps:
- Check if docker is running:
[mochoa@localhost github-public]$ docker version Client: Version:      1.12.4 API version:  1.24 Go version:   go1.6.4 Git commit:   1564f02 Built:        Mon Dec 12 23:50:16 2016 OS/Arch:      linux/amd64 
Server: Version:      1.12.4 API version:  1.24 Go version:   go1.6.4 Git commit:   1564f02 Built:        Mon Dec 12 23:50:16 2016 OS/Arch:      linux/amd64[mochoa@localhost dockerfiles]$ ./buildDockerImage.sh -v -x.. lot of outputs here ..[mochoa@localhost dockerfiles]$ docker images|grep oracle oracle/database                                        11.2.0.2-xe              ba74688a297e        39 hours ago        1.206 GBoracle/database                                        12.1.0.2-ee              af209128066e        6 days ago          11.72 GB[mochoa@localhost xe-apex-latest]$ ./buildDockerImage.sh .. lot of outputs here ..[mochoa@localhost xe-apex-latest]$ docker images|grep apex oracle/apex-5.0.4_en                                   11.2.0.2-xe              89ef9b5e1ced        18 hours ago        1.531 GB

As usual my first test with every Oracle RDBMS release is by Java inside the database, for this test I am using a heavy/intensive Java application inside the RDBMS named Scotas OLS.
Basically it provides full text search functionality by using an embedded Solr version running inside the DB.
By now Oracle 12cR2 is available at the Oracle Cloud DBAAS, the screen below is the services started to do the test
As I mentioned above Scotas OLS provides searching functionality through new domain index, during a creation or rebuild of the index and full table scan is done in batch of 5000 rows, this full scan is executed using a query like:
SELECT /*+ DYNAMIC_SAMPLING(L$MT 0) */ L$MT.rowid,..others_columns..
     FROM TABLE_NAME L$MT where rowid in (select /*+ cardinality(L$PT, 10) */ * from table(?) L$PT) for update nowaittable pipeline function receives as arguments a batch of 5000 rowids wrapped using ODCIRidList PLSQL Type.
The Java code running inside the DB look like:
                while (rs.next()) {
                    addCmd.clear();
                    addCmd.solrDoc = new SolrInputDocument();
                    String rowid = rs.getString(1);
                    addCmd.solrDoc.addField("rowid", rowid);
                    if (includeMasterColumn) {
                        Object o = rs.getObject(2);
                        readColumn(addCmd.solrDoc,this.colName,o);
                    }
                    for (int i = 0; i < numExtraCols; i++) {
                        Object o = rs.getObject(i + offset);
                        readColumn(addCmd.solrDoc,extraCols[i],o);
                    }
                    if (log.isTraceEnabled())
                      log.trace("Adding: " + addCmd.solrDoc.toString());
                    try {
                        processor.processAdd(addCmd);
                    } catch (Exception e) {
                        ....
                    }
                }
the scan time observed between the two DBAAS instances is summarized in this graph
Average scan time is 890 ms per batch in 12cR2 versus 1648 ms in 12cR1, GOOD JOB Oracle OJVM dev team for this improvement!!!!

The idea of this post is to show my simple test (first touch) of Oracle Cloud services:

• Oracle Compute Cloud Service
• Oracle Database Cloud Service

the idea is to create an Oracle instance and run a heavy IO testing using Oracle Resource manager.
Both DBs are running in similar hardware shapes as show in these screen shots:

If you connect using ssh the underlying hardware are:

• 4 x Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz - 16Gb RAM for IAAService
• 4 x Intel(R) Xeon(R) CPU E5-2699 v3 @ 2.30GHz - 16Gb RAM for DBAAService

Database release in both cases is:
Oracle Database 12c Enterprise Edition Release 12.1.0.2.0 - 64bit Production

Latest release of Docker project integrate a Swarm mode to manage your cluster environment.
The idea of this post is to show how to use Docker Swarm to deploy an ElasticSearch 5.0 cluster, to be similar to a production DataCenter I am using docker-machine to deploy six nodes of my Swarm cluster, similar deployment could be replicated in a public cloud such as AWS or DigitalOcean by simple change a parameter in docker-machine create command.
My six nodes where created using:
docker-machine create --driver virtualbox manager1
docker-machine create --driver virtualbox manager2
...mochoa@localhost:~/es$ docker service scale es_ingest=2
es_ingest scaled to 2
mochoa@localhost:~/es$ docker logs -f es_master.1.8fuhvyy9scidk9fzmmme9k14l
[2016-09-13 18:42:37,811][INFO ][cluster.service          ] [c251c9e143c1] added {{223fe8d80047}{f7GkmloETKe4UZsJy0nOsg}{fO6738D4T-ORB6GFFSSbsw}{10.0.1.9}{10.0.1.9:9300},}, reason: zen-disco-node-join[{223fe8d80047}{f7GkmloETKe4UZsJy0nOsg}{fO6738D4T-ORB6GFFSSbsw}{10.0.1.9}{10.0.1.9:9300}]mochoa@localhost:~/es$ docker service rm es_ingest es_data es_master
es_ingest
es_data
es_master

Oracle provides a way to build several Docker images at their GitHub Official Repo. But IMO they have two big problems:

• unlike the Docker way, you have to download manually binary installation files
• size of the images is big due the use of Docker COPY command (ADD or COPY section)

To mitigate these two factors I tested the local repo idea, a new Docker image which exposes using HTTP protocol all Oracle binary files in your DataCenter without broke the license term.
You could build a local repository image adding or removing Oracle binary packages, for example:
FROM gliderlabs/alpine:3.4
# Maintainer
# ----------
MAINTAINER Marcelo Ochoa
RUN apk --no-cache add nginx && mkdir /run/nginx
# Oracle Linux 64Bits 12.1.0.2
COPY linuxamd64_12102_database_1of2.zip /var/lib/nginx/html
COPY linuxamd64_12102_database_2of2.zip /var/lib/nginx/html
# Oracle Linux XE 11.2.0
COPY oracle-xe-11.2.0-1.0.x86_64.rpm.zip /var/lib/nginx/html
# Oracle JDK 8b101
COPY server-jre-8u101-linux-x64.tar.gz /var/lib/nginx/html
# Oracle WebLogic 12.2.1.1.0
COPY fmw_12.2.1.1.0_wls_Disk1_1of1.zip /var/lib/nginx/html
EXPOSE 80
CMD ["nginx", "-g", "daemon off;"]files marked in italic/bold should be downloaded first (manually from OTN) and located in same directory as your local repo Dockerfile, for example:
mochoa@localhost:~/jdeveloper/mywork/docker/localrepo$ ls -l
total 3805068
-rwxr-xr-x 1 mochoa mochoa         48 ago 12 10:23 buildDockerImage.sh
-rw-rw-r-- 1 mochoa mochoa        493 ago 15 18:19 Dockerfile
-rw-rw-r-- 1 mochoa mochoa  832500826 ago 15 18:01 fmw_12.2.1.1.0_wls_Disk1_1of1.zip
-rw-r--r-- 1 mochoa mochoa 1673544724 jul 16 19:19 linuxamd64_12102_database_1of2.zip
-rw-r--r-- 1 mochoa mochoa 1014530602 jul 16 19:20 linuxamd64_12102_database_2of2.zip
-rw-rw-r-- 1 mochoa mochoa  315891481 jul 18 12:20 oracle-xe-11.2.0-1.0.x86_64.rpm.zip
-rwxr-xr-x 1 mochoa mochoa         97 ago 12 10:25 run-repo.sh
-rw-rw-r-- 1 mochoa mochoa   59874321 ago 15 17:32 server-jre-8u101-linux-x64.tar.gzto build and run your local repo use buildDockerImage.sh and run-repo.sh:
# docker build -t "localrepo:1.0.0" .# docker run --name localrepo --hostname localrepo --detach=true --publish=8000:80 localrepo:1.0.0
echo "Local repo is at: http://$(docker inspect --format='{{ .NetworkSettings.IPAddress }}' localrepo):80/"finally if We use a modified version of Oracle Official Dockerfile:
mochoa@localhost:~/jdeveloper/mywork/docker/docker-images/OracleDatabase/dockerfiles$ git diff 11.2.0.2/Dockerfile.xe
diff --git a/OracleDatabase/dockerfiles/11.2.0.2/Dockerfile.xe b/OracleDatabase/dockerfiles/11.2.0.2/Dockerfile.xe
index 00ed28a..cade77c 100644
--- a/OracleDatabase/dockerfiles/11.2.0.2/Dockerfile.xe
+++ b/OracleDatabase/dockerfiles/11.2.0.2/Dockerfile.xe
@@ -50,7 +50,7 @@ ARG ORACLE_PWD

 # Copy binaries
 # -------------
-COPY $INSTALL_FILE_1 $CONFIG_RSP $INSTALL_DIR/
+COPY $CONFIG_RSP $INSTALL_DIR/

 # Update yum
 # ----------
@@ -61,8 +61,9 @@ WORKDIR $INSTALL_DIR

 # Install Oracle Express Edition
 # ------------------------------
-RUN unzip $INSTALL_FILE_1 && \
-    rm $INSTALL_FILE_1 &&    \
+RUN curl -sSL http://192.168.200.55:8000/$INSTALL_FILE_1 -o $INSTALL_FILE_1 && \
+    unzip $INSTALL_FILE_1 && \
+    rm -f $INSTALL_FILE_1 && \
     rpm -i Disk1/*.rpm &&    \
     sed -i -e "s|###ORACLE_PWD###|$ORACLE_PWD|g" $INSTALL_DIR/$CONFIG_RSP && \
     /etc/init.d/oracle-xe configure responseFile=$CONFIG_RSP
@@ -91,5 +92,6 @@ RUN echo "DEDICATED_THROUGH_BROKER_LISTENER=ON"  >> $ORACLE_HOME/network/admin/l
     echo "DIAG_ADR_ENABLED = off"  >> $ORACLE_HOME/network/admin/listener.ora;

 EXPOSE 1521 8080
+VOLUME ["/u01/app/oracle/oradata","/u01/app/oracle/fast_recovery_area"]

 CMD $ORACLE_BASE/runOracle.sh
PD: I don't know why official Oracle RDBMS images includes the Oracle Data-files (data), containers are immutable, Do not store data in containers!!! I prefer a build Docker image which creates outside the Database data-files.

Sometime Oracle XE is used in development environment because at the early stage of the project an Oracle RDBMS EE is not necessary.
There are many Docker images at the public repo but many of them do not include the possibility to customize some parameters and I am not sure if the legal condition of Oracle XE redistribution license is meet.
So, using similar ideas as in Oracle RDBMS 12c Docker image script I made this simple Dockerfile script:
FROM oraclelinux:5
MAINTAINER marcelo.ochoa@gmail.com
RUN groupadd -g 54321 oinstall
RUN groupadd -g 54322 dba
RUN useradd -m -g oinstall -G oinstall,dba -u 54321 oracle
RUN yum -y install bc glibc make binutils gcc libaio perl wget unzip && yum clean all
RUN chown -R oracle:oinstall /home/oracle
RUN chmod g+rx /home/oracle
RUN chmod o+rx /home/oracle
ADD xe.rsp /home/oracle/
ADD oracle-xe-11.2.0-1.0.x86_64.rpm /home/oracle/
RUN rpm2cpio /home/oracle/oracle-xe-11.2.0-1.0.x86_64.rpm | cpio -idmv
RUN rm -f /home/oracle/oracle-xe-11.2.0-1.0.x86_64.rpm
RUN mkdir /u01/app/oracle/oradata
RUN mkdir /u01/app/oracle/product/11.2.0/xe/config/log
RUN mkdir -p /u01/app/oracle/diag/rdbms/xe/XE/trace/
RUN umask 0027
RUN mkdir -p /u01/app/oracle/admin/XE/adump
RUN mkdir -p /u01/app/oracle/admin/XE/dpdump
RUN mkdir -p /u01/app/oracle/admin/XE/pfile
RUN mkdir -p /u01/app/oracle/admin/cfgtoollogs/dbca/XE
RUN mkdir -p /u01/app/oracle/admin/XE/dbs
RUN mkdir -p /u01/app/oracle/fast_recovery_area
RUN umask 0022
RUN touch /u01/app/oracle/diag/rdbms/xe/XE/trace/alert_XE.log
RUN chown -R oracle:oinstall /u01/app/oracle
RUN chmod u+x /etc/init.d/oracle-xe
RUN sed -i -e 's/%memory_target%/1G/g' /u01/app/oracle/product/11.2.0/xe/config/scripts/init.ora
RUN sed -i -e 's/%memory_target%/1G/g' /u01/app/oracle/product/11.2.0/xe/config/scripts/initXETemp.ora
RUN find /u01/app/oracle/product/11.2.0/xe -name "*.sh" -exec chmod u+x {} \;
RUN echo ". /u01/app/oracle/product/11.2.0/xe/bin/oracle_env.sh" >>/home/oracle/.bashrc
COPY manage-oracle.sh /home/oracle/
EXPOSE 1521 8080
VOLUME ["/u01/app/oracle/oradata","/u01/app/oracle/fast_recovery_area"]
CMD [ "sh" , "-c" ,  "/home/oracle/manage-oracle.sh" ]Note that also that in above Dockerfile script instead of using rpm tool to install Oracle XE rpm file I am using rpm2cpio, this is because the original rpm file provided by Oracle do some actions that are not valid on Docker environment, I am doing all necessary actions inside the Dockerfile then.docker run --privileged=true --name xe --hostname xe --detach=true --publish=1521:1521 --publish=8080:8080 oracle-xe:11.2.0-1.0docker run --privileged=true --volume=/home/data/db/xe/fast_recovery_area:/u01/app/oracle/fast_recovery_area --volume=/home/data/db/xe/oradata:/u01/app/oracle/oradata --name xe --hostname xe --detach=true --publish=1521:1521 --publish=8080:8080 oracle-xe:11.2.0-1.0

At the time of writing this post Building an Oracle NoSQL cluster using Docker there was no official Oracle NoSQL images to use as base image, also Docker has no native networking functionality.
So with these two new additions making a NoSQL cluster is too easy :)
First step, creating my NoSQL image with an automatic startup script, here a Docker file:
FROM oracle/nosql
MAINTAINER marcelo.ochoa@gmail.com
ADD start-nosql.sh /start-nosql.sh
RUN chmod +x /start-nosql.sh
CMD /start-nosql.shdocker build -t "oracle-nosql/net" .

Following Neto's blog post about graphOra (Docker Image) – Oracle Real Time Performance Statistics I did my personal test using a Docker image for 12c.
First, I started a 12c Docker DB using:
# docker run --privileged=true --volume=/var/lib/docker/db/ols:/u01/app/oracle/data --name ols --hostname ols --detach=true --publish=1521:1521 --publish=9099:9099 oracle-12102Next starting Graphite Docker image:
# docker run --name graphs-db -p 80 -p 2003 -p 2004 -p 7002 --rm -ti nickstenning/graphiteNext installing graphOra repository:
# docker run -ti --link ols:oracle-db netofrombrazil/graphora --host oracle-db --port 1521 --sid ols --create
Enter sys password: -------
Creating user graphora
Grant access for user graphora to create sessions
Grant select privilege on V$SESSION_EVENT, V$SYSSTAT, V$STATNAME for user graphora
---
GraphOra is ready to collect your performance data!# docker run -ti --link ols:oracle-db --rm --link graphs-db netofrombrazil/graphora --host oracle-db --port 1521 --sid ols --interval 10 --graphite graphs-db --graph-port 2003
phyReads: 0 phyWrites: 0 dbfsr: 43.30 lfpw: 43.30
phyReads: 0 phyWrites: 0 dbfsr: 0.00 lfpw: 0.00
phyReads: 0 phyWrites: 0 dbfsr: 0.00 lfpw: 0.00Note on parameters used