Some time ago I published a post regarding the Database Migration Service (DMS) on AWS. The service had just appeared at that time and looked very promising. It is still appealing and has a good use and potential to mature and be better. I was able to test and try it in several cases moving data to AWS and between RDS databases inside AWS. During those exercises, I ran into some issues and wanted to make other people aware of some things to keep in mind when starting or planning to use the DMS. Most of my experience with DMS is related to migrations to and from Oracle databases. So the following tips are about Oracle migrations and replications.
Before planning any kind of a logical replication based on transaction logs please check what kind of data types you have and whether you have primary keys on all your replicated tables. The primary key existence in some cases is not only desirable but is required to properly replicate the objects. As an example, if you want to replicate a table with some lob objects using DMS you must have a primary key for the table because if you don’t your lob columns will be excluded from replication and you end up with null values instead of the lobs on the target database.
Check for datatypes used for the tables. All logical replications have some limitations in terms of replicated data types. You may find some of your tables can be replicated with some limitations or not replicated at all.
Let’s speak about DMS. When you plan your DMS replication tasks keep in mind that you can combine replication of several schemas to one task. It may significantly reduce load on the source system. Each migration or replication task may apply a considerable load to the source system. In my experience, we hit 100% CPU and max IO load running 8 replication tasks on the source m3.large system.
Remember you cannot change the created task. You are able to stop, start or resume the task but you cannot alter any rules or parameters for the task. It will maybe change soon but currently, it is not possible yet.
If you choose the truncate option for your migration and replication task it may change your metadata. You may find your indexes and constraints to be renamed and you even can lose some of the indexes. In my case, I got renamed primary key and unique key constraints and couple of indexes were lost. Please be careful. After that case, I tried to choose “do nothing” mode and do everything by myself preparing the target for data migration and replication.
You can use RDS snapshot to instantiate your data if you plan to setup a replication between two RDS instances of the same type. In our case, it was done to minimize downtime when the database was migrated to an encrypted storage. When you use RDS snapshot you can use snapshot creation time as “Custom CDC start time” for your DMS replication task.
If you use one universal user for your endpoints to replicate multiple schemas you will need to use transformation rules because the replication will try to use schema from endpoint as destination target by default. By other words, you set up a target endpoint using user “DMS” and try to replicate schema SCOTT it will use schema “DMS” as a destination by default if you don’t have the transformation rules.
You have to enable minimal supplemental logging on database level and supplemental logging for all columns explicitly for each replicated table. Even you enable supplemental logging for all columns on database level using “alter database” you still need to add it on table level. The DMS task will be aborted without it complaining about the lack of supplemental logging for all column for the table.
If you create a table in one of your replicated schema you need to add a clause for supplemental logging like “CREATE TABLE …ADD SUPPLEMENTAL LOG DATA (ALL) COLUMNS; ” . In that case, the table and data will be replicated.
It is better to use AWS cli to get most of the service since all new features and additions will be in cli first. As an example, if you want a debug logging for your replication you have to use AWS cli.
If you have only unique index on a table and it is function based index your data migration and replication task can fail. Also, it may fail if you have more than one unique index on a table.
I hit some issues with monitoring. The idea looks good but it requires some fixing. It looks like it doesn’t work correctly in Firefox and Safari. At least for me, it was not working right.
The status of a task may not tell you everything. Sometimes it shows state “ERROR” but nevertheless, it works and replicates data behind the scenes. So, it can be bit misleading. I look to statistics and monitoring pages for the task to get the full picture.
As a summary, I can say the service deserves attention and can be considered as a valuable option when you plan your migration strategy and AWS DMS team works hard to make it better. Happy migrations!
Oracle Traffic Director (OTD) is software load balancer from Oracle that receive requests from client and forward it to back-end application like WebLogic , WebCenter, Portal etc. In this first post on OTD, I’ll be covering basic concepts related to OTD (More on installation & configuration of OTD later in future posts) 1. Oracle […]
The post Oracle Traffic Director (OTD) 12c : Basics & Overview appeared first on Oracle Trainings for Apps & Fusion DBA.
Web-based content in Oracle E-Business Suite 12 runs on the Oracle Application Framework (OAF or "OA Framework") user interface libraries and infrastructure. Since the release of Oracle E-Business Suite 12.2 in 2013, we have released several cumulative updates to Oracle Application Framework to fix performance, security, and stability issues.
These updates are provided in cumulative Release Update Packs, and
cumulative Bundle Patches that can be applied on top of the Release
"Cumulative" means that the latest RUP or Bundle Patch contains
everything released earlier.
The latest OAF update for EBS 12.2.5 is now available:
- Oracle Application Framework (FWK) Release 12.2.5+ Bundle 7 (Patch 24561446:R12.FWK.C)
Where is the documentation for this update?
Instructions for installing this OAF Release Update Pack are here:
- Oracle E-Business Suite Release 12.2 Upgrade Considerations for OA Framework-based Applications (Note 1927975.1)
Who should apply this patch?
All EBS 12.2.5 users should apply this patch. Future OAF patches for EBS 12.2.5 will require this patch as a prerequisite.
What's new in this update?
This bundle patch is cumulative: it includes all fixes released in previous EBS 12.2.5 bundle patches.
This latest bundle patch includes new fixes for the following critical issues:
- The number values in Total column in the Budget table are partially hidden after applying some set of ATG and Mobile Application patches.
- Failed to return to the base page after submitting the popup due to client side validation.
- The rawtext with style class set to OraInstrcutionText under HideShowHeader with style class OraHeaderSub does not word-wrap.
- Tables are not get rendered properly when scrolling when there are more than 300 records in the table.
- Page/form image icons are not aligned on home page with the function name links in Navigator and Favorites regions in Chrome and Firefox browsers.
- The document catalog column was displayed on inline attachment popup windows even if document catalog option was disabled.
- In HTTPS-enabled environments, navigating to 3-Frame JSP page shows a blank page since the request to load mixed active content has been blocked.
In one of our project, we were using CS 7.2 P02 (upgraded later with P05) in correlation with xPlore 1.5 P02. With these versions, we wanted to setup the IndexAgents in HTTPS to have a completely secured environment. We choose to follow the xPlore documentation and use the Groovy script that EMC is now providing to facilitate the setup in HTTPS of the IndexAgents & Primary DSearch.
The first thing to note is that the documentation is describing some parts of how to setup the SSL in xPlore but it is not the complete setup and some other parts are missing! You can use one of my previous blog to know how to setup an IndexAgent in HTTPS using the Groovy Script: click here. The second thing to note is that after setting our IndexAgents in HTTPS, we faced an issue where the status of the IndexAgents weren’t available anymore (it was working in HTTP)! After following the steps described in the other blogs, please take a look at the explications below in case you are having the same issue.
So after doing what is described in the link above, the status of the IndexAgents in Documentum Administrator was “Not responding” and we weren’t able to start/stop them. Same behavior via iapi, the status command returned a 200 code. Unlike the HTTP response codes, the IndexAgents codes are:
- 0 => Everything is running & working
- 100 => The JBoss instance hosting the IndexAgent is running BUT the IndexAgent isn’t started (E.g.: not in the status “Start in Normal Mode”)
- 200 => The JBoss instance hosting the IndexAgent isn’t running OR not responding
Now to check the status of the IndexAgent, you can do the following using iapi:
API> ?,c,select object_name, index_name from dm_ftindex_agent_config; object_name index_name ----------------------------------------------- ------------------------------------------------ xplore_server_01_9200_IndexAgent DOCBASE_ftindex_01 (1 row affected) API> apply,c,,FTINDEX_AGENT_ADMIN,NAME,S,DOCBASE_ftindex_01,AGENT_INSTANCE_NAME,S,xplore_server_01_9200_IndexAgent,ACTION,S,status ... q0 API> next,c,q0 ... OK API> dump,c,q0 ... USER ATTRIBUTES name : xplore_server_01_9200_IndexAgent status : 200 SYSTEM ATTRIBUTES APPLICATION ATTRIBUTES INTERNAL ATTRIBUTES API> exit Bye
You can see above a status 200… I can assure you that the JBoss instance and IndexAgent are up, running & working properly (we had to start it using the GUI since it was the only way to check the status)! Then why is a status code 200 returned? Well that’s the purpose of this blog!
Like I said before, we choose to setup the SSL for the IndexAgents using the Groovy script and that’s the reason why we faced this issue in the first place… The Groovy script is setup to automatically update the file “standalone.xml” to add some configuration for the SSL. This is what the script will add:
[xplore@xplore_server_01 ~]$ grep -A4 "connector name=\"https\"" /app/xPlore/jboss7.1.1/server/DctmServer_Indexagent_DOCBASE/configuration/standalone.xml <connector name="https" protocol="HTTP/1.1" scheme="https" socket-binding="https" secure="true"> <ssl name="https" password="***********" certificate-key-file="/app/xPlore/dsearch/admin/./../../jboss7.1.1/server/DctmServer_Indexagent_DOCBASE/configuration/my.keystore" cipher-suite="TLS_RSA_WITH_AES_128_CBC_SHA"/> </connector>
Now what is wrong with this configuration? It actually looks quite good… We were working on this specific issue for a few months with EMC without any special findings (no errors in logs, nothing in the DFC/RPC/HTTP_POST traces, aso…) when we noted that the Groovy script only added a “normal” Cipher to the standalone.xml file but our Content Servers were setup at that time using the default SSL mode of Documentum which is commonly named “Anonymous SSL”. As shown above, the SSL Cipher that the Groovy script added is “TLS_RSA_WITH_AES_128_CBC_SHA”. The problem is that the JBoss instance hosting the IndexAgent is restricting the Ciphers available but the docbase on the other side isn’t able to use this Cipher because it is using the “Anonymous SSL”. Therefore there is mismatch in the Ciphers available and the communication can’t be completed.
So what is the solution to that? Well we simply need to extend the Cipher list to find a Cipher that is available for both the docbase and the IndexAgent. To solve this issue, you just have to replace the “cipher-suite” as follow:
[xplore@xplore_server_01 ~]$ grep -A4 "connector name=\"https\"" /app/xPlore/jboss7.1.1/server/DctmServer_Indexagent_DOCBASE/configuration/standalone.xml <connector name="https" protocol="HTTP/1.1" scheme="https" socket-binding="https" secure="true"> <ssl name="https" password="***********" certificate-key-file="/app/xPlore/dsearch/admin/./../../jboss7.1.1/server/DctmServer_Indexagent_DOCBASE/configuration/my.keystore" cipher-suite="TLS_DH_anon_WITH_AES_128_CBC_SHA,TLS_RSA_WITH_AES_128_CBC_SHA"/> </connector>
With this setup, you just have to restart the JBoss instance hosting the IndexAgent, reinitialize the Content Server and after doing that, the status will be available again:
API> apply,c,,FTINDEX_AGENT_ADMIN,NAME,S,DOCBASE_ftindex_01,AGENT_INSTANCE_NAME,S,xplore_server_01_9200_IndexAgent,ACTION,S,status ... q0 API> next,c,q0 ... OK API> dump,c,q0 ... USER ATTRIBUTES name : xplore_server_01_9200_IndexAgent status : 100 SYSTEM ATTRIBUTES APPLICATION ATTRIBUTES INTERNAL ATTRIBUTES API> apply,c,,FTINDEX_AGENT_ADMIN,NAME,S,DOCBASE_ftindex_01,AGENT_INSTANCE_NAME,S,xplore_server_01_9200_IndexAgent,ACTION,S,start ... q1 API> next,c,q1 ... OK API> dump,c,q1 ... USER ATTRIBUTES name : xplore_server_01_9200_IndexAgent status : 0 SYSTEM ATTRIBUTES APPLICATION ATTRIBUTES INTERNAL ATTRIBUTES API> exit Bye
As you might know, the Cipher that we added contains “_anon_”. This means Anonymous (no authentication) and this also means that it is not the most secure of all Ciphers but this is necessary if you are using the default SSL mode in Documentum (without specifically setting up your own SSL Certificates). That’s why EMC recommends you to setup a non-Anonymous SSL mode but that’s a little bit more difficult to setup than just using the default one. Might be the subject of another blog
Cet article Documentum story – Status of IndexAgent not visible in HTTPS (xPlore) est apparu en premier sur Blog dbi services.
- ORA-01036 Error in Node.js app with Oracle Database Node.js Add-on (OPN Innovation and Modernisation Center (EMEA))
via OPN Innovation and Modernisation Center (EMEA) http://ift.tt/1AAiVSD
- Sponsored: 64% off Code Black Drone with HD Camera
Our #1 Best-Selling Drone--Meet the Dark Night of the Sky!
Redwood Shores, Calif.—Oct 26, 2016
Oracle today introduced a series of new data management platform innovations within the Oracle Marketing Cloud that enable customers to seamlessly define, select, count, and deliver audience IDs based on the users’ environment. This allows customers to create audience segments tied to the customer journey across devices, channels, media environments, and ID spaces. Powered by the Oracle ID Graph, this release helps marketing and media professionals deliver on their marketing strategies by delivering relevant experiences to customers on whichever device they choose.
As consumers increasingly interact with brands across a range of different devices and environments, marketers and publishers face the challenge of unifying consumer IDs from different platforms and devices in order to gain a holistic view of individual buyers. To empower marketers and publishers to quickly and efficiently manage these multiple environments, Oracle has released major enhancements to the Audience Builder user interface within the Oracle Data Management Platform to help marketers visualize and manage ID source, with planned functionality to link these audiences via a self-service UI through the Oracle ID Graph. The new user interface released today makes it easier for customers to manage and select ID types in order to identify, build, plan and analyze audiences based on the device or platform within which the ID was originally created. This lays the foundation for the next iteration of the platform, which will further enable cross-device marketing by bridging IDs and scoring linkages between those IDs. This set of enhancements accelerates the convergence of adtech and martech within and outside of the Oracle technology stack.
“The explosive growth in connected devices has created immense opportunities for marketers to engage with consumers at the moments that truly matter,” said Steve Krause, group vice president, product management, Oracle Marketing Cloud. “To capitalize on this opportunity, marketers need to be able to manage audiences based on the engaged environment. This is critical to intelligent linking for use cases like cross-device targeting and that is why we have revamped the Audience Builder user interface within the Oracle Data Management Platform to help address device agnostic identity management.”
Audience Builder enables marketers to plan audiences for their campaigns using a combination of their own data and third-party data sources. Through the new Audience Builder users are able to:
- Enhance Audience Segmentation and Engagement: Understand audience composition by device and environment to increase the effectiveness of engagements.
- Optimize Media Buys: By identifying the devices driving conversions, marketers can modify campaign strategies to focus the conversion points and eliminate wasted impressions.
- Increase Campaign Effectiveness: By managing advertising dollars based on targeting environment, customers can personalize campaigns against the source ID for relevant retargeting and prospecting.
- Improve Mobile App Targeting at Scale: Enables marketers to activate 1st and 3rd-party audiences on channels against their mobile ad IDs based on behaviors like mobile app installs.
The new Audience Builder user interface is the latest in a series of innovative enhancements to the Oracle Data Management Platform. In the last year, Oracle has:
- Integrated Native Tools for Data Ingest and Custom Audience Creation: Self-serve tools available in the platform for custom audience creation powered by Oracle Data Cloud’s Audience Discovery Platform, as well as embedded offline onboarding functionality via Oracle OnRamp, are now native functionalities of the platform.
- Enhanced Personalization: Maxymiser integration enables marketers to tap into 1st, 2nd and 3rd party data for testing and optimization initiatives. This makes it easier for marketers to deliver experiences that are relevant to customers even before the first page view.
- Enhanced Partner Integration: Enables partners to self-integrate into the Oracle Data Management Platform by setting up their endpoint for turnkey data campaign delivery. Since the new self-service portal was launched in November 2015, more than 20 new partners have been added.
The Oracle Marketing Cloud is part of the Oracle Customer Experience (CX) Cloud, one of the industry’s most complete CX solutions. Oracle CX Cloud empowers organizations to improve experiences, enhance loyalty, differentiate their brands, and drive measurable results by creating consistent, connected, and personalized brand experiences across all channels and devices.
PR for Oracle
Oracle offers a comprehensive and fully integrated stack of cloud applications and platform services. For more information about Oracle (NYSE:ORCL), visit www.oracle.com.Trademarks
Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.Safe Harbor
The preceding is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. The development, release, and timing of any features or functionality described for Oracle's products remains at the sole discretion of Oracle Corporation.
A Guest Post by Oracle's Jack Berkowitz, VP of AI Apps Development (pictured left)
I'm thrilled to report that industry watchers liked what they heard at OpenWorld about our next-generation cloud strategy. One in particular, Ventana Research, named Oracle a winner of its 2016 Technology Innovation Awards for our Adaptive Intelligent Applications. How cool is that! These decision science and data-driven apps are part of the hottest trend in technology right now: AI and machine learning.
But our AI notoriety doesn’t stop at one award. Our OpenWorld announcement generated a lot of analyst and press interest (see list of published articles below). We even received favorable coverage from some of the most skeptical pundits. Here are a few that stand out:
“I see Salesforce as the first mover, but Oracle has significant long-term advantages.” – Doug Henschen, Constellation Research
"Oracle's new Adaptive Intelligent solutions take this value proposition a step further. They are set apart from others by allowing the intelligent applications to learn from billions of anonymized consumer and business profiles available from Oracle." – Dave Schubmehl, IDC, quoted in eWeek
“It is good to see a major vendor start to talk about specific use cases and put a bit more business context around the current buzz around machine learning.” - Vinnie Mirchandani, Enterprise Irregulars
Real products. Real benefits.
But this isn’t just us talking a good game. (In fact, there’s no shortage of hype as evidenced by articles such as this: Salesforce's "Einstein" receives mixed reviews.) Our Dev team is on track to meet the product rollout plan we shared with analysts and press at OpenWorld.
The first product we plan to release in 2017 is Adaptive Intelligent Offers. It enhances Oracle Commerce and Marketing Clouds with the ability to deliver offers and product recommendations to consumers at a specific moment. There are a number of competitive solutions out there, but they don’t have the same decision science/machine-learning algorithms that we do and require manual tuning of rules.
In addition, our apps don’t require huge data science departments to use them. Here’s why that’s an advantage. Some customers, such as the largest retailers, will want to build proprietary machine learning platforms, but most won’t have access to the talent, data, and infrastructure required. Gartner predicts that many organizations will opt for packaged applications because the data science needed to create these systems is so complex.
No one can match Oracle's winning combination
So what sets our apps apart? It's the combo of these four elements:
- The most data with five billion anonymous consumer and business profiles
- The broadest domain expertise and delivery platform with 12,500 SaaS customers across CX, HCM, ERP, EPM, and SCM
- Data science, machine learning, and AI expertise in consumer research, systems optimization, IOT predictive analytics, business analytics and prediction
- Scalable cloud infrastructure that is automated
We're also working on AI Actions for sales and service, AI Discounts for procurement and finance, AI Dynamic Planning and Bidding for logistics, and AI Candidate Experience for HR.
 “Top 10 Strategic Technology Trends for 2017”, Gartner, October 2016.
Analyst and Press Articles Covering Adaptive Intelligent Apps
CMSWire: “Oracle Ups its Cloud Strategy with Adaptive Intelligent Applications #OOW16,” Derek Walter, Sept. 20, 2016
CMSWire: “Oracle Wraps Up OpenWorld With Promises of a Smarter Cloud,” Derek Walter, Sept. 26, 2016
Constellation Research: “Oracle Vs Salesforce on AI: What to Expect When,” Doug Henschen, Sept. 20, 2016
CRN: “Oracle Delivers Artificial Intelligence Solutions To Partners,” Joseph Tsidulko, Sept. 19, 2016
Destination CRM: “At OpenWorld 2016, Oracle Unveils Adaptive Intelligent Cloud Applications,” Oren Smilansky, Sept. 20, 2016
Enterprise Apps Today: “Oracle Updates Apps for Next Gen Cloud,” Sean Michael Kerner, Sept. 20, 2016
Enterprise Irregulars: “Oracle AI: Adaptive Intelligence,” Vinnie Mirchandani, Sept. 21, 2016
eWeek: “Oracle Launches Major AI Initiative With ‘Intelligent cloud Apps',” David Needle, Sept. 19, 2016
FierceCMO: “AI brings great potential to B2B marketing,” Patrick Gorman, Sept. 30, 2016
Fortune: “Salesforce and Oracle Vie to Prove Whose Apps Will Be Smarter,” Heather Clancy, Sept. 19, 2016
IoT Evolution World: “OpenWorld: Oracle is Dropping All Sorts of IoT Cloud Knowledge,” Ken Briodagh, Sept. 19, 2016
ITWeb: “Oracle ups cloud game in a bid to outdo rivals,” Admire Moyo, Sept. 20, 2016
The Register: “Oracle: We're going to be the practical AI people, we swear it,” Shaun Nichols, Oct. 11, 2016
ZDNet: “Oracle and Salesforce and IBM? Oh My! Here comes AI sprawl,” Larry Dignan, Sept. 19, 2016
ZDNet: “Oracle vs. Salesforce on AI: What to expect when,” Doug Henschen, Sept. 21, 2016
There is a new technology available from PeopleSoft that makes it very easy to configure Fluid Appovals. It’s called Enterprise Common Components Page Composer or often referred to as just Page Composer. With Page Composer, administrators can modify the contents, layout and format of Fluid Approval Pages without having to customize the underlying PeopleTools managed objects (pages and records).
Fluid Approvals are often one of the first product areas that many of you distribute to users to use on mobile devices. They are also an area of the product that each organization likes to tweak to meet their needs. Previous versions of mobile-ready approvals were built with IScripts and Mobile Application Platform (MAP), so customizing the UI was difficult and required developers. Now with Page Composer, changes are made using a simple drag-drop UI that doesn’t require complex programming.
The beauty of the Page Composer architecture makes it so that no PeopleSoft page is modified or customized, and the changes take effect immediately. That means the change does not need to be managed when updating underlying PeopleSoft objects and code. In Fluid Approvals, each approval type and device size has its own composition (page layout) providing full flexibility to customers when deciding what data to display.
It’s important to point out that in order to use Page Composer, you must use the new Fluid Approvals that are built for use with Page Composer. Application teams will make the first set of approvals available in the following PeopleSoft Update Images, FSCM image 20, HCM image 20, CRM image 12 and ELM image 15. It’s also important to mention that Page Composer only supports Fluid Approvals and is not something that can be used with any Fluid page. We’ll let you know if that changes.You can read more about Page Composer in the Tech Update found on MOS Note 764222.1.