Questions & Answers to replatforming webcast

Q&AA list of questions being asked to my most recent webcast on Oracle replatforming together with Madora Consulting.

I decided to put the Q&A on the blog so anyone can benefit instead of just the 100+ attendees we had on the webcast (awesome!)

Disclaimer: Some Q&A are about licensing and my answers are according to the best I know, but always consult with an independent expert (like my copresenter Keith Dobbs who I reviewed the Q&A with). Also the topic is highly controversial so don’t feel offended if you disagree ;)

Question: For normal data guard should i have to pay for normal DB EE or SE licenses or no need ?

Bart: Data Guard is free but requires both primary and standby host to be fully licensed. It requires Enterprise Edition. If you use advanced features such as running read-only queries on the standby, then you need Active Data Guard licenses on both nodes. Note that there is no technical difference between “normal” and “active” Data Guard. You need to pay for the license as soon as you use the Active Data Guard features.

Question: I’d like to know about replication Virtualized Oracle DB Server using Vsphere , should I worry about consistency?

Bart: VMware (vSphere) does not buffer any IO and passes any writes directly to the underlying storage. Consistency is therefore not dependent on whether or not you are running virtualized or physical. If you use any kind of storage replication (such as Dell EMC Recoverpoint, SRDF, VPLEX etc) then you must make sure that all volumes for a database are replicated as one “consistency” group, such that after a failover you don’t get timing issues between the different database volumes. As said, this is not depending on whether you run virtual or physical.

Also, I highly recommend testing D/R failover at least every 6 months or so, to make sure your D/R strategy is working in case you really need it. I have experience with customers who added local volumes but forgot to replicate those volumes which resulted in non-recoverable D/R systems. Using Dell EMC snapshot features of your storage array allows you to perform D/R testing during normal weekdays so your primary database and DR replication stay active even while you perform D/R testing (a method that you cannot use the same way with Data Guard unless you break replication or use other workarounds).

Question: How to use VMware in a DR site and the production site has engineered system ? Without high cost

Bart: Interesting scenario. Assuming Engineered System is either Exadata (most likely) or Supercluster (where the storage is the same as Exadata), the challenge here is that Oracle does not allow connectivity of non-Oracle storage, so we cannot use classic SAN replication here. We need to discuss in more detail but a few options you may consider:

  • Backup Exadata with EMC Data Domain, replicate to another Data Domain in the D/R site and use that to recover your database (you can use it to daily update your standby using replicated backup files). This would be a very cost efficient solution but the trade-off is that it would take some time to recover after a failover (so the RTO – recovery time objective) would not be close to zero.
  • Use Data Guard (the free version i.e. not using Active DG features) to replicate to a (Dell EMC) system. There are some disadvantages here and care must be taken that you don’t use Hybrid Columnar Compression – as this would have to be de-compressed first in order to start the standby database
  • Best solution: re-platform production from Engineered System to a Dell EMC proven solution and then you can use any D/R method you like, as well as save on license cost as we have shown in the presentation

Question: How are Madora (consulting) funded / paid for ? How does the local DEll EMC (Think DELL folks) Business Units generate revenue from a Madora engagement?

Bart: I assume this question was asked by one of our Dell EMC colleagues on the call. But sharing it with everyone because no secrets…

I’ve had this question before several times so it’s an important point, and we should have that in a future version of the presentation. There is no formal partnership between Madora and Dell EMC, because Madora must be able to provide a completely independent advice to customers and not be biased because being sub-contracted by Dell EMC. So for Dell EMC sales teams there is no direct revenue stream, however, the chance of winning infrastructure deals increases because

  • customers can save money and spend it on the infrastructure that is needed to achieve those savings
  • customers can eliminate uncertainty and make sure they are – and stay – compliant and don’t have to worry about project risks
  • bringing in license expertise can eliminate the objections raised by the customer on licensing, support, audits etc.

The relationship is purely an informal one but that said, we’ve cooperated succesfully on a number of occasions.

Question: How and why can a Customer move to Standard Edition?

Bart: If you have a database on Enterprise Edition and want to move to Standard Edition (Standard Edition 2 – SE2 – in future deployments as of Oracle version 12) then you must consider if there are any Oracle features in use that are not available in SE/SE2. If that is not the case, you can install a server with SE and move the datafiles, then restart the database. Again, test the scenario before go live as the devil is often in the details. Also you need to make sure you have enough SE(2) licenses because you are not allowed to run Standard Edition using Enterprise Edition licenses (it is essentially a different product). Madora may be able to help converting or re-negotiating licenses. Oracle will probably push back because it’s a huge reduction in their revenue stream.

Question: Can you convert Exadata quarter rack to ovm and reduce the licenses of DB and reuse them in DR site ? without buying new licenses.

Bart: (assuming OVM is Oracle Virtual Machine) – As long as the total number of physical processors is licensed then that should be possible from a licensing standpoint. Note that with OVM you have many of the same (perceived or real) license issues as with VMware (but Oracle doesn’t tell you that). Note that I would not recommend running OVM for consolidation of mission critical systems as it’s not best practice and does not offer the same level of performance management, 3rd party tooling integration and features as VMware, however, on Exadata it may be your only viable option next to running physical. I haven’t seen any customers yet who run OVM on Exadata for mission-critical workloads yet (but there sure will be a few out there – let me know your experiences!) so make sure you know what you are doing before starting such a journey.

Final thought:

I get many more questions during my customer meetings. Maybe I should put a static FAQ page on the topic that I update with new questions, answers and insights… Let me know what you think!

This post first appeared on Dirty Cache by Bart Sjerps. Copyright © 2011 – 2017. All rights reserved. Not to be reproduced for commercial purposes without written permission.

Oracle snapshots and clones with ZFS

Another Frequently Asked Question: Is there any disadvantage for a customer in using Oracle/SUN ZFS appliances to create database/application snapshots in comparison with EMC’s cloning/snapshot offerings?

Oracle marketing is pushing materials where they promote the ZFS storage appliance as the ultimate method for database cloning, especially when the source database is on Exadata. Essentially the idea is as follows: backup your primary DB to the ZFS appliance, then create snaps or clones off the backup for testing and development (more explanation in Oracle’s paper and video). Of course it is marketed as being much cheaper, easier and faster than using storage from an Enterprise Storage system such as those offered by EMC.

Oracle Youtube video

Oracle White paper

In order to understand the limitations of the ZFS appliance you need to know the fundamental workings of the ZFS filesystem. I recommend you look at the Wikipedia article on ZFS (here http://en.wikipedia.org/wiki/ZFS) and get familiar with its basic principles and features. The ZFS appliance is based on the same filesystem but due to it being an appliance, it’s a little bit different in behaviour.

So let’s see what a customer gets when he decides to go for the Sun appliance instead of EMC infrastructure (such as the Data Domain backup deduplication  system or VNX storage system).

Read more of this post

Duplexing Oracle Redo logs?

A customer asked me recently what EMC’s advice is regarding duplexing Oracle redo logs. There is a thought behind this – Oracle redo logs are sensitive to data corruption – if redo logs are corrupt, there is no way to nicely recover the database to a technically consistent state (at least not without restoring data from backups).

This is what Oracle tells you:

Oracle recommends that you multiplex your redo log files; the loss of the log file information can be catastrophic if a recovery operation is required.

 

Duplexed Redo

Duplexed Redo


Read more of this post

Pinpointing I/O bottlenecks on Linux

Does this story sound familiar?

The end users of a database application start complaining about poor system response and long running batch jobs. The DBA team starts investigating the problem. DBA’s look at their database tools such as Enterprise Manager, Automatic Workload Repository (AWR) reports, etc. They find that storage I/O response times are too high (such as an average of 50 milliseconds or more) and involve the storage team to resolve it.

The storage guys, in turn, look at their tooling – in case of EMC this could be Navisphere Analyzer, Symmetrix Performance Analyzer (SPA) or similar tools. They find completely normal response times – less than 10 milliseconds average.

The users still complain but the storage and database administrators point to each other to resolve the problem. There is no real progress in solving the problem though.

Two Way Communications

Read more of this post

How to set disk alignment in Linux

As you might know, if disk partitions containing Oracle datafiles are not aligned with the underlying storage system, then some I/O’s can suffer from some overhead as they are effectively translated in two I/O’s.

If you want more info, google for “EMC disk alignment” and you’ll find plenty of information, explaining the issue.

Update 28-03-2013: I wrote a follow-up for this post describing the same thing for Linux (Red Hat / CentOS / OEL) versions 6. For that, you might want to jump straight to the new post as this one gets a bit outdated ;-)
Read more of this post

How does Oracle keep data consistent on filesystems?

Sometimes there is questioning on how EMC can provide consistent snapshots on storage systems, given the fact that file systems like ext3, ufs and the like may keep write data in file system (server) cache.
Read more of this post