Fedora 21 has arrived and it’s now delivered in three focussed flavours (Workstation, Server and Cloud). This of course resulted in the usual articles from me.
As always, read the warnings before you start down this path.
From an Oracle installation perspective, it’s almost identical to Fedora 20. I chose to use the server flavour and install the “MATE Desktop” package group. I suspect others may prefer to start with the workstation release. Either way it should be fine.
As I suspected, switching my main desktop from Fedora to the MacBook means I care significantly less about this release than before, but I still have some upgrades I’ll need to plug through.
The new versions of the WebLogic 12cR2 and ADF could not have come at a worse time for me. My top priority is learning about the 12cR1 version of the database. Second, is getting to grips with Cloud Control 12cR3. Third on the list is getting up to speed with the changes in WebLogic 12cR2 and ADF. Unfortunately, my personal priorities don’t quite match my work priorities, so WebLogic 12cR2 has moved up the list for a while. As a result, I did some installations last night.
I’ll have to play with this for a few days to get my head round it as I may be doing a test installation at work early next week.
A word of warning. I am a self-confessed WebLogic newbie. If you ask me questions about it I will be redirecting you to people who actually know what they are talking about…
I finally got my server problems sorted out yesterday, so I was able to do a 12c virtual RAC installation. You can see what I did here:
The setup is pretty much the same as the 11gR2 setup. So there’s no real drama at all.
With the basic installation articles out of the way I can start having a play with the functionality.
I’ve now got 12c running under VirtualBox on my server (Oracle Linux), desktop (Fedora 18), laptop (OS X) and desktop at work (Windows 7). If I can just get it running on my Nexus 4 and Nexus 7…
I did an EM Cloud Control 12cR2 installation at work yesterday. The database repository was 188.8.131.52 on HP-UX and the middle tier was installed on RHEL 5.8. The installation was pretty much the same as the 12cR1 version. Over the next few days I’ll be testing out some of the features to decide if we can move across to it permanently.
Today I did two run throughs of single server installations on Oracle Linux 5.8 and 6.3. There are a couple of minor differences, but nothing to worry about. You can see what I did here:
The installations are a little small, so they are not too fast, but it’s good enough to test things out.
Update: It’s been a while since I used the 12c version, so I’ve had to relearn a few simple things. I thought I might as well write the down in case it helps others.
While I was at Open World I tried a few times to get hold of the new Cloud Control software, but the hotel network wasn’t up to the job, so I had to wait until I got home.
The installation is pretty simple compared to previous versions of Grid Control and it installs fine on both Oracle Linux 5.x and 6.x. As always it’s a little greedy on the memory front, with the recommendation for a small installation being 4G for the Cloud Control and 2G for the repository database. That’s not including the OS requirement. On the subject of the repository database, you can use a number of 10g and 11g versions, but anything before 184.108.40.206 requires additional patches, so I stayed with 220.127.116.11.
You can see what I did here.
I had this comment today related to RAC installation.
“thanks for the feedback, but for newbies this is where it gets confusing. No clear guidelines”
This post is not specifically about this comment, but it does bring up the issue I keep going back to again and again…
One of the things that annoys me about the Oracle marketing machine is they still try to make out all Oracle products are accessible for newbies. Oh really? Are you seriously telling me that Oracle RAC and Oracle Grid Control 11g are accessible for newbies?
I’ve been using Oracle products for about 17 years. I’ve been using Linux for about 13 years. I’ve been administering RAC for about 10 years. I don’t claim to be an international consultant to the stars, but I have a long history with this stuff. I’m not saying this to brag, just to put this into context. With all this experience I still don’t think this stuff is easy.
Check out the Oak Table Members list. Excluding myself, this is a “who’s who” of the people you would love to have on your site to show you how Oracle stuff really works. If you were part of the Oak Table mailing list you would see these people are still struggling with the idiosyncracies of some of this Oracle stuff. There are lots of RAC related issues under discussion all the time.
Knowing all this, do you really think you can roll up off the street and do a good job of installing and administering this stuff in a production environment? Do you think it is OK to be an SQL Server DBA on Windows today and start a job as an Oracle DBA on Linux tomorrow? I see this happening all the time because bosses don’t understand how complicated this technology can be. People do one Oracle installation on Windows and think the logical next step is RAC or Exadata.
I’m happy that Oracle have invested time and money in making Oracle *easier* to install and administer, but trying to tell people that it is easy is totally the wrong message. A week long course or a 2-Day DBA manual is not going to get someone up to speed.
For the next marketing slogan I suggest,
“Oracle. It’s f*ckin’ complicated, but it’s really cool!”
Rant over … until the next time…
Fedora 14 is here and so are the obligatory articles:
My attitude to Fedora and Ubuntu as changed today, with most of that shift due to VirtualBox.
Before I switched to VirtualBox I was always reliant on my OS being able to run VMware Server. Over the years I had repeatedly encountered problems running VMware Server on Ubuntu and Fedora. Not all of them show stoppers, but enough to put me off them as my main desktop OS. Why did I stick with VMware Server? Just because it supported shared virtual disks, which allowed me to easily create virtual RAC installations. Version 3.2.8 of VirtualBox included support for shared disks for the first time, so I ditched VMware Server and launched full scale into using VirtualBox.
While I was playing around with Fedora 14 I was thinking how cool it would be to have a newer OS on my desktop that could run Google Chrome, then it dawned on me that now I can. I’ve been free of VMware Server for a while now and I hadn’t realized the knock-on effect of that.
My years of using RHEL mean I feel a little more comfortable with Fedora than Ubuntu, but to be honest all I do on a desktop is fire up VirtualBox, use a browser (preferably Chrome) and use a terminal for SSH. Virtually everything else is done in VMs.
Now, do I waste a few days assessing the various options for my desktop, or do I just stick with CentOS and deal with the fact I can’t use Chrome on it?
Well a new day and a fresh pair of eyes and it all went well. I was out this afternoon so I started the final bit of the installation running and it had completed successfully when I got home.
I’m not totally sure what was causing the previous problems. I had been taking snapshots of my VM at regular intervals over the last couple of days and I guess something must have been dodgy because when I threw it all away and started again from scratch it worked fine. Here is the installation guide listing all the steps:
So does this success change my opinion of this release? Not at all. The installation is a mess and I think Oracle really do need to have a shrink-wrapped install, even if it is a 8G download. It would reduce the barrier to entry and I just think it feels a little poor that you have to manually install a bunch of patches before you can get the product working. Makes me wince a little.
I’ve spent the last couple of days failing to install OEM Grid Control 11g on OEL 5.5 x86-64 with a 11gR2 database for the repository. The installation process is horrid. You have to manually install and patch the database and middleware software before starting the GC installation. So you end up with a whopping 7.1G of software, not including patches and the OS.
The docs are not ideal. There is a lot of cross-referencing to bugs, patches and metalink notes, which means I’ve often had about 10 browser tabs open while performing the installation. I can only assume that somewhere in the mish-mash of the docs I’ve missed something out.
I feel really disappointed with the installation process for this release. In my opinion there should be a single installation that includes the middleware, database and all necessary patches. I like to think of GC as a shrink-wrapped product I can install separate to everything else and leave alone. Not any more…
I’m off to ODTUG this week, so I’m not sure I can be bothered to waste more time on this until I get back. Perhaps someone there will be able to explain to me what the hell is going on with it.
I’ve had my first play with 11gR2 today:
Nothing too unpredictable really.
I guess the most noticeable change is the new installer. I didn’t have an issue with the old installer, but a few friends from the Microsoft world had pointed out how dated it looked. The new is much cleaner, and although it will take some getting used to, I think it is a step in the right direction.
I’ve decided that I’m only going to do 64-bit installations from now on. I see little point doing the 32-bit installations, as I hope I will never work on one again.