My initial batch of 12c installations resulted in these two articles.
Judging by my web stats, they’ve proved very popular. According to Google Analytics, there are 3-4 active visitors on them pretty much 24×7.
My progress from there has been thwarted somewhat by technical issues. The server I normally use for my Virtual RAC installations is being a little funky at the moment. Rather than mess about I figured I would just reinstall the base OS. Initially it was running Fedora 18, but I decided to go for Oracle Linux instead. This server is used entirely for running VirtualBox VMs, so stability is more important than being bleeding edge. My desktop is still Fedora though.
The Oracle Linux install was fine. The 6.4 spin of Oracle Linux detected all my hardware, including the RAID stuff, and it installed with no worries. The problem is getting all the VMs back on to it from my backups. It’s taking an age! 🙂
Hopefully tonight, or tomorrow I can get back on the case with the 12c RAC installation. Yury from Pythian has already done a run through, which you can see here. There doesn’t seem to be much difference to the infrastructure setup compared 11gR2 RAC, which is good.
Cloud Control 12c is a great product. Yes, it is suffering from bloat, but generally it is a really great tool. I’m always encouraging people to ditch the DBControl and switch to Cloud Control!
Having said that, one area that annoys the hell out of me is the job management, which feels really clumsy. I started to write this post, then felt a bit guilty because I hadn’t actually bothered to raise an enhancement request, so that’s what I’ve just done!
My main gripes are the following.
- Editing Jobs: You can edit some parameters of jobs, but not all. This is really frustrating and often means you have to delete and recreate a job in order to make a minor change.
- Create Like: There is still no support for “Create Like” of a database backup job. Most of our jobs are database backup jobs, so this omission is quite annoying.
- Run Now: There is no “Run Now” type of functionality on the scheduled jobs. You either have to create a new one-off job (with out Create Like), or alter the schedule, then remember to alter it back once the job has finished. A “Run Now” option that doesn’t affect the normal schedule would be really handy.
- Backup Destination: If you use “CONFIGURE CHANNEL DEVICE TYPE DISK FORMAT ‘/path/to/backup/%U’;” from RMAN, or use the Backup Settings screen in Cloud Control, which does the same thing, the path used for the backup works fine. If on the other hand you use the “Over Default Settings” option while creating the database backup job, the settings seem to be completely ignored. This should either work, or be removed as an option because it catches me out all the time!
OK, the last one is a bug, not an enhancement request, but I thought I would throw that into the mix anyway. 🙂
We are in the process of upgrading and moving databases at the moment, so amending the backup jobs is a big thing for us. Yesterday afternoon, evening and night was littered with a chorus of expletives in relation to Cloud Control job management, which gave me the activation energy to post this rant. 🙂
Note. Even as I rant, I am still convinced centralising your jobs in Cloud Control is a good thing!
Update: My enhancement request is being split into three separate SRs and they are being put forward to the development team as formal enhancement requests. Fingers crossed they will get implemented.
The fourth point is a “problem between keyboard and chair”. I misunderstood what this feature was supposed to do. My bad. 🙂
On a yoga course a teacher said to the group, “Don’t try to remember everything I say. Some things will come back to haunt you later.”
So I was reviewing a couple of chapters of Marcelle Kratochvil‘s multimedia book and she mentioned DBFS. That jogged a memory of the DBFS demo stand at OOW12, where the guy told me that 12c will (probably) have WebDAV support for DBFS.
DBFS is a neat feature, but it’s a little frustrating if you are using any OS other than Linux because you are forced to use a client utility with limited functionality, rather than accessing it like a regular file system as you can on Linux using the FUSE project. If this WebDAV functionality does get released in 12c it will make it accessible from pretty much any OS or browser.
Not surprisingly, it uses the XML DB infrastructure, which has supported WebDAV since its introduction on Oracle 9i (here). I seem to remember hearing XML DB is a mandatory feature from Oracle 12c onward, which means you will be able to use DBFS on any database running 12c or above.
Remember, all this was prefixed with a Safe Harbor slide, so there are no guarantees this will make it to the production release…
I did an EM Cloud Control 12cR2 installation at work yesterday. The database repository was 18.104.22.168 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.
Question: What happens when 12c Cloud Control runs out of disk space?
Answer: It doesn’t work very well. 🙂
I have a 12c Cloud Control installation on an Oracle Linux 6.1 VM and I was pushing an agent to both nodes of an 22.214.171.124 RAC, also on OL6.1 VMs. The agent installation seemed to go fine and the agent upload to CC was fine, but when I tried to discover the database on the nodes it went a bit loopy. After a little messing about I noticed my disk was maxed out on the 12c CC server. Bummer!
So I turned off the VM, added another virtual disk, turned it back on and added the new disk to the existing volume. Bob’s your uncle!
Once Cloud Control was running the database discovery worked fine and it was full steam ahead. 🙂
Before I realized my mistake I was starting to think this was OL6.1 related, but as it stands I’ve still seen no ill effects of running Cloud Control or RAC on Oracle Linux 6.1.
I mentioned the day before Open World I put a Virtual RAC on Oracle Linux 6.1 article live. Although the procedure was complete, some of the screen shots were from an old article as I didn’t have time to redo them before my flight. 🙂 I’ve just run through the procedure again and taken new screen shots. As a result, I’ve allowed the article to display on the front page of the website, which is why you will see it listed as a new article there.
This kinda rounds out the whole Oracle on 6.1 stuff as there has been a single instance installation guide out for ages and more recently the Cloud Control installation, which references it.
Remember, it’s still not certified yet, but it’s coming.
Update: It’s finally certified. See here.
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 126.96.36.199 requires additional patches, so I stayed with 188.8.131.52.
You can see what I did here.