In honor of all of those earth-bound, quasi-sentient, primates that believe the world is going to end in a few days
I present proof that Larry Ellison and Oracle are part of the grand Mayan conspiracy
and that I have personally initiated the earth's destruction in accordance with prophesy.
SQL> exec dbms_internal_logstdby.cancel_future;
PL/SQL procedure successfully completed.
Everyone with at least one free synapse can just smile and go about enjoying the day.
Thursday 29 November, 2012
One of the true joys of being a geek not totally given over to the Boolean world is the ability to appreciate when technology meets biology meets a sense of humor.
Exhibit A, at left, is for your consideration.
I hope her feelings won't be hurt ... but I'd be just as happy to meet telomerase.
Thursday 08 November, 2012
Does your data center look like this?
I can not tell you which of my former customers is the proud owner of this worm infestation but rest assured I've held this
picture for quite some time to make sure no could could yell "foul" in my direction.
How many connections are terminated in mid-air?
How many connections are terminated on the data center floor?
How many cables are labeled?
How many cables are color coded to indicate what they are?
What do you think are the odds someone could navigate your way through this mess and find the correct pair without accidentally
plunging their infrastructure into a black hole?
Network Admins ... trust AND verify!
Monday 05 November, 2012
I am now deep into the 12c Beta and have come to the realization that
Morgan's Library, as this site has been structured for the past 12 years, needs to be updated to match the new architecture of
Container databases and Edition Based Redefinition.
Thus a major change is underway which will be implemented beginning shortly after the release of 12c likely in December of this
year. Watch for it and let me know what you think when I release it ... I'd like to do so now but with the NDA in place that would
be a bit akward.
The current library is based on a single database named ORABASE with a tablespace named ORADATA and primarily using a schema named
UWCLASS which originated with my first year teaching at the University of Washington. The graphic, below, shows what I expect will be the new design.
The concept will be that the root database will now be named ORAROOT and will have
three PDBs: ORADEV, ORATEST, and ORABASE all with the familiar tables, SERVER, SERV_INST, AIRPLANES, but these object names will
refer to the editioning views on top of the base tables not the tables themselves.
The vast majority of the demos, those of DML and DDL will continue to be based solely upon the ORABASE "production" database but
demos best demontrated in a different environment, for example those related to new functionality may be baed on the CDB or one of
the down-level PDBs.
Thursday 04 October, 2012
Morgan's quote of the day:
Never attribute to conspiracy ...
what can explained by complacency
Thursday 04 October, 2012
OpenWorld is wrapping up ... time to go pay homeage to Larry's little boats.
Wednesday 03 October, 2012
The only twin brothers in the ACE program pose with an ODA.
Thursday 27 September, 2012
A limerick for Chris Date by Helen Lofstrom ...
composed somewhere between Tallinn Estonia and Seattle during our return from Harmony.
Wise sage Mr. Date said to me,
What you think to be true may not be.
Of NULL: "It's too vague; avoid it like plague!"
And "NOT is NOT not, don't you see?"
Wednesday 19 September, 2012
There is something about getting ready to go to OpenWorld each year that makes me rummage through old emails and old directories
looking for unfinished business. Well I have too much of it to face at the moment but I've pulled together some pictures and am putting them here just to take a little low
hanging fruit off the top of the pile.
Heli's son Mateus had something to do with this.
It was 2010 and we
were at Sandanski Bulgaria. Stanley wanted to get medieval
Who was I to say "No!"
People often want to know what color of red Stanley is ...
I believe we have finally identified it ... Amanita muscaria red.
What do you think?
Oracle Biplane ... pic from the house looking
west over Lake Washington toward downtown Seattle
Blue Angel 5 Inverted
Blue Angels Flying In "Close" Formation
Blue Angels: Two Inverted
Tuesday 18 September, 2012
Google Chrome ... by any reasonable definition ... is malware!
If there ever was a day when Google lived by the phrase "don't be evil" that day is distant history. Yesterday I installed Google
Chrome to try it out: It was Ok but nothing all that special. Today I opened Firefox and found that Google had intentionally
modified another company's software without asking permission. That was enough motivation to get me to uninstall Chrome. No one has permission to
alter another vendor's software, and my personal customizations, without asking permission.
And that is when I discovered the evil. It turns out that just unstalling Chrome is only the beginning. The next afront to my
dignity was that Google had intentionally changed the default search engine in Firefox ... I changed it back and delete Chrome as a search
choice. But that still wasn't enough. No they weren't happy with just a little evil so I had to go in and remove their browser
extension to Firefox, also installed without permission and not removed during the uninstall, by going to the address bar and
typing in "about:addons." And that still didn't do it. Open a new tab and the now hated Babylon search engine still pops up as a
search default on the new tab ignoring my choice of a default homepage and a search engine. So now I am in the registry ... and the words "Babylon" and
"Chrome" will not exist when I am done. Because, as I am doing it, I can see their uninstall program intentionally did
not perform a real uninstall ... it did not clean up the mess it made
... and that too was intentional too. Right down to installing babylon.xml in Program Files\Mozilla Firefox\searchplugins and
not removing it when being asked to do so. Also, if doing this, in the address bar type "about:config" and remove everything with
the word "babylon" in it. You can do this by right-clicking on everything "Babylon" and choosing "Reset." Then scan the hard disk for
other directories with the names Chrome and Babylon and remove them too ... you'll find them well hidden. Then reboot.
Google ... fair warning ... one more violation and I am switching to Bing or Yahoo: You have less than zero credibility and that
extends to Android. I will now not be replacing my current phone with one running your O/S.
Also a quick note to Mozilla ... you need to proivde an easier way to get rid of malware if you want me to continue to use your
Google Chrome is malware! Do not install it!
Monday 10 September, 2012
Andy Mendelsohn has given the public a small peek under the kimono. I can not, and
will not, comment beyond what is here.
So "No!" I am not violating my NDA.
Saturday 08 September, 2012
Some of us write every day ... some of us only when motivated ... today I am motivated.
Why you might ask? Because I've run into another one of those inexplicable situations where a DBA is technically competent,
knows what is going on, and discounts it as an issue not requiring attention. Here are the facts.
A new database has been stood up
RMAN backups are being taken and validated as being "good"
But the RMAN log, for every backup reports a warning like this:
START_TIME END_TIME INPUT_TYPE STATUS
-------------------- -------------------- ------------- -----------------------
05-SEP-2012 23:30:02 05-SEP-2012 23:32:01 ARCHIVELOG COMPLETED WITH WARNINGS
06-SEP-2012 03:30:02 06-SEP-2012 03:31:18 ARCHIVELOG COMPLETED WITH WARNINGS
06-SEP-2012 07:30:02 06-SEP-2012 07:31:24 ARCHIVELOG COMPLETED WITH WARNINGS
06-SEP-2012 11:30:02 06-SEP-2012 11:31:20 ARCHIVELOG COMPLETED WITH WARNINGS
06-SEP-2012 15:30:02 06-SEP-2012 15:31:10 ARCHIVELOG COMPLETED WITH WARNINGS
06-SEP-2012 19:30:02 06-SEP-2012 19:34:37 ARCHIVELOG COMPLETED WITH WARNINGS
06-SEP-2012 20:00:02 06-SEP-2012 20:30:25 DB INCR COMPLETED WITH WARNINGS
The reason for the warning is benign ... the current RMAN script is trying to delete archived redo logs still required by the
system for Data Guard.
So what is the issue that has me motivated enough to call this out in my blog? The backups are good ... what's my issue? After all
... I am perfectly willing to stipulate that the current backup is good and the warning not an indication of impending failure?
Simply put it is the fact that when there is a warning ... one that "everyone" knows can be ignored ... they will
precisely that: ignore
it. When the OEM alert indicates a backup with a different warning no one is going to take the time to log onto the host, review the RMAN log,
and see if the warning has changed to something less benign. Thus ingnoring the current warning is a guarantee that, with time, an
important warning will not serve the intended purpose of actual providing a warning: it too will be ignored.
If there are no warnings then, by definition, people will be alerted by the presence of a new warning message and read it.
So my friends ... given a choice of a bad practice or a good one what should be done here? How about rewriting the RMAN script so
it doesn't produce the warnings? Something that would take ... oh ... maybe three to five minutes.
Tuesday 28 August, 2012
Morgan's quote of the day:
Your SQL didn't go bad ... it was born that way
Sunday 8 July, 2012
One thing that should be antithetical for all Oracle DBAs and
Developer is a belief system not founded in objective reality:
Such belief systems should left to the likes of network admins and Windows devs.
Cartoonists have an elegant way of saying, simply, that which
needs t be said ... XKCD has done it again.
Saturday 7 July, 2012
I have no idea where I was
or what I'd been drinking
but the person that resembles
me in these pictures is not an
imposter ... so as long as these
made it to Facebook they can
make to my blog too.
A great group of good looking people ... whoever they are ;-)
Tuesday 3 July, 2012
Hans Forbrich and I were selected to do presentations at OpenWorld on the newly released (September 2011)
Oracle Database Appliances, ODAs, and it makes sense as I am consulting for a client that has, so far, purchased and deployed 18 of them, and Hans is tall, good looking,
and absolutely faboulous at presenting technical subjects to a audience with a wide ranging level of expertise.
A week or two ago I was up in Vancouver and Victoria BC presenting at OUG meetings and Oracle provided an OGA to"wow" everyone by having green lights and
reasonably quiet fans. They were right ... everyone wanted to see the little beast ... and I wanted to take it home with me. Hans just happened to be in Victoria
teaching an Oracle RAC class, brought his students, and somehow we convinced Oracle to let us "smuggle" the ODA back across the border to the U.S. Oracle
put only two provisions on us ... (1) we had to actually return it to Oracle and (2) we had to do something useful with it in preparation for OpenWorld.
Now ... with many hours of experience under my belt not just from real-world deployments at my customer site but from dissecting the gift ODA I have published a new
page to Morgan's Library that should serve as a reasonably good roadmap to ODA configuration based on the June 2012 software release (Database version 18.104.22.168).
My feelings at this time having personally built well in excess of 200 RAC clusters ... now that ODAs and Exadatas exist ... you need a psychiatrist if you think it
is a good idea to build one yourself ... with two, and only one caveats. They are: (1) you can prove, with math, that you need more than two nodes and (2) can can prove,
again with math, that you need more physical horsepower than can be provided with anything short of IBM P795s and clustered EMC VMAX SANs plus.
Saturday 16 June, 2012
... One small step for a woman,
one giant leap for mankind
Saturday 09 June, 2012
I've been a terrible correspondent again ... but here is a quick post with a link to point anyone interested to
the pictures from the trip to Russia, Finland, and Latvia including the Harmony conferences outside Helsinki and inside Riga. I will enhance this in the next week or
two before heading to Canada. And what I am I doing in Canada you might ask ... more writing to do. [ Click Here].
Tuesday 05 June, 2012
Want another perspective of how important we,
and our little planet, are?
Venus is only slightly small than earth.
Thermonuclear global warming is coming.
The question of what we choose to do in the
next couple of billion years with our time is
still an open question.
Sunday 29 April, 2012
Oracle signage: Schiphol Airport
Sunday, 22 April - Saturday 28, 2012
Finally in Amsterdam and outside Schiphol. Arrived at the Hilton DoubleTree, a Novotel-like setting,
to catch a few hours sleep, and then go out walking until dinner time when I've designs on the world's best Indonesian food for dinner.
Being in an Amsterdam reminds me of a great teacher we lost a few years back so he is honored on the left side of this section.
More about the Maastricht conference later other than to say the members truly represent the Oracle community with a substantial
number of members here for Apps and JDE ... not just database. But those with the database as their specialty are very technical and very eager to learn more.
I've left Maastricht for lunch and a long walk in Luxembourg. Why not ... I've days before I have to fly out.
Spent last night in Antwerp Belgium, after travelling from Bruxelles, OMG! Antwerp is a gem.
I am totally impressed and definitely planning to return.
Travelled back to Schiphol via Rotterdam, Delft, and Den Haag. Spending the night here and first thing in the morning returning to Seattle.
My customer, they I'll not name them, should be very happy upon my return.
Not only were their offices quiet for a week but they now can proudly claim an application written in the Netherlands, Luxembourg, and Belgium.
[Click Here] to view the trip map and pictures.
More pictures are coming as I continue work on the web pages
Now I know this link won't be especially popular with some that follow my blog by to honour Mr. Carlin,
[Click Here] to read something that will keep the synapses firing.
Friday, 20 April, 2012
Tomorrow morning I leave for the airport to go to my first conference since Mark's passing and I do so
with mixed feelings. Too short a period of time and in a few days I will be walking the walk in front of an Oracle user conference
that will bring me tremendous happiness and remind me of something lost to both myself our community.
On a positive note two wonderful things happened today. The first is I received a copy of Rich Niemiec's new book "Oracle Database 11g Release 2
Performance Tuning Tips & Techniques." It is coming on the plane with me and I plan to devour it. I will write a review on my site for it
as soon as I can.
The other bit of good news is that I have the wonderful good fortune of working for a truly competent IT leadership team and while I can not go into details
here ... I wish everyone could do what I did ... present the facts ... and then receive a response based purely upon those facts leaving aside all of the
politics and emotions that too often disappoint everyone and turn IT projects into well publicized failures.
But I will say this about the subject in the hope that others may take away something of value.
The most senior DBA and I produced a proposal that pointed directly to a mixed environment of ODAs and Exadatas with ASM, RAC, Data Guard Physical Standbys
and other state-of-the-art proven technology as the way to go. Today I watched two such people in upper management do precisely that and it was a welcome sight.
Now for a good night's sleep and off to the airport at sunrise.
Friday, 13 April, 2012
So much I wish
I could say
Monday, 19 March, 2012
It was exactly 23 months ago, to the day, I was here at the University of Oslo,
teaching a class on Oracle, when members of Alice's staff found me the only flight leaving OSL going anywhere, and luckily that "anywhere" was
Northwest to Keflavic Airport in Iceland.
Bryn Llewellyn joined me in presenting at the university and this evening for dinner, accompanied by Bryn, made a pilgrimage to
Statholdergaarden. I would describe in delicious details the meal ... but it was indescribable.
I am already looking forward to returning next year to again worship at a white linen alter.
But back to the university. Bryn spoke about Edition Based Redefinition and I spent a bit less than an hour on developer topics and then we had lunch with the
DBA team during which I presented on DBA Best Practices. Both of us received copies of a book "IT-Historien @ Universitetet i Oslo"
written by Per H. Jacobsen. My very first adult book ... all in Norwegian I can not read the words but the pictures are great as they cover 40 years of
IT history much of which I participated in so it is very meaningful. Thank you Alice.
I will be posting pictures soon so watch coming installments.
Monday, 12 March, 2012
Today we started configuring the first of three Oracle Database Appliances (ODA) in the lab.
These servers are absolutely beautiful. As everyone that has known me knows ... I really enjoyed the physical elegance of the Apple XServers
that we used to create the 24 node RAC cluster on the third floor of Moscone West in 2005: These machines are just as functionally elegant.
The top four drive bays are SSDs for redo ...the bottom 20 bays hold 4TB of triple mirrored ASM: As fast as what one would expect from an engineered solution.
Full configuration time, from initial boot-up, 51 minutes.
Here's the back of one of the ODAs.
The door in the upper right is open showing the two 500GB RAID 1 disks that hold the Linux O/S and Oracle database binaries.
Sunday, 11 March, 2012
The online documentation at tahiti, on compression,
has been creating more than a small amount of confusion in the OTN forums so I pulled together some of the issues and took them directly to the writers.
The following is the response edited only to remove personal information.
From: Janis G@oracle.com [mailto:firstname.lastname@example.org]
Sent: Monday, March 05, 2012 2:43 PM
To: Daniel Morgan; eric p
Subject: Please try this Tahiti search in 11.2 : resolution of Advanced Compression Doc Issue
Thank you for pointing out a few problems for us.
The licensing description is described in the Oracle Licensing Guide as follows:
Oracle Advanced Compression
Oracle Advanced Compression provides comprehensive data compression capabilities to compress all types of data, backups,
and network traffic in an application transparent manner. With Advanced Compression, Oracle includes table compression targeted at OLTP workloads,
resulting in reduced storage consumption and improved query performance while incurring minimal write performance overhead.
Advanced Compression can be used to compress any unstructured content using SecureFiles Compression.
Deduplication provides the ability to automatically eliminate redundant copies of SecureFiles data. RMAN offers multiple backup compression levels,
to meet the needs of various environments, including faster compression and better compression ratio choices.
Note that the RMAN DEFAULT COMPRESS option does not require Advanced Compression. With Advanced Compression,
Data Pump exports can now be compressed to reduce disk space requirements. Note that the COMPRESSION=METADATA_ONLY option for Data Pump does not require
Advanced Compression, and is the default behavior. Finally, Data Guard can compress redo data resulting in reduced network traffic and faster gap resolution.
The Oracle Advanced Compression option contains the following features:
o Data Guard Network Compression
o Data Pump Compression (COMPRESSION=METADATA_ONLY does not require the Advanced Compression option)
o Multiple RMAN Compression Levels (RMAN DEFAULT COMPRESS does not require the Advanced Compression option)
o OLTP Table Compression
o SecureFiles Compression and Deduplication
As you can see, not everything under this heading is about compression.
Based on your email (thank you), we found several documents that indicated they had a feature named Advanced Compression,
when they really had a feature of Advanced Compression. Based on this, we are changing Spatial GeoRaster Developer's Guide to discuss Oracle Database SecureFiles
Compression and will change the Database Backup and Recovery User's Guide to discuss the Multiple RMAN Compression.
So, the question now is, have you enabled one of these features and need syntax information for it?
Or do you need syntax information for another form of compression that you are not able to find. In either case, let me know and I'll see what I can do.
Sunday, 12 February, 2012
I didn't mention this yesterday as I was hoping United Airlines would redeem itself: It failed.
The level of incompetence is the stuff of infamy. Did United lose my checked suitcase from Boston to Denver? No. Did they take
it to the plane in Denver? Yes. I saw it looking out the window from seat 12F. Did they then drive away with it and 14 other people's checked baggage?
And what marvelous gift did they give me when they didn't phone me after it was delivered to Seattle to tell me it had arrived?
They gave me the gift of breaking it and then claiming the breakage wasn't their responsibility. Of course not United. The wheels falling off a roller-bag
when you lost it is purely coincidental. But don't worry United ... I am going to extract my revenge and it is going to cost your many times the cost of a
new case. Perhaps you are not aware of this but other airlines exist. So, in April, I am flying Air France.
Want to guess which airline will be taking me to Finland in May? Good guess ... someone other than the UNfriendly skies.
Here's a couple picture I took over the Cascade's just prior to arriving in Seattle. A latte' to the first person that can name that volcano.
Saturday, 11 February, 2012
Flying United back to Seattle with a brief plane-change in Denver.
I have a new souvenir: One that will take its place along with my conference badges.
I suspect it is easier to speak at OpenWorld than at Harvard: But, so far, that is just an unverified suspicion.
Friday, 10 February, 2012
A week's worth of teaching Oracle is over ... time to pack for the return to Seattle.
Tomorrow morning, at least in theory, it is going to snow here. Everyone in Boston seems to understand the implications of their lost winter
just as did those in Chicago when I stopped at O'Hare on my way here. One has to wonder how it is that a subset of our population can be so hard-of-thinking.
Just completed the final act in a not-very-complex, but highly enjoyable, deception. It is always fun when you can create a totally unanticipated
reunion between people that haven't seen each other in a long time.
Thursday, 09 February, 2012
And yet another great day in Cambridge. The sky is blue ... the sun is shining nothing is green ...
it is not Seattle. Sorry East Coast but given a choice between sun and green ... I choose green.
That said ... there is something awfully nice about being back on campus: That I will miss.
Wednesday, 08 February, 2012
Another great day in Cambridge ... and a reminder for anyone that might have forgotten,
and I take that personally, DataPump does not export grants on objects owned by SYS. Why I don't know so don't ask. ;-)
Tuesday, 07 February, 2012
Harvard University: Longwood Campus
Shattuck International House
Monday, 06 February, 2012
Had dinner with Frank Kulash, someone you must already know,
if you spend any time at all on the OTN Database forums. Frank has the skill, and the patience, to provide a continuous stream of high-quality,
well considered, attitude-free advice. I am going to recommend Frank as an ACE for both the quality and the quantity of his contribution to the community.
But on another subject dear to my heart ... we met at a restaurant in North Boston's "Little Italy" I had not previously visited: Rabia's.
This Italian seafood restaurant is world-class and, in addition to the wonderful menu items you can find on their website ... check out the chalkboard specials.
I started with the Portabello mushroom appetizer, my main was a lobster casserole,
and finished with a Tiramisu better than any available in the Pacific Northwest.
Sunday, 05 February, 2012
I sitting in my faculty housing apartment on the campus at Harvard University ... Veritas indeed!
Academically I didn't have the work ethic, or the attention span, to walk through these doors as a student.
Tomorrow, at 09:00, I walk through them as an instructor. To say I am humbled, something those that know me well may find hard to believe,
would be an understatement.
Tomorrow evening I am sitting down for dinner with Frank Kulash: One of the brightest lights in the arena of Oracle developers.
His posts in the OTN forum are nothing short of excellent. Perhaps by the time the tiramisu is gone he will agree to be an Oracle ACE.
Thursday, 19 January, 2012
Here at Morgan's Library, are dedicated to the freedom, and freedom to exchange both knowledge and information.
If you are in the U.S. please contact our President, and your legislators in the Senate and the House of Representatives.
Tell them you are their constituent, that you oppose SOPA and PIPA, and to get your vote they must oppose them too.
If you are not in the US please, in whatever manner is appropriate in your country and your culture join the fight for freedom from oppression.
So to quote the song:
Why, why are laws things you can buy?
They got paid off, should be laid off, re-election denied
Our web means more than lawyers, lobbies, and lies
So speak up before the internet dies
Speak up before the internet dies