May 31, 2012 at 10:57 am
Nice interesting question, thanks.
Very surprised that more than half the answers so far got it wrong.
I didn't know the database id was always 32767, but didn't need two as it said "select 4" and I knew 2 that weren't true out of the six options provided, so that 32767 had to be one of the four true ones.
Agree with Hugo that the use of "persisted" might be a bit misleading.
Tom
May 31, 2012 at 12:39 pm
Nice question, thanks!
May 31, 2012 at 12:44 pm
Wow, great question! Thanks!
May 31, 2012 at 1:25 pm
May 31, 2012 at 2:45 pm
EL Jerry (5/31/2012)
Thank you for the question, Ron.I also got the right answers by elimination of the first 2 false statements.
"El" Jerry.
+1 but like Tom I was wondering about the 32767. Didn't seem like a logical choice for an ID but knew the other 2 were wrong.
May 31, 2012 at 3:32 pm
Cliff Jones (5/31/2012)
EL Jerry (5/31/2012)
Thank you for the question, Ron.I also got the right answers by elimination of the first 2 false statements.
"El" Jerry.
+1 but like Tom I was wondering about the 32767. Didn't seem like a logical choice for an ID but knew the other 2 were wrong.
Go back to the very first days of Microsoft when they were working for IBM to develop a relational DB. In those days, Cobol and Fortran were perhaps the 2 most used languages in computing (outside of the basic Assembly language used by IBM to develop their operating system). In those days 18K of memory was huge. Indirect addressing of memory had yet to be developed, with all those restrictions the largest value that could stored in 16 bits was 32767. So, I am only guessing, that someone at MS with a long memory picked it as the value least likely to create a problem. No facts to back up my assumption, a pure guess on my part, take it for what it is worth .... which of course could be nothing. If you are curious enough use google to look up the IBM 900 or 901 model to learn what the "good old days" of computing were.
May 31, 2012 at 3:44 pm
bitbucket-25253 (5/31/2012)
Cliff Jones (5/31/2012)
EL Jerry (5/31/2012)
Thank you for the question, Ron.I also got the right answers by elimination of the first 2 false statements.
"El" Jerry.
+1 but like Tom I was wondering about the 32767. Didn't seem like a logical choice for an ID but knew the other 2 were wrong.
Go back to the very first days of Microsoft when they were working for IBM to develop a relational DB. In those days, Cobol and Fortran were perhaps the 2 most used languages in computing (outside of the basic Assembly language used by IBM to develop their operating system). In those days 18K of memory was huge. Indirect addressing of memory had yet to be developed, with all those restrictions the largest value that could stored in 16 bits was 32767. So, I am only guessing, that someone at MS with a long memory picked it as the value least likely to create a problem. No facts to back up my assumption, a pure guess on my part, take it for what it is worth .... which of course could be nothing. If you are curious enough use google to look up the IBM 900 or 901 model to learn what the "good old days" of computing were.
Correct me if I'm wrong. Back in the good old days of GW-BASIC the highest line number a program could have was precisely 32767.
"El" Jerry
May 31, 2012 at 3:59 pm
the maximum number of databases per instance is 32767, therefore the maximum DBID = 32767, I would say thats the reason this number was chosen for the resource database, its as far out of the way of potential user database dbids as you can get.
---------------------------------------------------------------------
May 31, 2012 at 4:11 pm
george sibbald (5/31/2012)
the maximum number of databases per instance is 32767, therefore the maximum DBID = 32767, I would say thats the reason this number was chosen for the resource database, its as far out of the way of potential user database dbids as you can get.
If I were using a sequence or an identity to implement new database ID's then 0 would be out of the way. That's what I was thinking was illogical.
May 31, 2012 at 4:19 pm
EL Jerry (5/31/2012)
bitbucket-25253 (5/31/2012)
Cliff Jones (5/31/2012)
EL Jerry (5/31/2012)
Thank you for the question, Ron.I also got the right answers by elimination of the first 2 false statements.
"El" Jerry.
+1 but like Tom I was wondering about the 32767. Didn't seem like a logical choice for an ID but knew the other 2 were wrong.
Go back to the very first days of Microsoft when they were working for IBM to develop a relational DB. In those days, Cobol and Fortran were perhaps the 2 most used languages in computing (outside of the basic Assembly language used by IBM to develop their operating system). In those days 18K of memory was huge. Indirect addressing of memory had yet to be developed, with all those restrictions the largest value that could stored in 16 bits was 32767. So, I am only guessing, that someone at MS with a long memory picked it as the value least likely to create a problem. No facts to back up my assumption, a pure guess on my part, take it for what it is worth .... which of course could be nothing. If you are curious enough use google to look up the IBM 900 or 901 model to learn what the "good old days" of computing were.
Correct me if I'm wrong. Back in the good old days of GW-BASIC the highest line number a program could have was precisely 32767.
"El" Jerry
The restriction was the fact that 16 bits (remember 1 bit was reserved - in most systems the high order bit - numbering being from right to left - was reserved to indicate when set that the number was negative, if not set the number was of course positive. So GW-BASIC was limited by memory construction.. the chips themselves ..
Boy oh boy this is taking me down memory lane to the days long past. When Hewlett Packard computers .. first model a 2116C, no hard drive, data read in and output using punched paper tape. An IBM 1401 using a RAMAC hard drive ... drive consisted of multiple platters each about 3 foot in diameter, stacked in a vertical housing standing about 6 ft tall, read / write heads (1 set) driven from platter to platter using of all things a steering gear shaft from a Ford motor vehicle. to raise and lower the heads to the required disc platter. Language used know as auto-coder ... but it did have its bright spots .. got chewed out for a printer miss alinement of its print heads by Admiral Rickover the founder of the U.S. nuclear navy and his people developing the first nuc submarine ... The one individual I wish I had the good fortune to meet, but never did was Adm Grace Hopper the founder of COBOL, (an acronym for Common Business Orientated Language ).
Well enough of the so called Good Old Days.
June 1, 2012 at 1:01 am
Nice question, thanks.
Need an answer? No, you need a question
My blog at https://sqlkover.com.
MCSE Business Intelligence - Microsoft Data Platform MVP
June 4, 2012 at 7:52 pm
Good question. Thanks for submitting.
http://brittcluff.blogspot.com/
August 9, 2012 at 8:49 am
Nice question!
December 19, 2013 at 1:13 am
A Very good addition in the post.
January 30, 2014 at 5:36 am
Does that mean that if one replaces resource db files of 2008R2 server with 2008, SQL would be rolled back to 2008 server? Or the roll back using resource database applies only to Service Packs.
Viewing 15 posts - 16 through 29 (of 29 total)
You must be logged in to reply to this topic. Login to reply