Re: Well - I feel like an idiot.

[Sean Corfield]

On Fri, 31 Dec 2004 14:18:07 -0500, Jim Davis
<[EMAIL PROTECTED]> wrote:
> > http://livedocs.macromedia.com/coldfusion/6/Developing_ColdFusion_MX_Appli
> > cations_with_CFML/sharedVars6.htm
> >
> > http://livedocs.macromedia.com/coldfusion/6/Developing_ColdFusion_MX_Appli
> > cations_with_CFML/sharedVars8.htm#1154837
> 
> Yeah - but they're wrong.  ;^)

And this is not likely to be corrected for Blackstone I'm afraid. The
docs need a serious going over in terms of locking and that just isn't
possible in the current timeframe. I suggest you visit the LiveDocs
pages and add comments about this...
-- 
Sean A Corfield -- http://www.corfield.org/
Team Fusebox -- http://www.fusebox.org/
Breeze Me! -- http://www.corfield.org/breezeme
Got Gmail? -- I have 6 invites to give away!

"If you're not annoying somebody, you're not really alive."
-- Margaret Atwood

~|
Special thanks to the CF Community Suite Silver Sponsor - RUWebby
http://www.ruwebby.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189096
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Stress Testing & Cold Fusions Limits

[Sean Corfield]

On Fri, 31 Dec 2004 16:38:42 -0600, Tony Hicks <[EMAIL PROTECTED]> wrote:
> Where you say cold fusion on a single box really peaks out?

It completely depends on your application and the traffic patterns it sees.

> I've been having a few issues but its cfmx related... problems
> that didn't exist in 5.. And I want to know a good next move.

One of the most common issues people encounter moving from CF5 to CFMX
is database driver performance. Do you have the latest 6.1 Updater
installed?

> Anyone
> got any recommendations on a good program to... bombard your site with
> traffic?

OpenSTA, WAST (Microsoft) are free. We use Segue Silk Performer here
at Macromedia but that costs $$$.
-- 
Sean A Corfield -- http://www.corfield.org/
Team Fusebox -- http://www.fusebox.org/
Breeze Me! -- http://www.corfield.org/breezeme
Got Gmail? -- I have 6 invites to give away!

"If you're not annoying somebody, you're not really alive."
-- Margaret Atwood

~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189095
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Stress Testing & Cold Fusions Limits

[Elyse Nielsen]

Hi,

Check out grinder out and see if it helps.

http://www.anticlue.net/archives/000395.htm

The one main drawback is that you need to have multiple sessions on
multiple PCs, because it overwhelms the local pc's nic card since they
normally are lower end.

The other program that is great for load testing is load runner from
Mercury Interactive.  I believe there is a 10 day trail version.

HTH,
Elyse
http://www.anticlue.net


On Fri, 31 Dec 2004 18:38:28 -0500, Rick Mason <[EMAIL PROTECTED]> wrote:
> Tony,
> 
> Finding out where an application "peaks out" is a complex question and
> depends on your code and the database etc.
> 
> Charlie Arehart spoke before our user group a few years back on the
> subject of load testing.  I also believe he wrote an article on the
> subject in CFDJ though I couldn't locate it with a quick search.
> 
> I've actually used one of the tools he spoke of that night by a
> company called OpenDemand and I believe they have a free web trial.
> There is also a free load tester from Microsoft and another one from a
> German company.
> 
> If you can locate either that article or Charlie's powepoint it should
> answer most of the questions you might have.  I know that I learned an
> awful lot about my application after I did some basic load testing.
> 
> Rick Mason
> 
> 
> On Fri, 31 Dec 2004 16:38:42 -0600, Tony Hicks <[EMAIL PROTECTED]> wrote:
> > You guys have some impressive knowledge.. I think I do too :) but I
> > admire yours..  I wanted to ask a question...
> >
> > Where you say cold fusion on a single box really peaks out? I mean
> > without clustering and everything... where would you say its too much
> > load? I've been having a few issues but its cfmx related... problems
> > that didn't exist in 5.. And I want to know a good next move.
> >
> > Also.. I want to toss traffic at my own site and let my error reporter
> > run and see how far I can crank it up.. good weekend project.. Anyone
> > got any recommendations on a good program to... bombard your site with
> > traffic?
> >
> >
> 
> 

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189094
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Stress Testing & Cold Fusions Limits

[Rick Mason]

Tony,

Finding out where an application "peaks out" is a complex question and
depends on your code and the database etc.

Charlie Arehart spoke before our user group a few years back on the
subject of load testing.  I also believe he wrote an article on the
subject in CFDJ though I couldn't locate it with a quick search.

I've actually used one of the tools he spoke of that night by a
company called OpenDemand and I believe they have a free web trial. 
There is also a free load tester from Microsoft and another one from a
German company.

If you can locate either that article or Charlie's powepoint it should
answer most of the questions you might have.  I know that I learned an
awful lot about my application after I did some basic load testing.


Rick Mason



On Fri, 31 Dec 2004 16:38:42 -0600, Tony Hicks <[EMAIL PROTECTED]> wrote:
> You guys have some impressive knowledge.. I think I do too :) but I
> admire yours..  I wanted to ask a question...
> 
> Where you say cold fusion on a single box really peaks out? I mean
> without clustering and everything... where would you say its too much
> load? I've been having a few issues but its cfmx related... problems
> that didn't exist in 5.. And I want to know a good next move.
> 
> Also.. I want to toss traffic at my own site and let my error reporter
> run and see how far I can crank it up.. good weekend project.. Anyone
> got any recommendations on a good program to... bombard your site with
> traffic?
> 
> 

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189093
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Jim Davis]

> -Original Message-
> From: Ben Rogers [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 31, 2004 5:43 PM
> To: CF-Talk
> Subject: RE: Well - I feel like an idiot.
> 
> > Which is exactly why you have to be consistent with locking.  ;^)
> 
> I don't see how consistency directly relates to what I was talking about.
> Modular code generally knows very little about its context: that's the
> whole
> point. You can be entirely consistent and still end up with deadlocks if
> one
> piece of code isn't aware of what's going on around it.

True - but that mean you shouldn't be consistent with your code.  At the
very least any extensions that you or your team's written would be unable to
cause such deadlocks.
 
> > Generally nesting concerns would only happen with scope level locking or
> > site-wide name-level locks.
> 
> The names used in "name level locks" are completely arbitrary. What
> happens
> when you call a third party CFC that just happens to use the same name
> that
> you use? Better yet, what happens when your cfc, which locks the session
> scope, accesses a singleton registry which locks the application scope?

Although name-level locking is arbitrary it's not at all random.  Steps can
be taken to minimize name conflicts (and if there are name conflicts you're
most likely going to have more issues than deadlocks).

For example in my persistent CFCs I use UUIDs as the lock names.
Furthermore in CFMX I don't personally use scope locking at all - that
eliminates most of the causes of contention that we're talking about.

As for a session lock being in place at the same time as an application lock
- so what?

You can only get a deadlock in CF if those locks are nested in BOTH places.
The CFC in question would have to lock the application then session (or vice
versa) and the singleton would have to do the reverse.

As far as I can see that's the only way to get a deadlock in that case.
 
> > In those cases, despite the encapsulation of
> > extensions you'll need to know something about the application
> > architecture to create them in first place (I think).
> 
> You're assuming you have access to and an intimate knowledge of all the
> code
> in question. I used a third party CFC in a recent project. It's encrypted,
> so I have no idea what locks it might create, what scope they are or what
> it
> names them, etc. We used the Component Kit in another project. That stores
> quite a bit of information in the server scope of all places. How did that
> information get there?

Then those are issues with your application.  Personally I would ask the
developer of those third party tools to document the locking strategy of the
extension.

For what it's worth I don't think that any CFMX extension should EVER scope
lock exactly for the reasons that you mention.  Only name locks should be
used and then steps should be taken to prevent name contention.

This isn't too dissimilar to the issues in the Java world with distributing
packages.  The whole "com.mysite.packagename" convention was promoted to
prevent this kind of name contention.

I think the same kind of methodology could be adapted to prevent lock name
contention as well.

> > True - but that would be a failure of design, not of locking.  If your
> > template is doing something outside of locks that would affect code
> inside
> > of locks like that I would think you'd need to reevaluate how you're
> doing
> > things.
> 
> All failures are a failure of design. :) The point I was trying to make is
> that you have to know a good deal about locking and how your code will be

To true.  I stand corrected.  ;^)

> This is clearly something we disagree on. I'm sure I'm biased because I've
> seen far more deadlock issues than race conditions.

I might just be the kind of code we see.  I tend to see a lot of racing due
to bad UI design (don't get me started on THAT however).  Framesets, popup
windows, poor design leading to multiple windows, etc.

A lot of cases are just potential race conditions - the clichéd example is
the single shopping cart linked to multiple windows.  I don't honestly think
I've ever seen a real problem related to that, but that's the example
everybody quotes.  ;^)
 
> If the novice is locking everything, then deadlocks are sure to ensue.
> ColdFusion does nothing to prevent the loss of data integrity in a
> deadlock
> situation.

I'm not sure about that (which means, of course, that I could be wrong).
Again - you need nested locks and an inconsistent nesting in multiple
templates.

I rarely see nested locks at all, much less in novice code.  Since CFMX was
released I can't remember a case where I've ever needed to nest locks.

In previous versions the classic issues was always the
application-then-session nesting.  This nearly ALWAYS showed up due to bad
code like this:








Of course even in earlier versions this is just plain bad code.

But in MX I just don't see nested locks much at 

Re: Well - I feel like an idiot.

[Tony Hicks]

I've never had a session write or a session read fail on me, because
thank God, right after I learned what a session was, someone sent me
to research CFLOCK.

But I've seen them fail on other cf sites... a session read actually..
no data was being written at the time... I was browsing a forum one
day and I don't know what happened but I posted a response to someone
and I was posting as a different user... A user who I'd never met so I
had no access to his account.

I don't know what happened.. it probably wasn't strictly a read...
something invisible in the site.. like a lastAccessed variable was
probably being set, and whadya know, I come out as a different user. I
had to email the guy and tell him to change his password because if it
happened to me, it could happen to others and while I didn't look up
his password, someone else could have.

That is (or at least was) the apparent flaw to cf's sessions.. its not
one field that gets switched... it was the whole session.

On Fri, 31 Dec 2004 17:53:15 -0500, Jim Davis
<[EMAIL PROTECTED]> wrote:
> > -Original Message-
> > From: Tony Hicks [mailto:[EMAIL PROTECTED]
> > Sent: Friday, December 31, 2004 5:35 PM
> > To: CF-Talk
> > Subject: Re: Well - I feel like an idiot.
> >
> > Many session-based sites run well without any cflocks but they're
> > smaller sites.. It doesn't seem to me to be a policy of knowing what
> > your doing or not, its more about the fact that you don't ever want
> > data mixed up, ever.
> >
> > To me its worth a few more lines of code to always protect my user's
> > data. I can see where locking or not locking application variables
> > could be harmless but I stand firm on sessions needing to be locked.
> 
> As I've been saying (in my opinion) if you're unsure, lock.
> 
> However if you are sure that not locking will not cause any issues, then its
> only better performance to drop needless code.
> 
> Most people would agree that many (if not all) session reads (for example)
> don't need to be locked (depending on the application of course).
> 
> For example consider the common practice of placing a "LastAccessed"
> timestamp in your session scope.  This variable is updated with the current
> when a user makes a request.
> 
> If two templates attempt to update the value at the same time it doesn't
> matter (at least to my application): the time is only used to determine if a
> session is "stale" so either time is fine as they're both essentially the
> same.
> 
> Personally I wouldn't lock that write in CFMX (in previous versions I
> would).
> 
> To me it's always about the current situation.  I agree in general that
> session data generally needs to be locked, but each situation presents
> different circumstances.
> 
> Jim Davis
> 
> 

~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189091
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Jim Davis]

> -Original Message-
> From: Tony Hicks [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 31, 2004 5:35 PM
> To: CF-Talk
> Subject: Re: Well - I feel like an idiot.
> 
> Many session-based sites run well without any cflocks but they're
> smaller sites.. It doesn't seem to me to be a policy of knowing what
> your doing or not, its more about the fact that you don't ever want
> data mixed up, ever.
> 
> To me its worth a few more lines of code to always protect my user's
> data. I can see where locking or not locking application variables
> could be harmless but I stand firm on sessions needing to be locked.

As I've been saying (in my opinion) if you're unsure, lock.

However if you are sure that not locking will not cause any issues, then its
only better performance to drop needless code.

Most people would agree that many (if not all) session reads (for example)
don't need to be locked (depending on the application of course).

For example consider the common practice of placing a "LastAccessed"
timestamp in your session scope.  This variable is updated with the current
when a user makes a request.

If two templates attempt to update the value at the same time it doesn't
matter (at least to my application): the time is only used to determine if a
session is "stale" so either time is fine as they're both essentially the
same.

Personally I wouldn't lock that write in CFMX (in previous versions I
would).

To me it's always about the current situation.  I agree in general that
session data generally needs to be locked, but each situation presents
different circumstances.

Jim Davis



~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189090
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Ben Rogers]

> Many session-based sites run well without any cflocks but they're
> smaller sites.. It doesn't seem to me to be a policy of knowing what
> your doing or not, its more about the fact that you don't ever want
> data mixed up, ever.

Which can happen as the result of deadlocks.

> To me its worth a few more lines of code to always protect my user's
> data.

I don't believe Jim or I are arguing against locking or that locking causes
code to become too verbose. We simply disagree over whether or not people
who do not understand locking should lock every block of code that is stored
in a shared scope.

> I can see where locking or not locking application variables
> could be harmless but I stand firm on sessions needing to be locked.

It depends more on the data and its use then the actual ColdFusion scope
involved. Some applications store data pertaining to user sessions in the
server scope. Personally, I'm not a big fan of this, but I'm sure others
have their reasons.

Ben Rogers
http://www.c4.net
v.508.240.0051
f.508.240.0057


~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189089
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Ben Rogers]

> Which is exactly why you have to be consistent with locking.  ;^)

I don't see how consistency directly relates to what I was talking about.
Modular code generally knows very little about its context: that's the whole
point. You can be entirely consistent and still end up with deadlocks if one
piece of code isn't aware of what's going on around it.

> Generally nesting concerns would only happen with scope level locking or
> site-wide name-level locks.

The names used in "name level locks" are completely arbitrary. What happens
when you call a third party CFC that just happens to use the same name that
you use? Better yet, what happens when your cfc, which locks the session
scope, accesses a singleton registry which locks the application scope? 

> In those cases, despite the encapsulation of
> extensions you'll need to know something about the application
> architecture to create them in first place (I think).

You're assuming you have access to and an intimate knowledge of all the code
in question. I used a third party CFC in a recent project. It's encrypted,
so I have no idea what locks it might create, what scope they are or what it
names them, etc. We used the Component Kit in another project. That stores
quite a bit of information in the server scope of all places. How did that
information get there?

> True - but that would be a failure of design, not of locking.  If your
> template is doing something outside of locks that would affect code inside
> of locks like that I would think you'd need to reevaluate how you're doing
> things.

All failures are a failure of design. :) The point I was trying to make is
that you have to know a good deal about locking and how your code will be
used to lock properly. Otherwise, you end up deadlocks instead of race
conditions. The end result is the same: inconsistent data. Consequently, I
don't believe that the "lock everything" paradigm is any better than not
locking at all.

> That depends - but I agree in theory.  I still stand by my belief that
> locking is safer than unlocking if you don't know what you're doing.

This is clearly something we disagree on. I'm sure I'm biased because I've
seen far more deadlock issues than race conditions.

> In most cases data integrity is more important than experience and since
> nested locks are the only way to create a deadlock in CF it seems to me
> rare that a novice would encounter the issue.

If the novice is locking everything, then deadlocks are sure to ensue.
ColdFusion does nothing to prevent the loss of data integrity in a deadlock
situation.

> Of course I think we both can agree that the preferred thing is to
> actually gain an understanding of threaded requests and locking before
> writing a CF application.  ;^)

Yes, but short of that, I would not recommend "locking everything". I would
recommend the opposite: do not lock unless you have a reason to. 

> In practice I actually write much more like you're talking about: very few
> locks.  Since I know (or at least think I know) what's what I feel
> confident in choosing when to an when not to lock.

And you've probably had very few deadlock issues. I do the same. I generally
have a handful of well named locks in any application. I've had very few
deadlock issues and, in almost a decade of Web programming, I can probably
count the number of race conditions I've seen on one hand.

> Then I get bitten in the ass by something the code that I started the
> thread with.  ;^)  (I know that wasn't really a locking issue per se, but 
> it's definitely in the ballpark.)

I know what you mean. It's always when I'm most sure of myself that I'm the
furthest from being right. :)

I actually had to write my own locking routines in ASP a couple of years
ago. ASP lacks anything analogous to the cflock tag. To make a long story
short, I discovered the hard way that locking was a far more intricate and
delicate process than I had thought. I came back to ColdFusion with a new
found respect for locking issues.

Ben Rogers
http://www.c4.net
v.508.240.0051
f.508.240.0057


~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189088
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Stress Testing & Cold Fusions Limits

[Tony Hicks]

You guys have some impressive knowledge.. I think I do too :) but I
admire yours..  I wanted to ask a question...

Where you say cold fusion on a single box really peaks out? I mean
without clustering and everything... where would you say its too much
load? I've been having a few issues but its cfmx related... problems
that didn't exist in 5.. And I want to know a good next move.

Also.. I want to toss traffic at my own site and let my error reporter
run and see how far I can crank it up.. good weekend project.. Anyone
got any recommendations on a good program to... bombard your site with
traffic?

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189087
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Well - I feel like an idiot.

[Tony Hicks]

Many session-based sites run well without any cflocks but they're
smaller sites.. It doesn't seem to me to be a policy of knowing what
your doing or not, its more about the fact that you don't ever want
data mixed up, ever.

To me its worth a few more lines of code to always protect my user's
data. I can see where locking or not locking application variables
could be harmless but I stand firm on sessions needing to be locked.


On Fri, 31 Dec 2004 16:52:00 -0500, Jim Davis
<[EMAIL PROTECTED]> wrote:
> > -Original Message-
> > From: Ben Rogers [mailto:[EMAIL PROTECTED]
> > Sent: Friday, December 31, 2004 4:28 PM
> > To: CF-Talk
> > Subject: RE: Well - I feel like an idiot.
> >
> > But what happens if you call a UDF, CFC method or custom tag within a
> > lock?
> > Each of these provides a degree of encapsulation, meaning you don't know
> > what they might be doing.
> 
> Which is exactly why you have to be consistent with locking.  ;^)
> 
> Generally nesting concerns would only happen with scope level locking or
> site-wide name-level locks.  In those cases, despite the encapsulation of
> extensions you'll need to know something about the application architecture
> to create them in first place (I think).
> 
> > > In other words I don't believe CF has the concept of deadlock victims,
> > but
> > > I could be wrong.  As far as I can tell when a lock request times out,
> > > it's dead - it either skips the locked code (being unable to obtain the
> > > lock) or ends the template with an error (depending on how you set up
> > the
> > > lock request).
> >
> > I used the term "deadlock victims" in the first paragraph to mean the
> > templates were victims of a deadlock situation, not that ColdFusion had
> > determined that they were deadlocked victims and treated them as such.
> > Sorry for the confusion.
> 
> No problem - this is just a topic I'm very interested in.
> 
> > > I'm way too tired to consider this clearly, but I'm not sure I see the
> > > potential for a deadlock where "some" locked code is run and would need
> > to
> > > be "rolled back" (I'm not sure how you would roll back arbitrary code
> > even
> > > then).
> >
> > The issue with race conditions is that you end up with data that's in an
> > inconsistent state. If you have two templates that get terminated because
> > of
> > a deadlock situation, both templates could have updated some information,
> > but not all. In other words, the end result of a race condition and a
> > deadlock is the same: inconsistent data.
> 
> True - but that would be a failure of design, not of locking.  If your
> template is doing something outside of locks that would affect code inside
> of locks like that I would think you'd need to reevaluate how you're doing
> things.
> 
> > Instead of a race condition (which can often be harmless), you caused a
> > deadlock. Neither process finished. Some data reflects the state before
> > each
> > template ran. Some data, after. With a deadlock, the user gets an error in
> > the process.
> 
> That depends - but I agree in theory.  I still stand by my belief that
> locking is safer than unlocking if you don't know what you're doing.
> 
> In most cases data integrity is more important than experience and since
> nested locks are the only way to create a deadlock in CF it seems to me rare
> that a novice would encounter the issue.
> 
> Of course I think we both can agree that the preferred thing is to actually
> gain an understanding of threaded requests and locking before writing a CF
> application.  ;^)
> 
> In practice I actually write much more like you're talking about: very few
> locks.  Since I know (or at least think I know) what's what I feel confident
> in choosing when to an when not to lock.
> 
> Then I get bitten in the ass by something the code that I started the thread
> with.  ;^)  (I know that wasn't really a locking issue per se, but it's
> definitely in the ballpark.)
> 
> Jim Davis
> 
> 

~|
Special thanks to the CF Community Suite Silver Sponsor - RUWebby
http://www.ruwebby.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189086
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Array issue

[Tony Hicks]

I didn't have a part in the suggestion but I'm glad it works. I saw it
and being a list fan like I am.. I would have suggested it too...

I would suggest one other thing though... your summary seems like to
goes beyond too much... just a thought..

I'd put pages relating to each module in a seperate folder.. I'd name
them m1, m2, m3m19 etc...

The code at the top of each page, or even in application.cfm would
look like this...


 (or whatever means admin is on)
  

  
  

  


And that would parse out the folder name that contained the level. You
check for mgranted to equal 1.

Now code to configure links to custom controls...


  
  
  
  
  
  
  
  
  



Now the code to get the right links for the right users.


  
· 
  #AppControls[a1][a2]# ·

  


Here's another way to do the above... Its a better method I'd say...
Just don't know if it will work.. never tried it.


  
#ArrayToList(AppControls[a1]," | ")#
  


~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189085
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Jim Davis]

> -Original Message-
> From: Ben Rogers [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 31, 2004 4:28 PM
> To: CF-Talk
> Subject: RE: Well - I feel like an idiot.
> 
> But what happens if you call a UDF, CFC method or custom tag within a
> lock?
> Each of these provides a degree of encapsulation, meaning you don't know
> what they might be doing.

Which is exactly why you have to be consistent with locking.  ;^)

Generally nesting concerns would only happen with scope level locking or
site-wide name-level locks.  In those cases, despite the encapsulation of
extensions you'll need to know something about the application architecture
to create them in first place (I think).

> > In other words I don't believe CF has the concept of deadlock victims,
> but
> > I could be wrong.  As far as I can tell when a lock request times out,
> > it's dead - it either skips the locked code (being unable to obtain the
> > lock) or ends the template with an error (depending on how you set up
> the
> > lock request).
> 
> I used the term "deadlock victims" in the first paragraph to mean the
> templates were victims of a deadlock situation, not that ColdFusion had
> determined that they were deadlocked victims and treated them as such.
> Sorry for the confusion.

No problem - this is just a topic I'm very interested in.
 
> > I'm way too tired to consider this clearly, but I'm not sure I see the
> > potential for a deadlock where "some" locked code is run and would need
> to
> > be "rolled back" (I'm not sure how you would roll back arbitrary code
> even
> > then).
> 
> The issue with race conditions is that you end up with data that's in an
> inconsistent state. If you have two templates that get terminated because
> of
> a deadlock situation, both templates could have updated some information,
> but not all. In other words, the end result of a race condition and a
> deadlock is the same: inconsistent data.

True - but that would be a failure of design, not of locking.  If your
template is doing something outside of locks that would affect code inside
of locks like that I would think you'd need to reevaluate how you're doing
things.

> Instead of a race condition (which can often be harmless), you caused a
> deadlock. Neither process finished. Some data reflects the state before
> each
> template ran. Some data, after. With a deadlock, the user gets an error in
> the process.

That depends - but I agree in theory.  I still stand by my belief that
locking is safer than unlocking if you don't know what you're doing.

In most cases data integrity is more important than experience and since
nested locks are the only way to create a deadlock in CF it seems to me rare
that a novice would encounter the issue.

Of course I think we both can agree that the preferred thing is to actually
gain an understanding of threaded requests and locking before writing a CF
application.  ;^)

In practice I actually write much more like you're talking about: very few
locks.  Since I know (or at least think I know) what's what I feel confident
in choosing when to an when not to lock.

Then I get bitten in the ass by something the code that I started the thread
with.  ;^)  (I know that wasn't really a locking issue per se, but it's
definitely in the ballpark.)

Jim Davis




~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189084
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Array issue

[Michael C.Grove]

Thanks
As always you guys came through.
The answer to my problem was 

I don't know much about lists or arrays, so I really don't know the
available functions.

As for why I am using an array I found a piece of code in a book that seem
to resolve an issue.
When a person logs into my site The system check one table to see if they
have a correct id and password, checks their status and their security
level.

If all is good and the users level is admin then they have access to
everything. If they are just a user, then the system runs an additional
query to see what system modules the user has access to. This is in a
separate table and simply looks for the user id and module id.

The modules that the user has access to are placed in an array and stored in
a session variable.

Depending on the page that the user is on and the section of the system they
may or may not see various links based on the module access.
I also place an if statement at the beginning of each individual module
checking whether or not they have access.

-Original Message-
From: CF-Talk [mailto:[EMAIL PROTECTED] 
Sent: Friday, December 31, 2004 2:01 PM
To: CF-Talk
Subject: CF-Talk: Digest every 8 hours

CF-Talk  31-Dec-04 Issue:1444
In this issue:
 Array issue
 Array issue
 Array issue
 Coldfusion & IIS behind firewall
 Extract year from date field
 Extract year from date field
 Extract year from date field
 Extract year from date field
 Extract year from date field
 Extract year from date field
 Extract year from date field
 Extract year from date field
 Extract year from date field
 Extract year from date field
 OT: query optimization
 OT: query optimization
 OT: query optimization
 Possible to pass formatting to Excel from CF?
 Possible to pass formatting to Excel from CF?
 QofQ: SELECT DISTINCT
 QofQ: SELECT DISTINCT
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.



~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189083
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Ben Rogers]

> I'm not sure how this relates in CF...
> 
> The only real way that I know of to get a real deadlock in CF is to nest
> lock requests.  Like the following:
> 
> 
>   
>   
> 

As far as I know, it's only possible via nested locks. There's actually a
good deal on deadlocks in the ColdFusion documentation:

  http://livedocs.macromedia.com/coldfusion/6.1/htmldocs/shared52.htm

  http://livedocs.macromedia.com/coldfusion/6.1/htmldocs/shared52.htm

> And even then is only a real problem if you are inconsistent with the
> nesting order of the lock requests.

But what happens if you call a UDF, CFC method or custom tag within a lock?
Each of these provides a degree of encapsulation, meaning you don't know
what they might be doing.

> I don't think CF does anything
> special about this however - the templates just timeout normally unable
> to finish their work.

I think you're correct. In fact, I was trying to argue that the problem is
that ColdFusion does not recognize deadlocks like an RDBMS would, for
instance. ColdFusion can't detect a deadlock situation, choose a deadlock
victim, and rollback the changes that that script has made.

> In other words I don't believe CF has the concept of deadlock victims, but
> I could be wrong.  As far as I can tell when a lock request times out, 
> it's dead - it either skips the locked code (being unable to obtain the 
> lock) or ends the template with an error (depending on how you set up the 
> lock request).

I used the term "deadlock victims" in the first paragraph to mean the
templates were victims of a deadlock situation, not that ColdFusion had
determined that they were deadlocked victims and treated them as such. Sorry
for the confusion.

> I'm way too tired to consider this clearly, but I'm not sure I see the
> potential for a deadlock where "some" locked code is run and would need to
> be "rolled back" (I'm not sure how you would roll back arbitrary code even
> then).

The issue with race conditions is that you end up with data that's in an
inconsistent state. If you have two templates that get terminated because of
a deadlock situation, both templates could have updated some information,
but not all. In other words, the end result of a race condition and a
deadlock is the same: inconsistent data.

Instead of a race condition (which can often be harmless), you caused a
deadlock. Neither process finished. Some data reflects the state before each
template ran. Some data, after. With a deadlock, the user gets an error in
the process.

Ben Rogers
http://www.c4.net
v.508.240.0051
f.508.240.0057


~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189082
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Jim Davis]

> -Original Message-
> From: Ben Rogers [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 31, 2004 3:18 PM
> To: CF-Talk
> Subject: RE: Well - I feel like an idiot.
> 
> > But the core advice of "always lock" is definitely the safer option.
> 
> I don't think I agree with the "always lock" mentality. This inevitably
> leads to deadlocks, and I don't believe ColdFusion handles deadlocks in a
> graceful way. By that I mean that ColdFusion does not roll back all
> operations in a lock (as if it were a transaction) when killing off
> deadlock
> victims.

I'm not sure how this relates in CF...

The only real way that I know of to get a real deadlock in CF is to nest
lock requests.  Like the following:






And even then is only a real problem if you are inconsistent with the
nesting order of the lock requests.  I don't think CF does anything special
about this however - the templates just timeout normally unable to finish
their work.

In other words I don't believe CF has the concept of deadlock victims, but I
could be wrong.  As far as I can tell when a lock request times out, it's
dead - it either skips the locked code (being unable to obtain the lock) or
ends the template with an error (depending on how you set up the lock
request).

I'm way too tired to consider this clearly, but I'm not sure I see the
potential for a deadlock where "some" locked code is run and would need to
be "rolled back" (I'm not sure how you would roll back arbitrary code even
then).

Jim Davis



~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189081
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Ben Rogers]

> But the core advice of "always lock" is definitely the safer option.

I don't think I agree with the "always lock" mentality. This inevitably
leads to deadlocks, and I don't believe ColdFusion handles deadlocks in a
graceful way. By that I mean that ColdFusion does not roll back all
operations in a lock (as if it were a transaction) when killing off deadlock
victims.

Consequently, you still end up with inconsistent or corrupted data, and the
end user sees hard error messages. So, unless you fully understand locking
and its side effects, I would not recommend that you lock your code. Even
understanding all this, I would recommend locking judiciously.

Ben Rogers
http://www.c4.net
v.508.240.0051
f.508.240.0057


~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189080
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: OT: query optimization

[Paul Malan]

>(Which join type is hidden behind the popup?)

Sorry I left that covered up this time...  Look here:
http://www.redmondinc.com/sos/exe_plan.gif
Is that sufficient?

>Can you give me the recordcount of the following queries:
 > SELECT DISTINCT proj_code, cust_job_num, cust_code, order_code
> FROM ordr;
 -- 712 rows
 
> SELECT DISTINCT tkt_code, tkt_date, truck_code, po,
> truck_net_wgt, hler_code, order_code
> FROM tick;
 -- 4724 rows

> SELECT DISTINCT prod_descr, order_code
> FROM ordl;
 -- 705 rows

That's probably what you were asking for in your first message; sorry
I misunderstood...

thanks!

-paul

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189079
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Extract year from date field

[Robert Orlini]

Thanks

-Original Message-
From: Joe Rinehart [mailto:[EMAIL PROTECTED]
Sent: Friday, December 31, 2004 1:51 PM
To: CF-Talk
Subject: Re: Extract year from date field


> How do I extract the year from a date field such as 2005-10-12?

> 
> select distinct(datepart(m,report_date)) as month from report
> where datepart(yy,report_date) = '2005'
> 

Not to be a wiseass...but if that's the query, it'll always be 2005 :)

It looks like your query gets all of the months of reports from 2005. 
If you wanted to get all of the distinct years for all reports, you
could do the following:


select distinct(datepart(yy,report_date)) as reportYear from report


#reportYear# 

-Joe

-- 
For Tabs, Trees, and more, use the jComponents:
http://clearsoftware.net/client/jComponents.cfm



~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189078
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Jim Davis]

> -Original Message-
> From: Tony Hicks [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 31, 2004 2:03 PM
> To: CF-Talk
> Subject: Re: Well - I feel like an idiot.
> 
> Either I like to stick my foot in my mouth, or I'm feeling gutsy but I
> was looking at livedocs for application variables on another thing and
> it did say for data integrity you SHOULD lock.
> 
> http://livedocs.macromedia.com/coldfusion/6/Developing_ColdFusion_MX_Appli
> cations_with_CFML/sharedVars6.htm
> 
> http://livedocs.macromedia.com/coldfusion/6/Developing_ColdFusion_MX_Appli
> cations_with_CFML/sharedVars8.htm#1154837

Yeah - but they're wrong.  ;^)

As I said - this is one of the biggest, most confused topics in CF.

The way I like to say it is that if you don't understand locking then ALWAYS
lock.  You'll lose some performance, but won't corrupt your app.

In the Pre-MX days failure to lock could mean server death however - the
internal structures used to store memory resident variables were not,
themselves, thread safe.  Failure to lock would often cause server memory
corruption.

This could even happen (because of those internal structures) that a simple
read could happen before a write was finished leaving the read with
corrupted data.

In MX the internal structures are thread safe themselves.  However that
doesn't mean you can't screw things up.  The biggest potential problems is
"race conditions" - conditions were two requests accessing the same
variables would cause confusion or incorrect data.

To prevent race conditions you lock your code - and if you don't know if
you'll have a race condition then you should probably lock to be safe.

Also you don't have to worry as much about reads - the read will always work
(even if it doesn't give you the data that you think it should).

In MX not locking will just cause application-level problems, not
server-level problems.  If you understand the issues at play you can safely
optimize your code by not locking in many cases.

But the core advice of "always lock" is definitely the safer option.

Jim Davis



~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189077
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Well - I feel like an idiot.

[Tony Hicks]

Either I like to stick my foot in my mouth, or I'm feeling gutsy but I
was looking at livedocs for application variables on another thing and
it did say for data integrity you SHOULD lock.

http://livedocs.macromedia.com/coldfusion/6/Developing_ColdFusion_MX_Applications_with_CFML/sharedVars6.htm

http://livedocs.macromedia.com/coldfusion/6/Developing_ColdFusion_MX_Applications_with_CFML/sharedVars8.htm#1154837

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189076
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Extract year from date field

[Joe Rinehart]

> How do I extract the year from a date field such as 2005-10-12?

> 
> select distinct(datepart(m,report_date)) as month from report
> where datepart(yy,report_date) = '2005'
> 

Not to be a wiseass...but if that's the query, it'll always be 2005 :)

It looks like your query gets all of the months of reports from 2005. 
If you wanted to get all of the distinct years for all reports, you
could do the following:


select distinct(datepart(yy,report_date)) as reportYear from report


#reportYear# 

-Joe

-- 
For Tabs, Trees, and more, use the jComponents:
http://clearsoftware.net/client/jComponents.cfm

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189075
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Array issue

[Michael T. Tangorre]

 

> From: Dawson, Michael [mailto:[EMAIL PROTECTED] 
> You need to use the ListFind() function.  CONTAINS, or ListContains()
> can get you in trouble.  ;-)
> 
> You might also check CFLib.org for an ArrayFind() function so 
> you don't
> have to build it yourself.

I would also ask why you are making your list an array? Yes, string
manipulation in CF is slow, but how involved is your usage of the string of
modules? Do you really need the array?

Mike


~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189074
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Extract year from date field

[Robert Orlini]

Thank you Sean. How can I apply the DISTINCT to the year as well?

It gives me an error when I add it in front of datepart(yy,report_date)

-Original Message-
From: Sean Corfield [mailto:[EMAIL PROTECTED]
Sent: Friday, December 31, 2004 12:49 PM
To: CF-Talk
Subject: Re: Extract year from date field


On Fri, 31 Dec 2004 10:43:43 -0500, Robert Orlini <[EMAIL PROTECTED]> wrote:
> How do I extract the year from a date field such as 2005-10-12?
> 
> 
> select distinct(datepart(m,report_date)) as month from report
> where datepart(yy,report_date) = '2005'
> 

You need to change your query so returns years as well as months -
otherwise there is no year to extract later on:


  select distinct(datepart(m,report_date)) as month,
 datepart(yy,report_date) as year from report
  where datepart(yy,report_date) = '2005'


Then you can refer to populatedropdown.year.

Mind you, if your query really does just search by year like that, you
*know* what the year is since you specified it in the query!
-- 
Sean A Corfield -- http://www.corfield.org/
Team Fusebox -- http://www.fusebox.org/
Breeze Me! -- http://www.corfield.org/breezeme
Got Gmail? -- I have 6 invites to give away!

"If you're not annoying somebody, you're not really alive."
-- Margaret Atwood



~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189073
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Array issue

[Dawson, Michael]

You need to use the ListFind() function.  CONTAINS, or ListContains()
can get you in trouble.  ;-)

You might also check CFLib.org for an ArrayFind() function so you don't
have to build it yourself.

MAD 

-Original Message-
From: Michael C.Grove [mailto:[EMAIL PROTECTED] 
Sent: Friday, December 31, 2004 11:54 AM
To: CF-Talk
Subject: Array issue

I have an array that I am using in a security table.

The system does a query for a UserID in a table and returns all of the
modules that the user has access to.
The ModuleID is simply the Autonumber in an access table.
The array builds fine.

Using the following code below this is what I get and where the error
is...

CODE
Contents: #ArrayToList(session.moduleaccess)#

RETURNED
Contents: 1,7,10,14,13

CODE
TRUEFALSE

RETURNED
TRUE

The error in this is that I need to see of the user has access to module
3.
He/She does not, but since they have access to module 13 the system sees
it has having a 3.

How do I check to see if an array contains a "3"?

 

-Original Message-
From: CF-Talk [mailto:[EMAIL PROTECTED]
Sent: Friday, December 31, 2004 6:01 AM
To: CF-Talk
Subject: CF-Talk: Digest every 8 hours

CF-Talk  31-Dec-04 Issue:1443
In this issue:
 CF+XPath
 cfapplication/clientvariable Error after Upgrade to MX 6.1
 cfapplication/clientvariable Error after Upgrade to MX 6.1
 CFC encapsulation
 CFC encapsulation
 CFC encapsulation
 My Init() in my CFC...am I on the right track?
 OT: query optimization
 OT: query optimization
 OT: query optimization
 OT: query optimization
 OT: query optimization
 OT: query optimization
 QofQ: SELECT DISTINCT
 QofQ: SELECT DISTINCT
 QofQ: SELECT DISTINCT
 query optimization
 shopping cart, session variables - best practices
 shopping cart, session variables - best practices
 Upgraded CF5 to CFMX - applications not working
 Upgraded CF5 to CFMX - applications not working
 Webservice output in browser 
 Webservice output in browser
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.





~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189072
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Array issue

[Michael C.Grove]

I have an array that I am using in a security table.

The system does a query for a UserID in a table and returns all of the
modules that the user has access to.
The ModuleID is simply the Autonumber in an access table.
The array builds fine.

Using the following code below this is what I get and where the error is...

CODE
Contents: #ArrayToList(session.moduleaccess)#

RETURNED
Contents: 1,7,10,14,13

CODE
TRUEFALSE

RETURNED
TRUE

The error in this is that I need to see of the user has access to module 3.
He/She does not, but since they have access to module 13 the system sees it
has having a 3.

How do I check to see if an array contains a "3"?

 

-Original Message-
From: CF-Talk [mailto:[EMAIL PROTECTED] 
Sent: Friday, December 31, 2004 6:01 AM
To: CF-Talk
Subject: CF-Talk: Digest every 8 hours

CF-Talk  31-Dec-04 Issue:1443
In this issue:
 CF+XPath
 cfapplication/clientvariable Error after Upgrade to MX 6.1
 cfapplication/clientvariable Error after Upgrade to MX 6.1
 CFC encapsulation
 CFC encapsulation
 CFC encapsulation
 My Init() in my CFC...am I on the right track?
 OT: query optimization
 OT: query optimization
 OT: query optimization
 OT: query optimization
 OT: query optimization
 OT: query optimization
 QofQ: SELECT DISTINCT
 QofQ: SELECT DISTINCT
 QofQ: SELECT DISTINCT
 query optimization
 shopping cart, session variables - best practices
 shopping cart, session variables - best practices
 Upgraded CF5 to CFMX - applications not working
 Upgraded CF5 to CFMX - applications not working
 Webservice output in browser 
 Webservice output in browser
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.
 Well - I feel like an idiot.



~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189071
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Extract year from date field

[Sean Corfield]

On Fri, 31 Dec 2004 10:43:43 -0500, Robert Orlini <[EMAIL PROTECTED]> wrote:
> How do I extract the year from a date field such as 2005-10-12?
> 
> 
> select distinct(datepart(m,report_date)) as month from report
> where datepart(yy,report_date) = '2005'
> 

You need to change your query so returns years as well as months -
otherwise there is no year to extract later on:


  select distinct(datepart(m,report_date)) as month,
 datepart(yy,report_date) as year from report
  where datepart(yy,report_date) = '2005'


Then you can refer to populatedropdown.year.

Mind you, if your query really does just search by year like that, you
*know* what the year is since you specified it in the query!
-- 
Sean A Corfield -- http://www.corfield.org/
Team Fusebox -- http://www.fusebox.org/
Breeze Me! -- http://www.corfield.org/breezeme
Got Gmail? -- I have 6 invites to give away!

"If you're not annoying somebody, you're not really alive."
-- Margaret Atwood

~|
Special thanks to the CF Community Suite Silver Sponsor - RUWebby
http://www.ruwebby.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189070
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Well - I feel like an idiot.

[Sean Corfield]

On Fri, 31 Dec 2004 08:48:06 -0500, Jeffry Houser <[EMAIL PROTECTED]> wrote:
>   I'm just gonna throw this out there.  Would this be better?:

No, it wouldn't. You would cause every single request to the server to
be single threaded at this point:

> 
> 
> 
>  ... long init process ...
> 
>  
> 
> 
> 

Under load, this creates a bottleneck.

The double checked lock is the correct approach:

if not inited
lock
if not inited
do init
/if
/lock
/if

The most common case in terms of execution is with the app already
inited so you need to ensure that does not lock. The startup case just
needs to be thread-safe (since that's when you're modifying
application scope data).
-- 
Sean A Corfield -- http://www.corfield.org/
Team Fusebox -- http://www.fusebox.org/
Breeze Me! -- http://www.corfield.org/breezeme
Got Gmail? -- I have 6 invites to give away!

"If you're not annoying somebody, you're not really alive."
-- Margaret Atwood

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189069
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Well - I feel like an idiot.

[Tony Hicks]

I understand completely what your saying, just had a different view on
it, that's all.

Two cfifs don't hurt anything really... I just try to avoid them
whenever possible.. I mean its possible to let cf do the deciding in
some cases without using CFIF/CFSWITCH/*shudder* IIF. Wanna see?

You were clear on the application.cfm bit.

Happy New Year folks

On Fri, 31 Dec 2004 10:03:24 -0500, Jim Davis
<[EMAIL PROTECTED]> wrote:
> > -Original Message-
> > From: Tony Hicks [mailto:[EMAIL PROTECTED]
> > Sent: Friday, December 31, 2004 9:38 AM
> > To: CF-Talk
> > Subject: Re: Well - I feel like an idiot.
> >
> > Humbly, I was wrong on the locks on application variables.. thats
> > information to file for later, thanks.
> 
> No need to be humble about it.  The pre-MX locking implementation was
> perhaps the biggest gaff in the history of CF and is still one of the most
> confusing things.  Let's just all be happy that they fixed it in MX.  ;^)
> 
> > But I can't see how both CFIFs are necessary... The outer cfif will
> > likely never affect anything.
> >
> > If two people hit the site, one request behind another, they'll both
> > fire all the way up to the cflock at the same speed, and as we all
> > know, that's where the magic comes in.. cflock single threads on the
> > init process and one is done before the other.. The outer cfif will
> > not catch anything that the inner cfif does not.
> 
> It will catch everybody AFTER this process is complete.
> 
> You're absolutely right about the behavior while the initialization is
> running - but after it's done the outer CFIF means that all visitors simply
> check the status and move on rather than obtain a lock and THEN check the
> status.
> 
> I may not have been clear, but this code is in Application.CFM - which means
> its run by every request.  Because it's so central I want this code to run
> as fast as possible.
> 
> > I do wonder if it might be smarter to set application.init to
> > SOMETHING before the init process inside the cflock... but I doubt
> > that would change anything either.
> 
> I considered it, but I didn't see any benefit to the init process.  That
> process, itself, doesn't have to be as optimized (it's only run once...
> now).  It's getting past it when it has been run that needs to be optimized.
> 
> > This is funny, we're arguing tiny fractions of time here.
> 
> Tiny fractions of fractions to be sure!  But isn't that what optimization is
> all about.  ;^)
> 
> This code, as an aside, is on a site that gets lots of traffic in short (two
> day) burst.  In case the little optimizations do actually pay off.
> 
> Jim Davis
> 
> 

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189068
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Possible to pass formatting to Excel from CF?

[Elyse Nielsen]

Hi,

Dave Ross did a great job explaining how to exploit POI which
basically allows you to create your own excel sheet.

Check out

http://www.d-ross.org/index.cfm?objectid=9C65ECEC-508B-E116-6F8A9F878188D7CA

HTH,
Anticlue
http://www.anticlue.net


On Fri, 31 Dec 2004 11:37:28 -0400, Dale Thelen <[EMAIL PROTECTED]> wrote:
> I have a conference registration page that stores data in a semi-colon 
> delimited .txt file.  I have a .cfm file that lets the user dump this out 
> into Excel form using cfcontent and cfheader and cfoutput query.  It opens in 
> Excel (not Excel inside Explorer) and prompts the user to open or save as 
> attendees.xls.  It works very slick, and I am not having any problems running 
> it.
> 
> I would like to be able to pass Excel formatting to the new Excel file such 
> as 'Freeze Panes', 'Sort', etcbasically trying to make it as dummy-proof 
> as possible.
> 
> Is this possible?  Here's my code template:
> 
> 
> 
> Attendees
> 
> 
> BODY {background-image:  url(../../images/picture.gif);
> background-attachment:fixed}
> 
> 
> 
> 
> 
> 
> 
>url="http://..filename.txt";
>   name="CurrentRegistered"
>   delimiter=";"
>   textqualifier="">
> 
> 
>   
>   
>  header_1
>  header_2
>  .
>  .
>  .
>  header_n
>   
>   
>   
>  
>  
> #data_1#
> #data_2#
> .
> .
> .
> #data_n#
>  
>  
>   
> 
> 
> 
> 
> I can affect the background color and cell borders inside the  
> settings, but is there a way to pass more dynamic features? (ie: freezing the 
> header row)
> 
> 

~|
Special thanks to the CF Community Suite Silver Sponsor - RUWebby
http://www.ruwebby.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189067
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Possible to pass formatting to Excel from CF?

[Dale Thelen]

I have a conference registration page that stores data in a semi-colon 
delimited .txt file.  I have a .cfm file that lets the user dump this out into 
Excel form using cfcontent and cfheader and cfoutput query.  It opens in Excel 
(not Excel inside Explorer) and prompts the user to open or save as 
attendees.xls.  It works very slick, and I am not having any problems running 
it.

I would like to be able to pass Excel formatting to the new Excel file such as 
'Freeze Panes', 'Sort', etcbasically trying to make it as dummy-proof as 
possible.

Is this possible?  Here's my code template:



Attendees


BODY {background-image:  url(../../images/picture.gif);
background-attachment:fixed}







http://..filename.txt";
   name="CurrentRegistered"
   delimiter=";"
   textqualifier="">


   
   
  header_1
  header_2
  .
  .
  .
  header_n
   
   
   
  
  
 #data_1#
 #data_2#
 .
 .
 .
 #data_n#
  
  
   




I can affect the background color and cell borders inside the  settings, 
but is there a way to pass more dynamic features? (ie: freezing the header row)

~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189066
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Extract year from date field

[Pascal Peters]

In cf?? Year(date)

Pascal

> -Original Message-
> From: Robert Orlini [mailto:[EMAIL PROTECTED]
> Sent: 31 December 2004 16:57
> To: CF-Talk
> Subject: RE: Extract year from date field
> 
> Thanks Jochem. How would I extract the year from 2005-10-12 please?
> 
> -Original Message-
> From: Jochem van Dieten [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 31, 2004 10:47 AM
> To: CF-Talk
> Subject: Re: Extract year from date field
> 
> 
> Robert Orlini wrote:
> > How do I extract the year from a date field such as 2005-10-12?
> >
> > I'm using datepart in the Query as:
> >
> > datepart(yy,report_date),
> >
> > but when I try to output it as:
> >
> >  #datepart("",
> populatedropdown.report_date)# 
> >
> > it gives an form field not found error.
> >
> > Here is the query:
> >
> > 
> > select distinct(datepart(m,report_date)) as month from report
> > where datepart(yy,report_date) = '2005'
> > 
> 
> This query only returns months, so how can you extract a year
> from that?
> 
> Jochem
> 
> 
> 
> 

~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189065
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Extract year from date field

[Jochem van Dieten]

Robert Orlini wrote:
> Thanks Jochem. How would I extract the year from 2005-10-12 please? 

Year().

Jochem

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189064
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Extract year from date field

[Ben Doom]

dateformat(variable, '')

That should do the trick.

--Ben

Robert Orlini wrote:
> Thanks Jochem. How would I extract the year from 2005-10-12 please? 


~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189063
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Extract year from date field

[Robert Orlini]

Thanks Jochem. How would I extract the year from 2005-10-12 please? 

-Original Message-
From: Jochem van Dieten [mailto:[EMAIL PROTECTED]
Sent: Friday, December 31, 2004 10:47 AM
To: CF-Talk
Subject: Re: Extract year from date field


Robert Orlini wrote:
> How do I extract the year from a date field such as 2005-10-12?
> 
> I'm using datepart in the Query as: 
> 
> datepart(yy,report_date), 
> 
> but when I try to output it as:
> 
>  #datepart("", 
> populatedropdown.report_date)# 
> 
> it gives an form field not found error.
> 
> Here is the query:
> 
> 
> select distinct(datepart(m,report_date)) as month from report
> where datepart(yy,report_date) = '2005'
> 

This query only returns months, so how can you extract a year 
from that?

Jochem



~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189062
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: QofQ: SELECT DISTINCT

[Elyse Nielsen]

Hi,

I agree with Joe, try doing the query as a stored procedure if applicable.

Also if it is still a long running stored procedure check out the
indexing of the query.

HTH,
Anticlue


On Fri, 31 Dec 2004 08:41:56 -0500, Joe Rinehart <[EMAIL PROTECTED]> wrote:
> Andrew,
> 
> How does the query you're querying get created in the first place?  Is
> there a way to 'clean' it further 'up' the chain?
> 
> -Joe
> 
> 
> On Fri, 31 Dec 2004 12:20:28 +, Andrew Dixon <[EMAIL PROTECTED]> wrote:
> > Hi Ryan.
> >
> > That custom tag is no help as I need to DISTINCT on three columns. As
> > a work around for the moment, I'm creating a table in the mySQL
> > datasource and then populating it, doing the select and then deleting
> > the table and it is about 100 times (really) quicker. The QofQ takes
> > about 2 seconds and the work around takes about 0.02 seconds, which
> > just seems silly as it is doing so much more work.
> >
> > Andrew.
> >
> >
> 
> 

~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189061
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Extract year from date field

[Jochem van Dieten]

Robert Orlini wrote:
> How do I extract the year from a date field such as 2005-10-12?
> 
> I'm using datepart in the Query as: 
> 
> datepart(yy,report_date), 
> 
> but when I try to output it as:
> 
>  #datepart("", 
> populatedropdown.report_date)# 
> 
> it gives an form field not found error.
> 
> Here is the query:
> 
> 
> select distinct(datepart(m,report_date)) as month from report
> where datepart(yy,report_date) = '2005'
> 

This query only returns months, so how can you extract a year 
from that?

Jochem

~|
Special thanks to the CF Community Suite Silver Sponsor - RUWebby
http://www.ruwebby.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189060
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Extract year from date field

[Robert Orlini]

How do I extract the year from a date field such as 2005-10-12?

I'm using datepart in the Query as: 

datepart(yy,report_date), 

but when I try to output it as:

 #datepart("", 
populatedropdown.report_date)# 

it gives an form field not found error.

Here is the query:


select distinct(datepart(m,report_date)) as month from report
where datepart(yy,report_date) = '2005'
  

Thank you.

Robert O.



~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189059
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: OT: query optimization

[Jochem van Dieten]

Paul Malan wrote:
> 
> You're right, it is still too long.  And it turns out it's taking
> about 15 seconds on average, even slower than I thought.  I updated
> the execution plan scrshot and the indexes as well.  (I did add the
> index to all three applicable tables.)   The link again is
> http:www.redmondinc.com/sos/index.html

The costliest step is still the final aggregation in order to 
perform the DISTINCT. I'm afraid I can't see enough details in 
the graphical execution plans to see what is really going on 
there. (Which join type is hidden behind the popup?)


> I had never seen your technique of using what looks like a subquery in
> the join segment of a query - I'm guessing further optimization is
> available using another one of those?

If that actually eliminates rows early in the process it can help 
a lot. Can you give me the recordcount of the following queries:

SELECT DISTINCT proj_code, cust_job_num, cust_code, order_code
FROM ordr;

SELECT DISTINCT tkt_code, tkt_date, truck_code, po, 
truck_net_wgt, hler_code, order_code
FROM tick;

SELECT DISTINCT prod_descr, order_code
FROM ordl;

(We don't need the same query for the truc table since 
truc.truck_code is the primary key and hence by definition unique.)

Jochem

~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189058
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: OT: query optimization

[Paul Malan]

> I think that is still rather long. Can you update the execution
> plan and index pages if you need it to go even faster?

You're right, it is still too long.  And it turns out it's taking
about 15 seconds on average, even slower than I thought.  I updated
the execution plan scrshot and the indexes as well.  (I did add the
index to all three applicable tables.)   The link again is
http:www.redmondinc.com/sos/index.html

I had never seen your technique of using what looks like a subquery in
the join segment of a query - I'm guessing further optimization is
available using another one of those?  In fact when I learn how to do
that I bet I can apply similar tricks to most of my reports and really
speed things up all across the app.

The funny thing is, our company is so new to modern technology, people
were buying me drinks just for getting them the data, even when hung
the db server for 90 seconds.  Maybe if I really get things humming
they'll buy me a new car or something...

(I have 2 sql books en route, by the way...  "I'm not a slacker, I'm
doin' the work.  I'm babystepping.")

Thanks again for your time and advice.

-Paul

~|
Special thanks to the CF Community Suite Silver Sponsor - New Atlanta
http://www.newatlanta.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189057
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Jim Davis]

> -Original Message-
> From: Tony Hicks [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 31, 2004 9:38 AM
> To: CF-Talk
> Subject: Re: Well - I feel like an idiot.
> 
> Humbly, I was wrong on the locks on application variables.. thats
> information to file for later, thanks.

No need to be humble about it.  The pre-MX locking implementation was
perhaps the biggest gaff in the history of CF and is still one of the most
confusing things.  Let's just all be happy that they fixed it in MX.  ;^)

> But I can't see how both CFIFs are necessary... The outer cfif will
> likely never affect anything.
> 
> If two people hit the site, one request behind another, they'll both
> fire all the way up to the cflock at the same speed, and as we all
> know, that's where the magic comes in.. cflock single threads on the
> init process and one is done before the other.. The outer cfif will
> not catch anything that the inner cfif does not.

It will catch everybody AFTER this process is complete.

You're absolutely right about the behavior while the initialization is
running - but after it's done the outer CFIF means that all visitors simply
check the status and move on rather than obtain a lock and THEN check the
status.

I may not have been clear, but this code is in Application.CFM - which means
its run by every request.  Because it's so central I want this code to run
as fast as possible.
 
> I do wonder if it might be smarter to set application.init to
> SOMETHING before the init process inside the cflock... but I doubt
> that would change anything either.

I considered it, but I didn't see any benefit to the init process.  That
process, itself, doesn't have to be as optimized (it's only run once...
now).  It's getting past it when it has been run that needs to be optimized.

> This is funny, we're arguing tiny fractions of time here.

Tiny fractions of fractions to be sure!  But isn't that what optimization is
all about.  ;^)

This code, as an aside, is on a site that gets lots of traffic in short (two
day) burst.  In case the little optimizations do actually pay off.

Jim Davis



~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189056
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Well - I feel like an idiot.

[Tony Hicks]

Humbly, I was wrong on the locks on application variables.. thats
information to file for later, thanks.

But I can't see how both CFIFs are necessary... The outer cfif will
likely never affect anything.

If two people hit the site, one request behind another, they'll both
fire all the way up to the cflock at the same speed, and as we all
know, that's where the magic comes in.. cflock single threads on the
init process and one is done before the other.. The outer cfif will
not catch anything that the inner cfif does not.

I do wonder if it might be smarter to set application.init to
SOMETHING before the init process inside the cflock... but I doubt
that would change anything either.

This is funny, we're arguing tiny fractions of time here..

On Fri, 31 Dec 2004 08:48:06 -0500, Jeffry Houser <[EMAIL PROTECTED]> wrote:
>  I'm just gonna throw this out there.  Would this be better?:
> 
> 
> 
> 
> ... long init process ...
> 
> 
> 
> 
> 
> 
> 
> Person 1 comes in, Application.init is not defined, so they start initting
> the app.
> Person 2 comes in, and waits for the lock.
> Person 1 inits the app and releases the lock
> Person 2 gets the lock, sees it is already defined, and is all set
> 
>  I can understand why your "fix" worked, however I think the above code
> will accomplish the same thing.
>  As someone else said, you'll grab the lock every time, but I believe that
> will be negligible in all cases except the one where the app is actually be
> initialized.
> 
> At 04:01 AM 12/31/2004, you wrote:
> 
> >Subject: Well - I feel like an idiot.
> >From: "Jim Davis" <[EMAIL PROTECTED]>
> >Date: Fri, 31 Dec 2004 00:00:41 -0500
> >Thread:
> >http://www.houseoffusion.com/cf_lists/index.cfm/method=messages&threadid=37377&forumid=4#189043
> >
> >I've been having trouble with my application starting up under load.
> >Couldn't figure it out.  Once it got going the application purred like a
> >contented cat, but the start up was another matter.
> >
> >Under even moderate load the app would kill JRun.  Not good.
> >
> >Finally figured it - and I'm a dumb ass.  I have code like this in the
> >system to handle start up tasks (which might take as long as 10 seconds to
> >complete):
> >
> >
> >
> >
> > ... long init process ...
> >
> > 
> >
> >
> >
> >
> >See the problem?  I didn't.  Of course (like most things this works in
> >development fine - but under load the first thread will begin the process
> >and ALL the other active threads will queue up waiting to START IT AGAIN!
> >
> >Damn I'm dumb.
> >
> >The following code stops this from happening.  Check the marker, then wait
> >for the lock - but then CHECK THE MARKER again to make sure you have to do
> >it!
> >
> >
> >
> >
> >
> > ... long init process ...
> >
> > 
> >
> >
> >
> >
> >
> >Completely eliminated all my start up issues.
> >
> >I suppose I could just get rid of the outer CFIF but then I'd be setting the
> >lock on every page - I've a gut feeling that's heavier than just doing the
> >CFIF.
> >
> >Jim Davis
> 
> 
> --
> Jeffry Houser, Web Developer, Writer, Songwriter, Recording Engineer
> AIM: Reboog711  | Phone: 1-203-379-0773
> --
> My Books: 
> My Recording Studio: 
> My Energetic Acoustic Rock Band: 
> --
> When did Reality Become TV
> 
> 
> 

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189054
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

RE: Well - I feel like an idiot.

[Jim Davis]

> -Original Message-
> From: Jeffry Houser [mailto:[EMAIL PROTECTED]
> Sent: Friday, December 31, 2004 8:48 AM
> To: CF-Talk
> Subject: Re: Well - I feel like an idiot.
> 
>   I'm just gonna throw this out there.  Would this be better?:
> 
> 
> 
> 
>  ... long init process ...
> 
>  
> 
> 
> 

Yeah - that's actually what I said in my original post.   It would
definitely work but I'm guessing (nothing more that that) that checking the
variable is lighter than obtaining the lock THEN checking the variable.

Since this code is in Application.CFM I want it to be run on the first
request then ignored as quickly as possible by all requests.

Jim Davis



~|
Special thanks to the CF Community Suite Silver Sponsor - RUWebby
http://www.ruwebby.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189055
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Well - I feel like an idiot.

[Jeffry Houser]

  I'm just gonna throw this out there.  Would this be better?:




 ... long init process ...

 





Person 1 comes in, Application.init is not defined, so they start initting 
the app.
Person 2 comes in, and waits for the lock.
Person 1 inits the app and releases the lock
Person 2 gets the lock, sees it is already defined, and is all set

  I can understand why your "fix" worked, however I think the above code 
will accomplish the same thing.
  As someone else said, you'll grab the lock every time, but I believe that 
will be negligible in all cases except the one where the app is actually be 
initialized.


At 04:01 AM 12/31/2004, you wrote:

>Subject: Well - I feel like an idiot.
>From: "Jim Davis" <[EMAIL PROTECTED]>
>Date: Fri, 31 Dec 2004 00:00:41 -0500
>Thread: 
>http://www.houseoffusion.com/cf_lists/index.cfm/method=messages&threadid=37377&forumid=4#189043
>
>I've been having trouble with my application starting up under load.
>Couldn't figure it out.  Once it got going the application purred like a
>contented cat, but the start up was another matter.
>
>Under even moderate load the app would kill JRun.  Not good.
>
>Finally figured it - and I'm a dumb ass.  I have code like this in the
>system to handle start up tasks (which might take as long as 10 seconds to
>complete):
>
>
>
>
> ... long init process ...
>
> 
>
>
>
>
>See the problem?  I didn't.  Of course (like most things this works in
>development fine - but under load the first thread will begin the process
>and ALL the other active threads will queue up waiting to START IT AGAIN!
>
>Damn I'm dumb.
>
>The following code stops this from happening.  Check the marker, then wait
>for the lock - but then CHECK THE MARKER again to make sure you have to do
>it!
>
>
>
>
>
> ... long init process ...
>
> 
>
>
>
>
>
>Completely eliminated all my start up issues.
>
>I suppose I could just get rid of the outer CFIF but then I'd be setting the
>lock on every page - I've a gut feeling that's heavier than just doing the
>CFIF.
>
>Jim Davis


--
Jeffry Houser, Web Developer, Writer, Songwriter, Recording Engineer
AIM: Reboog711  | Phone: 1-203-379-0773
--
My Books: 
My Recording Studio: 
My Energetic Acoustic Rock Band: 
--
When did Reality Become TV 



~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189053
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: QofQ: SELECT DISTINCT

[Joe Rinehart]

Andrew,

How does the query you're querying get created in the first place?  Is
there a way to 'clean' it further 'up' the chain?

-Joe


On Fri, 31 Dec 2004 12:20:28 +, Andrew Dixon <[EMAIL PROTECTED]> wrote:
> Hi Ryan.
> 
> That custom tag is no help as I need to DISTINCT on three columns. As
> a work around for the moment, I'm creating a table in the mySQL
> datasource and then populating it, doing the select and then deleting
> the table and it is about 100 times (really) quicker. The QofQ takes
> about 2 seconds and the work around takes about 0.02 seconds, which
> just seems silly as it is doing so much more work.
> 
> Andrew.
> 
> 

~|
Special thanks to the CF Community Suite Silver Sponsor - CFDynamics
http://www.cfdynamics.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189052
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Coldfusion & IIS behind firewall

[David Manriquez]

Hi CF-Talkers!

  I havea little problem here , hope someone can help me.

I have two sites in IIS configured with one Coldfusion MX server. both sites 
are named site1.myhost.xx and site2.myhost.xx and are viewable from the local 
network working perfectly.!
Well i tried to put one of them online ( internet viewable ) settings some 
parameters on my firewall and DNS.

The problem : I can see only the site1.myhost.xx itself with and index.cfm 
behind scene but if i call it directly or call another file like 
http://site1.myhost.xx/index.cfm , i get an error 404

i have set only the port 80 open for internet connections..
Coldfusion uses another one?..

What' going on here?.. anyone can help me?.. please!


Happy New Year

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189051
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: QofQ: SELECT DISTINCT

[Andrew Dixon]

Hi Ryan.

That custom tag is no help as I need to DISTINCT on three columns. As
a work around for the moment, I'm creating a table in the mySQL
datasource and then populating it, doing the select and then deleting
the table and it is about 100 times (really) quicker. The QofQ takes
about 2 seconds and the work around takes about 0.02 seconds, which
just seems silly as it is doing so much more work.

Andrew.

~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189050
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: OT: query optimization

[Jochem van Dieten]

Paul Malan wrote:
> That is a huge improvement!  The execution time is lower than 10
> seconds now - that's amazing.

I think that is still rather long. Can you update the execution 
plan and index pages if you need it to go even faster?


> For clarification, when I set up the index all I did was give it a
> name and tell it which column to index, then pressed OK.  Nothing more
> to it?  Do I need to do anything to tell the db to create the index,
> or does it just do that when I name it?

That should be enough. Did you do that for the tick, ordr and 
ordl tables or just for one?

Jochem

~|
Special thanks to the CF Community Suite Silver Sponsor - RUWebby
http://www.ruwebby.com

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189049
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: 
http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=11502.10531.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54

Re: Well - I feel like an idiot.

[I-Lin Kuo]

> change anything... Also reads and writes to
> application variables
> should be locked so the outer CFIF is not
> necessarily a good idea.

The outer CFIF is necessary so that once
initialization is over, subsequent requests won't even
try to obtain a lock. Otherwise, every request will
try to obtain a lock.

By the way, this is a familiar idiom in java called
"double-checked locking". I'm glad it works in CF,
because it may fail in java. See 
http://www.cs.umd.edu/~pugh/java/memoryModel/DoubleCheckedLocking.html

=
I-Lin Kuo
Macromedia CF5 Advanced Developer
Sun Certified Java 2 Programmer



__ 
Do you Yahoo!? 
The all-new My Yahoo! - Get yours free! 
http://my.yahoo.com 
 


~|
Special thanks to the CF Community Suite Gold Sponsor - CFHosting.net
http://www.cfhosting.net

Message: http://www.houseoffusion.com/lists.cfm/link=i:4:189048
Archives: http://www.houseoffusion.com/cf_lists/threads.cfm/4
Subscription: http://www.houseoffusion.com/lists.cfm/link=s:4
Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
Donations & Support: http://www.houseoffusion.com/tiny.cfm/54