Search Results

Search found 25520 results on 1021 pages for 'boost test'.

Page 209/1021 | < Previous Page | 205 206 207 208 209 210 211 212 213 214 215 216  | Next Page >

  • listenbacklog directive not working in apache2.2

    - by andrés
    I was trying to make apache 2.2 reject connections if MaxClients was reached, to do this I found the directive ListenBacklog.To test it, I configured apache in the following way: <IfModule mpm_prefork_module> StartServers 5 MinSpareServers 5 MaxSpareServers 10 MaxClients 10 ListenBacklog 1 MaxRequestsPerChild 0 </IfModule> I've made a little script in JMeter to test this. The test launchs 50 users in 1 second (it requests a phpinfo page) but none is rejected, they all wait! I don't understand how this directive works... my operating system is Ubuntu.

    Read the article

  • How would i down-sample a .wav file then reconstruct it using nyquist? - in matlab [closed]

    - by martin
    This is all done in MatLab 2010 My objective is to show the results of: undersampling, nyquist rate/ oversampling First i need to downsample the .wav file to get an incomplete/ or impartial data stream that i can then reconstuct. Heres the flow chart of what im going to be doing So the flow is analog signal - sampling analog filter - ADC - resample down - resample up - DAC - reconstruction analog filter what needs to be achieved: F= Frequency F(Hz=1/s) E.x. 100Hz = 1000 (Cyc/sec) F(s)= 1/(2f) Example problem: 1000 hz = Highest frequency 1/2(1000hz) = 1/2000 = 5x10(-3) sec/cyc or a sampling rate of 5ms This is my first signal processing project using matlab. what i have so far. % Fs = frequency sampled (44100hz or the sampling frequency of a cd) [test,fs]=wavread('test.wav'); % loads the .wav file left=test(:,1); % Plot of the .wav signal time vs. strength time=(1/44100)*length(left); t=linspace(0,time,length(left)); plot(t,left) xlabel('time (sec)'); ylabel('relative signal strength') **%this is were i would need to sample it at the different frequecys (both above and below and at) nyquist frequency.*I think.*** soundsc(left,fs) % shows the resaultant audio file , which is the same as original ( only at or above nyquist frequency however) Can anyone tell me how to make it better, and how to do the various sampling at different frequencies?

    Read the article

  • DDD Melbourne -lessons leant

    - by Michael Freidgeim
    I've attended DDD Melbourne and want to list the interesting points, that I've leant and want to follow. To read more: * Moles-Mocking Isolation framework for .NET. Documentation is here.   (See also Mocking frameworks comparison created October 4, 2009 ) * WebFormsMVP * PluralSight   http://www.pluralsight-training.net/offers/default.aspx?cc=trial   * ELMAH: Error Logging Modules and Handlers *Rhino.Mocks   * VS UI Test Recorder -see posts Visual Studio 2010 Coded UI Test User Guide. Note that Microsoft Test Manager (MTM) toolis a separate application, that can be started from Program files/VS 2010 menu.It is not a menu inside Visual Studio.   * CodeContract- seems great in Debug. Will be good if in production  will be possible runtime configuration, ability to log instead of throw exception. Current recommendation to customize Debug.Assert is not trivial The programmer is free to use the customization provided by Debug.Assert using assert listeners to obtain whatever runtime behavior they desire (e.g., ignoring the error, logging it, or throwing an exception).   // Clears the existing list of assert listener (the default pop-up box) System.Diagnostics.Debug.Listeners.Clear(); // Install your own listener System.Diagnostics.Debug.Listeners.Add(MyTraceListener); Note that you can't catch specific ContractException, but can catch generic Exception(see How come you cannot catch Code Contract exceptions?)   Books recommended "Working effectively with legacy code" by Michael Feathers (corresponding article)   Fowler, Martin Refactoring: Improving the Design of Existing Code, slides http://jaoo.dk/jaoo1999/schedule/MartinFowlerRefractoring.pdf

    Read the article

  • connecting to redis from wsgi application

    - by W_P
    I am running redis and httpd with mod_wsgi on the same server. I have a wsgi application using a virtualenv and the follow virtualhost conf: WSGIPythonPath /var/www/wsgi <VirtualHost *:80> ServerName test.example.com DocumentRoot /var/www/html WSGIScriptAlias / /var/www/wsgi/myapp/app.wsgi CustomLog logs/test.example.com-access-log common ErrorLog logs/test.example.com-error-log <Directory /var/www/wsgi/myapp> Order allow,deny Allow from all WSGIScriptReloading On </Directory> </VirtualHost> I have a sites.pth in my /var/www/wsgi directory, adding myapp to the PYTHONPATH, and this is my /var/www/wsgi/myapp/app.wsgi: activate_this = '/var/www/envs/myapp/bin/activate_this.py' execfile( activate_this, dict(__file__=activate_this) ) from myapp.application import app as application Sorry, I am trying to edit the question to have the rest of my info by it keeps timing out before the edits go through

    Read the article

  • Lightning talk: Coderetreat

    - by Michael Williamson
    In the spirit of trying to encourage more deliberate practice amongst coders in Red Gate, Lauri Pesonen had the idea of running a coderetreat in Red Gate. Lauri and I ran the first one a few weeks ago: given that neither of us hadn’t even been to a coderetreat before, let alone run one, I think it turned out quite well. The participants gave positive feedback, saying that they enjoyed the day, wrote some thought-provoking code and would do it again. Sam Blackburn was one of the attendees, and gave a lightning talk to the other developers in one of our regular lightning talk sessions: In case you can’t watch the video, I’ve transcribed the talk below, although I’d recommend watching the video if you can — I didn’t have much time to do the transcribing! So, what is a coderetreat? So it’s not just something in Red Gate, there’s a website and everything, although it’s not a very big website. It calls itself a community network. The basic ideas behind coderetreat are: you’ve got one day, and you split it into one hour sections. You spend three quarters of that coding, and do a little retrospective at the end. You’re supposed to start fresh each, we were told to delete our code after every session. We were in pairs, swapping after each session, and we did the same task every time. In fact, Conway’s Game of Life is the only task mentioned anywhere that I find for coderetreat. So I don’t know what we’ll do next time, or if we’re meant to do the same thing again. There are some guiding principles which felt to us like restrictions, that you have to code in crazy ways to encourage better code. Final thing is that it’s supposed to be free for outsiders to join. It’s meant to be a kind of networking thing, where you link up with people from other companies. We had a pilot day with Michael and Lauri. Since it was basically the first time any of us had done anything like this, everybody was from Red Gate. We didn’t chat to anybody else for the initial one. The task was Conway’s Game of Life, which most of you have probably heard of it, all but one of us knew about it when did the coderetreat. I won’t got into the details of what it is, but it felt like the right size of task, basically one or two groups actually produced something working by the end of the day, and of course that doesn’t mean it’s necessarily a day’s work to produce that because we were starting again every hour. The task really drives you more than trying to create good code, I found. It was really tempting to try and get it working rather than stick to the rules. But it’s really good to stop and try again because there are so many what-ifs when you’ve finished writing something, “what if I’d done it this way?”. You can answer all those questions at a coderetreat because it’s not about getting a product out the door, it’s about learning and playing with ideas. So we had all these different practices we were trying. I’ll try and go through most of these. Single responsibility is this idea that everything should do just one thing. It was the very first session, we were still trying to figure out how do you go about the Game of Life? So by the end of forty-five minutes hadn’t produced very much for that first session. We were still thinking, “Do we start with a board, how do we represent all these squares? It can be infinitely big, help, this is getting really difficult!”. So, most of us didn’t really get anywhere on the first one. Although it was interesting that some people started with the board, one group started with the FateDecider class that decides whether things live or die. A sort of god class, but in a good way. They managed to implement all of the rules without even defining how the squares were arranged or anything like that. Another thing we tried was TDD (test-driven development). I’m sure most of you know what TDD is: Watch a test, watch it fail for the right reason Write code to pass the test, watch it pass Refactor, check the test still passes Repeat! It basically worked, we were able to produce code, but we often found the tests defined the direction that code went, which is obviously the idea of TDD. But you tend to find that by the time you’ve even written your first assertion, which is supposed to be the very first thing you write, because you write your tests backwards from the assertions back to the initial conditions, you’ve already constrained the logic of the code in some way by the time you’ve done that. You then get to this situation of, “Well, we actually want to go in a slightly different direction. Can we do this?”. Can we write tests that don’t constrain the architecture? Wrapping up all primitives: it’s kind of turtles all the way down. We had a Size, which has a Width and Height, which both derive from Dimension. You’ve got pages of code before you’ve even done anything. No getters and setters (use tell don’t ask instead): mocks and stubs for tests are required if you want to assert that your results are what you think they should be. You can’t just check the internal state of the code. And people found that really challenging and it made them think in a different way which I think is really good. Not having mutable state: that was kind of confusing because we weren’t quite sure what fitted within that rule and what didn’t, and I think we were trying too hard to follow the rule rather than the guideline. No if-statements: supposed to use polymorphism instead, but polymorphism still requires a factory with conditional behaviour. We did something really crazy to get around this: public T If(bool condition, Func<T> left, Func<T> right) { var dict = new Dictionary<bool, Func<T>> {{true, left}, {false, right}}; return dict[condition].Invoke(); } That is not really polymorphism, is it? For-loops: you can always replace a for-loop with recursion, but it doesn’t tend to make it any more readable unless it’s the kind of task that really lends itself to that. So it was interesting, it was good practice, but it wouldn’t make it easier it’s the kind of tree-structure algorithm where that would help. Having a limit on the number of levels of indentation: again, I think it does produce very nice, clean code, but it wasn’t actually a challenge because you just extract methods. That’s quite a useful thing because you can apply that to real code and say, “Okay, should this method really be going crazy like this?” No talking: we hated that. It’s like there’s two of you at a computer, and one of you is doing the typing, what does the other guy do if they’re not allowed to talk. The answer is TDD ping-pong – one person writes the tests, and then the other person writes the code to pass the test. And that creates communication without actually having to have discussion about things which is kind of cool. No code comments: just makes no difference to anything. It’s a forty-five minute exercise, so what are you going to put comments in code for? Finally, this is my fault. I discovered an entertaining way of doing the calculation that was kind of cool (using convolutions over the state of the board). Unfortunately, it turns out to be really hard to implement in C#, so didn’t even manage to work out how to do that convolution in C#. It’s trivial in some high-level languages, but you need something matrix-orientated for it to really work. That’s most of it, really. The thoughts that people went away with: we put down our answers to questions like “What have you learnt?” and “What surprised you?”, “How are you going to do things differently?”, and most people said redoing the problem is really, really good for understanding it properly. People hate having a massive legacy codebase that they can’t change, so being able to attack something three different ways in an environment where the end-product isn’t important: that’s something people really enjoyed. Pair-programming: also people said that they wanted to do more of that, especially with TDD ping-pong, where you write the test and somebody else writes the code. Various people thought different things about immutables, but most people thought they were good, they promote functional programming. And TDD people found really hard. “Tell, don’t ask” people found really, really hard and really, really, really hard to do well. And the recursion just made things trickier to debug. But most people agreed that coderetreats are really cool, and we should do more of them.

    Read the article

  • Can I still use unity 2d [duplicate]

    - by dragonloverlord
    This question already has an answer here: Is it possible to change Unity 3D to 2D and will I gain any performance boost after that? 3 answers I can not run unity 3d on my Chromebook but unity 2d in Ubuntu 12.04 works fine so is it possible to run unity low graphics mode on Ubuntu 14.04 as an alternative? If I can run low graphics mode as an alternative then how would I go about that? If I can not then what would be a good unity like alternative for Ubuntu 14.04?

    Read the article

  • Need to add request headers to every request in Apache

    - by 115748146017471869327
    I'm trying to add a header value to every request via Apache (ver 2.2). I've edited my VirtualHost to include the following vaiations: (I've tried both RequestHeader and Header, add and set in all of these cases) RequestHeader set X-test_url "Test" or <Directory /> RequestHeader set X-test_url "Test" </Directory> or <Location ~ "/*" > RequestHeader set X-test_url "Test" </Location> It's hard to explain how I've gotten to this point, but I have to get this done in Apache. Again I'm trying to add the header value to every request. Thanks.

    Read the article

  • Sucky MSTest and the "WaitAll for multiple handles on a STA thread is not supported" Error

    - by Anne Bougie
    If you are doing any multi-threading and are using MSTest, you will probably run across this error. For some reason, MSTest by default runs in STA threading mode. WTF, Microsoft! Why so stuck in the old COM world?  When I run the same test using NUnit, I don't have this problem. Unfortunately, my company has chosen MSTest, so I have a lot of testing problems. NUnit is so much better, IMO. After determining that I wasn't referencing any unmanaged code that would flip the thread into STA, which can also cause this error, the only thing left was the testing suite I was using. I dug around a little and found this obscure setting for the Test Run Config settings file that you can't set using its interface. You have to open it up as a text file and add the following setting:  <ExecutionThread apartmentState="MTA" /> This didn't break any other tests, so I'm not sure why it's not the default, or why there is nothing in the test run configuration app to change this setting. Here is the code I was testing:  public void ProcessTest(ProcessInfo[] infos) {    WaitHandle[] waits = new WaitHandle[infos.Length];    int i = 0;    foreach (ProcessInfo info in infos)    {       AutoResetEvent are = new AutoResetEvent(false);       info.Are = are;       waits[i++] = are;         Processor pr = new Processor();       WaitCallback callback = pr.ProcessTest;       ThreadPool.QueueUserWorkItem(callback, info);    }      WaitHandle.WaitAll(waits); }

    Read the article

  • Cannot create file in directory even though it's writable by a group I belong to

    - by Alan Berndt
    I have a directory structure owned by a certain group, and I am a member of the group that owns these directories. I am able to create files in one directory, but not in another, even though the permissions are the same. alan@bricky:/mnt/storage/media$ stat Music Music\ \(Lossy\)/ File: `Music' Size: 34 Blocks: 8 IO Block: 4096 directory Device: fb00h/64256d Inode: 4215424 Links: 3 Access: (2775/drwxrwsr-x) Uid: ( 1001/ media) Gid: ( 1001/ media) Access: 2011-08-19 11:45:03.182586898 -0700 Modify: 2011-08-19 11:44:01.412840027 -0700 Change: 2011-08-19 11:45:02.734603240 -0700 File: `Music (Lossy)/' Size: 6 Blocks: 8 IO Block: 4096 directory Device: fb00h/64256d Inode: 1512056832 Links: 2 Access: (2775/drwxrwsr-x) Uid: ( 1001/ media) Gid: ( 1001/ media) Access: 2011-08-19 11:45:03.190586606 -0700 Modify: 2011-08-19 10:34:46.526530313 -0700 Change: 2011-08-19 11:45:02.738603094 -0700 alan@bricky:/mnt/storage/media$ touch Music/test alan@bricky:/mnt/storage/media$ touch Music\ \(Lossy\)/test touch: cannot touch `Music (Lossy)/test': Permission denied

    Read the article

  • Glibc importance of error ...

    - by Oz123
    Hi Everyone, I am following LFS 6.7, and I reached the point where I compile glibc-2.12.1 . I mounted the LFS partition with the atime option: here is a confirm on that I think: /dev/sdb1 on /mnt /lfs type ext4 (rw) I get the following errors on making the test, and I have no clue if I should try to resolve them, or just ignore them and go on ... rpc/types.h sunrpc/rpc/svc_auth.h sunrpc/rpcsvc/bootparam.h sysvipc/sys/ipc.h \ sysvipc/sys/msg.h sysvipc/sys/sem.h sysvipc/sys/shm.h termios/termios.h \ termios/sys/termios.h termios/sys/ttychars.h time/time.h time/sys/time.h \ time/sys/timeb.h wcsmbs/wchar.h wctype/wctype.h > \ /sources/glibc-build/begin-end-check.out make[1]: Target `check' not remade because of errors. make[1]: Leaving directory `/sources/glibc-2.12.1' make: *** [check] Error 2 root:/sources/glibc-build# grep Error glibc-check-log make[2]: *** [/sources/glibc-build/math/test-float.out] Error 1 make[2]: *** [/sources/glibc-build/math/test-ifloat.out] Error 1 make[1]: *** [math/tests] Error 2 make[2]: [/sources/glibc-build/posix/annexc.out] Error 1 (ignored) make: *** [check] Error 2 thanks in advance, Oz

    Read the article

  • IE Kerberos failure on some machines with CNAME web server (with SPN for host's A record)

    - by Eric Thames
    It's fairly well known that IE doesn't like to do Kerberos against hosts that are registered in DNS as CNAMEs. What happens is that IE turns around and uses the underlying A record for the host for looking up the Service Principal Name (SPN). On a test network we are able to get Kerberos working by having the SPN registered for the A record of the host, so that Kerberos authentication happens successfully when accessing the web server via it's CNAME in the browser. Kerberos authentication works properly when directly accessing the web server with the A record host in the URL, but for various reasons that are beyond my control, it is desired to use the CNAME. On the production network, this same configuration fails though and I can't figure out why. Any thoughts? This is a java web application using the SPNEGO library - not IIS. Kerberos authentication is working properly in both the test and production networks (and has been confirmed to not fail back to NTLM), but the CNAME access only works in test.

    Read the article

  • A Reusable Builder Class for Ruby Testing

    - by Liam McLennan
    My last post was about a class for building test data objects in C#. This post describes the same tool, but implemented in Ruby. The C# version was written first but I originally came up with the solution in my head using Ruby, and then I translated it to C#. The Ruby version was easier to write and is easier to use thanks to Ruby’s dynamic nature making generics unnecessary.  Here are my example domain classes: class Person attr_accessor :name, :age def initialize(name, age) @name = name @age = age end end class Property attr_accessor :street, :manager def initialize(street, manager) @street = street @manager = manager end end and the test class showing what the builder does: class Test_Builder < Test::Unit::TestCase def setup @build = Builder.new @build.configure({ Property => lambda { Property.new '127 Creek St', @build.a(Person) }, Person => lambda { Person.new 'Liam', 26 } }) end def test_create assert_not_nil @build end def test_can_get_a_person @person = @build.a(Person) assert_not_nil @person assert_equal 'Liam', @person.name assert_equal 26, @person.age end def test_can_get_a_modified_person @person = @build.a Person do |person| person.age = 999 end assert_not_nil @person assert_equal 'Liam', @person.name assert_equal 999, @person.age end def test_can_get_a_different_type_that_depends_on_a_type_that_has_not_been_configured_yet @my_place = @build.a(Property) assert_not_nil @my_place assert_equal '127 Creek St', @my_place.street assert_equal @build.a(Person).name, @my_place.manager.name end end Finally, the implementation of Builder: class Builder # defaults is a hash of Class => creation lambda def configure defaults @defaults = defaults end def a(klass) temp = @defaults[klass].call() yield temp if block_given? temp end end

    Read the article

  • HTTP Headers - need to check OPTIONS

    - by chris
    I've received the results of a pen test and there is some unwanted exposure in the HTTP OPTIONS where the fact that Frontpage Server Extensions was available (now removed) was reported. I need to run a check on the OPTIONS to see whether it has been removed. The test output from the report is below, I need to recreate it to establish that it has gone but don't know how to check the OPTIONS, I can only find tools that seem to check the "HEAD / HTTP/1.1". Does anyone know how to test this - I'm running a Windows environment? Many thanks OPTIONS / HTTP/1.1 Host: www.website.com HTTP/1.1 200 OK Allow: OPTIONS, TRACE, GET, HEAD Content-Length: 0 Server: Microsoft-IIS/6.0 Public: OPTIONS, TRACE, GET, HEAD, POST MS-Author-Via: MS-FP/4.0 X-Powered-By: ASP.NET MicrosoftOfficeWebServer: 5.0_Pub Date: Fri, 01 Feb 2010 16:09:15 GMT

    Read the article

  • Welcome files are not loaded! Need help with Railo, mappings and J2EE configuration!

    - by mrt181
    I have installed a J2EE Server (tried it with Glassfish3, Tomcat6 and Resin4) on Win7 64bit and deployed Railo3.1. I have then added a virtual host to the J2EE server, i.e. Resin: <host host-name="railo"> C:/resin/webapps/railo In the Railo Admin i have added this mapping: Virtual Physical / C:/webapps/ When i access http://railo:8080/ my index.cfm welcome file in C:/webapps/ is loaded (index.cfm is definded in Railos web.xml). When i try to access http://railo:8080/test which contains the same index.cfm i get an 500 Servlet Exception java.io.FileNotFoundException: C:\webapps\test (access denied) (on all J2EE Servers i tried so far). http://railo:8080/test/index.cfm works fine. I already tried to add index.cfm to Resins welcome-file-list in app-default.xml to no avail. I want to be able to access deployed apps without this url: http://localhost:8080/app/ Instead i want to use this: http://app:8080/

    Read the article

  • Bit by bit comparison of using Java or Python for unit testing frameworks and Selenium

    - by Anirudh
    Currently we are in the process of finalizing which language out of Java, Python should be used for Automation using selenium webdriver and a suitable unit testing frameworks. I have made use of Junit, TestNG and webdriver while using with Java and have designed frameworks without much fuss before. I am new to python though I came across pyhton's unit testing frameworks like unittest, pyunit, nose e.t.c but I have doubts if they would be as successful as testNG or Java. I would like to analyze point by point when used with selenium webdriver as below: 1)I have read that as Python is an interpreted language hence it's execution is slower, so say if I have to run 1000 test cases which take about 6 hours to run in Java, would python take considerably longer time for the same test cases like 8 hours? 2)Can the Python unit testing framework be as flexible as a Java unit testing framework like testNG in terms or Grouping the tests, parallel execution, skipping test. e.t.c 3)Also one point that I think of is that Python with selenium webdriver doeasn't have as big or learned community as we have for Java with webdriver, say if I run into trouble with something I am more likely to find an answer for Java as compared to python? 4)Somewhat related to point 3, is it safe to rely on tools, plugins or even webderiver's python's binding as a continuously well maintained? 5)One major drawback as I see while using python's unit testing framework is lack of boilerplate code or libraries for nicely illustrative HTML reports preferably historical reports with Pie charts, bar graphs and timelines as we have in case of Java like Allure, TestNG's default reports, reportNG or Junit reports with the help of ANT as shown below Allure Reports Junit Historical reports Also I would like to emphasize on the fact if there is a way for one to write the framework in java and make libraries or utilities according to out application in webdriver which can easily be called or integrated in with python code or modules? That would actually solve the problem for us as the client would be able to use the code we write in Java and make use of the same or call it from their python modules?

    Read the article

  • PHP throwing XDebug errors ONLY in command line mode...

    - by Wilhelm Murdoch
    Hey, all! I've been having a few problems running PHP-based utilities within the command line ever since I enabled the XDebug. It runs just fine when executing script through a browser, but once I try an execute a script on the command line, it throws the following errors: h:\www\test>@php test.php PHP Warning: PHP Startup: Unable to load dynamic library 'E:\development\xampplite\php\ext\php_curl.dll' - The specified module could not be found in Unknown on line 0 PHP Warning: Xdebug MUST be loaded as a Zend extension in Unknown on line 0 h:\www\test> The script runs just fine after this, but it's something I can't seem to wrap my head around. Could it be a path issue within my php.ini config? I'm not sure if that's the case considering it throws the same error no matter where I access the @php environmental variable. Also, all paths within my php.ini are absolute. Not really sure what's going on here. Any ideas? Thanks!

    Read the article

  • Failure to toubleshoot a juju charm deployment

    - by Bruno Pereira
    My environments.yaml looks like this: environments: test: type: local control-bucket: juju-a14dfae3830142d9ac23c499395c2785999 admin-secret: 6608267bbd6b447b8c90934167b2a294999 default-series: oneiric juju-origin: distro data-dir: /home/bruno/projects/juju juju bootstrap runs perfect: 2011-11-22 19:19:31,999 INFO Bootstrapping environment 'test' (type: local)... 2011-11-22 19:19:32,004 INFO Checking for required packages... 2011-11-22 19:19:33,584 INFO Starting networking... 2011-11-22 19:19:34,058 INFO Starting zookeeper... 2011-11-22 19:19:34,283 INFO Starting storage server... 2011-11-22 19:19:40,051 INFO Initializing zookeeper hierarchy 2011-11-22 19:19:40,247 INFO Starting machine agent (origin: distro)... [sudo] password for bruno: 2011-11-22 19:23:16,054 INFO Environment bootstrapped 2011-11-22 19:23:16,079 INFO 'bootstrap' command finished successfully Deploy from a known good charm is accepted (tried it with one that I am trying to create): juju deploy --repository=/home/bruno/projects/charms_repo/ local:teamspeak 2011-11-22 19:28:49,929 INFO Charm deployed as service: 'teamspeak' 2011-11-22 19:28:49,962 INFO 'deploy' command finished successfully After this I can see that juju debug-log shows activity and I can see the network indicator going on and off and activity on my hard-disk. Wait... Looking at juju status I get: services: teamspeak: charm: local:oneiric/teamspeak-1 relations: {} units: teamspeak/0: machine: 0 public-address: 192.168.122.226 relations: {} state: start_error juju debug-log does not help and I have no files under /var/log/juju or /var/lib/juju. Last juju debug-log only shows this: 2011-11-22 19:45:20,790 Machine:0: juju.agents.machine DEBUG: Units changed old:set(['wordpress/0']) new:set(['wordpress/0', 'teamspeak/0']) 2011-11-22 19:45:20,823 Machine:0: juju.agents.machine DEBUG: Starting service unit: teamspeak/0 ... 2011-11-22 19:45:21,137 Machine:0: juju.agents.machine DEBUG: Downloading charm local:oneiric/teamspeak-1 to /home/bruno/projects/juju/bruno-test/charms 2011-11-22 19:45:22,115 Machine:0: juju.agents.machine DEBUG: Starting service unit teamspeak/0 2011-11-22 19:45:22,133 Machine:0: unit.deploy INFO: Creating container teamspeak-0... 2011-11-22 19:47:04,586 Machine:0: unit.deploy INFO: Container created for teamspeak/0 2011-11-22 19:47:04,781 Machine:0: unit.deploy DEBUG: Charm extracted into container 2011-11-22 19:47:04,801 Machine:0: unit.deploy DEBUG: Starting container... 2011-11-22 19:47:07,086 Machine:0: unit.deploy INFO: Started container for teamspeak/0 2011-11-22 19:47:07,107 Machine:0: juju.agents.machine INFO: Started service unit teamspeak/0 How can I troubleshot what is happening here?

    Read the article

  • Click Once Deployment Process and Issue Resolution

    - by Geordie
    Introduction We are adopting Click Once as a deployment standard for Thick .Net application clients.  The latest version of this tool has matured it to a point where it can be used in an enterprise environment.  This guide will identify how to use Click Once deployment and promote code trough the dev, test and production environments. Why Use Click Once over SCCM If we already use SCCM why add Click Once to the deployment options.  The advantages of Click Once are their ability to update the code in a single location and have the update flow automatically down to the user community.  There have been challenges in the past with getting configuration updates to download but these can now be achieved.  With SCCM you can do the same thing but it then needs to be packages and pushed out to users.  Each time a new user is added to an application, time needs to be spent by an administrator, to push out any required application packages.  With Click Once the user would go to a web link and the application and pre requisites will automatically get installed. New Deployment Steps Overview The deployment in an enterprise environment includes several steps as the solution moves through the development life cycle before being released into production.  To make mitigate risk during the release phase, it is important to ensure the solution is not deployed directly into production from the development tools.  Although this is the easiest path, it can introduce untested code into production and result in unexpected results. 1. Deploy the client application to a development web server using Visual Studio 2008 Click Once deployment tools.  Once potential production versions of the solution are being generated, ensure the production install URL is specified when deploying code from Visual Studio.  (For details see ‘Deploying Click Once Code from Visual Studio’) 2. xCopy the code to the test server.  Run the MageUI tool to update the URLs, signing and version numbers to match the test server. (For details see ‘Moving Click Once Code to a new Server without using Visual Studio’) 3. xCopy the code to the production server.  Run the MageUI tool to update the URLs, signing and version numbers to match the production server. The certificate used to sign the code should be provided by a certificate authority that will be trusted by the client machines.  Finally make sure the setup.exe contains the production install URL.  If not redeploy the solution from Visual Studio to the dev environment specifying the production install URL.  Then xcopy the install.exe file from dev to production.  (For details see ‘Moving Click Once Code to a new Server without using Visual Studio’) Detailed Deployment Steps Deploying Click Once Code From Visual Studio Open Visual Studio and create a new WinForms or WPF project.   In the solution explorer right click on the project and select ‘Publish’ in the context menu.   The ‘Publish Wizard’ will start.  Enter the development deployment path.  This could be a local directory or web site.  When first publishing the solution set this to a development web site and Visual basic will create a site with an install.htm page.  Click Next.  Select weather the application will be available both online and offline. Then click Finish. Once the initial deployment is completed, republish the solution this time mapping to the directory that holds the code that was just published.  This time the Publish Wizard contains and additional option.   The setup.exe file that is created has the install URL hardcoded in it.  It is this screen that allows you to specify the URL to use.  At some point a setup.exe file must be generated for production.  Enter the production URL and deploy the solution to the dev folder.  This file can then be saved for latter use in deployment to production.  During development this URL should be pointing to development site to avoid accidently installing the production application. Visual studio will publish the application to the desired location in the process it will create an anonymous ‘pfx’ certificate to sign the deployment configuration files.  A production certificate should be acquired in preparation for deployment to production.   Directory structure created by Visual Studio     Application files created by Visual Studio   Development web site (install.htm) created by Visual Studio Migrating Click Once Code to a new Server without using Visual Studio To migrate the Click Once application code to a new server, a tool called MageUI is needed to modify the .application and .manifest files.  The MageUI tool is usually located – ‘C:\Program Files\Microsoft SDKs\Windows\v6.0A\Bin’ folder or can be downloaded from the web. When deploying to a new environment copy all files in the project folder to the new server.  In this case the ‘ClickOnceSample’ folder and contents.  The old application versions can be deleted, in this case ‘ClickOnceSample_1_0_0_0’ and ‘ClickOnceSample_1_0_0_1’.  Open IIS Manager and create a virtual directory that points to the project folder.  Also make the publish.htm the default web page.   Run the ManeUI tool and then open the .application file in the root project folder (in this case in the ‘ClickOnceSample’ folder). Click on the Deployment Options in the left hand list and update the URL to the new server URL and save the changes.   When MageUI tries to save the file it will prompt for the file to be signed.   This step cannot be bypassed if you want the Click Once deployment to work from a web site.  The easiest solution to this for test is to use the auto generated certificate that Visual Studio created for the project.  This certificate can be found with the project source code.   To save time go to File>Preferences and configure the ‘Use default signing certificate’ fields.   Future deployments will only require application files to be transferred to the new server.  The only difference is then updating the .application file the ‘Version’ must be updated to match the new version and the ‘Application Reference’ has to be update to point to the new .manifest file.     Updating the Configuration File of a Click Once Deployment Package without using Visual Studio When an update to the configuration file is required, modifying the ClickOnceSample.exe.config.deploy file will not result in current users getting the new configurations.  We do not want to go back to Visual Studio and generate a new version as this might introduce unexpected code changes.  A new version of the application can be created by copying the folder (in this case ClickOnceSample_1_0_0_2) and pasting it into the application Files directory.  Rename the directory ‘ClickOnceSample_1_0_0_3’.  In the new folder open the configuration file in notepad and make the configuration changes. Run MageUI and open the manifest file in the newly copied directory (ClickOnceSample_1_0_0_3).   Edit the manifest version to reflect the newly copied files (in this case 1.0.0.3).  Then save the file.  Open the .application file in the root folder.  Again update the version to 1.0.0.3.  Since the file has not changed the Deployment Options/Start Location URL should still be correct.  The application Reference needs to be updated to point to the new versions .manifest file.  Save the file. Next time a user runs the application the new version of the configuration file will be down loaded.  It is worth noting that there are 2 different types of configuration parameter; application and user.  With Click Once deployment the difference is significant.  When an application is downloaded the configuration file is also brought down to the client machine.  The developer may have written code to update the user parameters in the application.  As a result each time a new version of the application is down loaded the user parameters are at risk of being overwritten.  With Click Once deployment the system knows if the user parameters are still the default values.  If they are they will be overwritten with the new default values in the configuration file.  If they have been updated by the user, they will not be overwritten. Settings configuration view in Visual Studio Production Deployment When deploying the code to production it is prudent to disable the development and test deployment sites.  This will allow errors such as incorrect URL to be quickly identified in the initial testing after deployment.  If the sites are active there is no way to know if the application was downloaded from the production deployment and not redirected to test or dev.   Troubleshooting Clicking the install button on the install.htm page fails. Error: URLDownloadToCacheFile failed with HRESULT '-2146697210' Error: An error occurred trying to download <file>   This is due to the setup.exe file pointing to the wrong location. ‘The setup.exe file that is created has the install URL hardcoded in it.  It is this screen that allows you to specify the URL to use.  At some point a setup.exe file must be generated for production.  Enter the production URL and deploy the solution to the dev folder.  This file can then be saved for latter use in deployment to production.  During development this URL should be pointing to development site to avoid accidently installing the production application.’

    Read the article

  • Apache/passenger/ree doesn't interpret .rb files

    - by Sergey
    I'm trying to get apache + passenger + ree to work. I think I did everything (except for setting up rails env - for now I wanna run just pure ruby) described here: http://rvm.beginrescueend.com/integration/passenger/ But when I try to go to localhost/test.rb it doesn't interpret that file and just download it. I don't know where should I look for mistakes, so here are a few files I think could be relevant: /var/log/apache2/error.log (these 2 lines are repeating) [Mon May 31 23:12:47 2010] [notice] Graceful restart requested, doing restart [Mon May 31 23:12:48 2010] [notice] Apache/2.2.14 (Ubuntu) PHP/5.3.2-1ubuntu4.2 with Suhosin-Patch Phusion_Passenger/2.2.11 configured -- resuming normal operations /etc/apache2/httpd.conf LoadModule passenger_module /home/sergey/.rvm/gems/ree-1.8.7-2010.01/gems/passenger-2.2.11/ext/apache2/mod_passenger.so PassengerRoot /home/sergey/.rvm/gems/ree-1.8.7-2010.01/gems/passenger-2.2.11 PassengerRuby /home/sergey/.rvm/bin/passenger_ruby /var/www/test.rb puts "test"

    Read the article

  • How to get .cgi files working in python with Apache on Ubuntu

    - by tapan
    I am not really sure whether to post this here or on SO. I think this has more to do with server administration so i am posting it here. I have apache 2.2 installed on my ubuntu 10.10 system with libapache2-mod-python. However when i put a .cgi file with python in it in my apache root (/var/www) it doesn't get executed and all i see is the script i have written. For example this should show the text "Test": #!/usr/bin/python print "Test" However the above script shows up in my browser. Any idea what conf files,etc. i'll have to go through and change to allow this to work ? (the file perms on test.cgi is 755 if that makes a difference.)

    Read the article

  • Diagnose Issues with USB External Hard Drive

    - by Chris Lieb
    I have a 1TB SATA hard drive in a third-party USB enclosure hooked up to my computer via USB. The drive has become increasingly unreliable, randomly unmounting and remounting in Windows, making strange noises on occasion, and every once in a while making lots of noise and then unmounting, only fixable by powering down the enclosure, waiting a while, then powering it back up again. I am unsure if the hard drive or the enclosure is to blame. The random unmount/remount could be the enclosure, but the strange sounds are likely related to the hard drive. Given this uncertainty, I want to figure out which of the two items needs to be replaced. However, I do not know of a way to test each item separately since I do not have an extra SATA-to-USB enclosure to test the hard drive with or an extra SATA hard drive to test the enclosure with. Does anyone know how to figure out what's to blame so that I don't have to replace both unless it is absolutely necessary.

    Read the article

  • Ogre 3d and bullet physics interaction

    - by Tim
    I have been playing around with Ogre3d and trying to integrate bullet physics. I have previously somewhat successfully got this functionality working with irrlicht and bullet and I am trying to base this on what I had done there, but modifying it to fit with Ogre. It is working but not correctly and I would like some help to understand what it is I am doing wrong. I have a state system and when I enter the "gamestate" I call some functions such as setting up a basic scene, creating the physics simulation. I am doing that as follows. void GameState::enter() { ... // Setup Physics btBroadphaseInterface *BroadPhase = new btAxisSweep3(btVector3(-1000,-1000,-1000), btVector3(1000,1000,1000)); btDefaultCollisionConfiguration *CollisionConfiguration = new btDefaultCollisionConfiguration(); btCollisionDispatcher *Dispatcher = new btCollisionDispatcher(CollisionConfiguration); btSequentialImpulseConstraintSolver *Solver = new btSequentialImpulseConstraintSolver(); World = new btDiscreteDynamicsWorld(Dispatcher, BroadPhase, Solver, CollisionConfiguration); ... createScene(); } In the createScene method I add a light and try to setup a "ground" plane to act as the ground for things to collide with.. as follows. I expect there is issues with this as I get objects colliding with the ground but half way through it and they glitch around like crazy on collision. void GameState::createScene() { m_pSceneMgr->createLight("Light")->setPosition(75,75,75); // Physics // As a test we want a floor plane for things to collide with Ogre::Entity *ent; Ogre::Plane p; p.normal = Ogre::Vector3(0,1,0); p.d = 0; Ogre::MeshManager::getSingleton().createPlane( "FloorPlane", Ogre::ResourceGroupManager::DEFAULT_RESOURCE_GROUP_NAME, p, 200000, 200000, 20, 20, true, 1, 9000,9000,Ogre::Vector3::UNIT_Z); ent = m_pSceneMgr->createEntity("floor", "FloorPlane"); ent->setMaterialName("Test/Floor"); Ogre::SceneNode *node = m_pSceneMgr->getRootSceneNode()->createChildSceneNode(); node->attachObject(ent); btTransform Transform; Transform.setIdentity(); Transform.setOrigin(btVector3(0,1,0)); // Give it to the motion state btDefaultMotionState *MotionState = new btDefaultMotionState(Transform); btCollisionShape *Shape = new btStaticPlaneShape(btVector3(0,1,0),0); // Add Mass btVector3 LocalInertia; Shape->calculateLocalInertia(0, LocalInertia); // CReate the rigid body object btRigidBody *RigidBody = new btRigidBody(0, MotionState, Shape, LocalInertia); // Store a pointer to the Ogre Node so we can update it later RigidBody->setUserPointer((void *) (node)); // Add it to the physics world World->addRigidBody(RigidBody); Objects.push_back(RigidBody); m_pNumEntities++; // End Physics } I then have a method to create a cube and give it rigid body physics properties. I know there will be errors here as I get the items colliding with the ground but not with each other properly. So I would appreciate some input on what I am doing wrong. void GameState::CreateBox(const btVector3 &TPosition, const btVector3 &TScale, btScalar TMass) { Ogre::Vector3 size = Ogre::Vector3::ZERO; Ogre::Vector3 pos = Ogre::Vector3::ZERO; Ogre::Vector3 scale = Ogre::Vector3::ZERO; pos.x = TPosition.getX(); pos.y = TPosition.getY(); pos.z = TPosition.getZ(); scale.x = TScale.getX(); scale.y = TScale.getY(); scale.z = TScale.getZ(); Ogre::Entity *entity = m_pSceneMgr->createEntity( "Box" + Ogre::StringConverter::toString(m_pNumEntities), "cube.mesh"); entity->setCastShadows(true); Ogre::AxisAlignedBox boundingB = entity->getBoundingBox(); size = boundingB.getSize(); //size /= 2.0f; // Only the half needed? //size *= 0.96f; // Bullet margin is a bit bigger so we need a smaller size entity->setMaterialName("Test/Cube"); Ogre::SceneNode *node = m_pSceneMgr->getRootSceneNode()->createChildSceneNode(); node->attachObject(entity); node->setPosition(pos); //node->scale(scale); // Physics btTransform Transform; Transform.setIdentity(); Transform.setOrigin(TPosition); // Give it to the motion state btDefaultMotionState *MotionState = new btDefaultMotionState(Transform); btVector3 HalfExtents(TScale.getX()*0.5f,TScale.getY()*0.5f,TScale.getZ()*0.5f); btCollisionShape *Shape = new btBoxShape(HalfExtents); // Add Mass btVector3 LocalInertia; Shape->calculateLocalInertia(TMass, LocalInertia); // CReate the rigid body object btRigidBody *RigidBody = new btRigidBody(TMass, MotionState, Shape, LocalInertia); // Store a pointer to the Ogre Node so we can update it later RigidBody->setUserPointer((void *) (node)); // Add it to the physics world World->addRigidBody(RigidBody); Objects.push_back(RigidBody); m_pNumEntities++; } Then in the GameState::update() method which which runs every frame to handle input and render etc I call an UpdatePhysics method to update the physics simulation. void GameState::UpdatePhysics(unsigned int TDeltaTime) { World->stepSimulation(TDeltaTime * 0.001f, 60); btRigidBody *TObject; for(std::vector<btRigidBody *>::iterator it = Objects.begin(); it != Objects.end(); ++it) { // Update renderer Ogre::SceneNode *node = static_cast<Ogre::SceneNode *>((*it)->getUserPointer()); TObject = *it; // Set position btVector3 Point = TObject->getCenterOfMassPosition(); node->setPosition(Ogre::Vector3((float)Point[0], (float)Point[1], (float)Point[2])); // set rotation btVector3 EulerRotation; QuaternionToEuler(TObject->getOrientation(), EulerRotation); node->setOrientation(1,(Ogre::Real)EulerRotation[0], (Ogre::Real)EulerRotation[1], (Ogre::Real)EulerRotation[2]); //node->rotate(Ogre::Vector3(EulerRotation[0], EulerRotation[1], EulerRotation[2])); } } void GameState::QuaternionToEuler(const btQuaternion &TQuat, btVector3 &TEuler) { btScalar W = TQuat.getW(); btScalar X = TQuat.getX(); btScalar Y = TQuat.getY(); btScalar Z = TQuat.getZ(); float WSquared = W * W; float XSquared = X * X; float YSquared = Y * Y; float ZSquared = Z * Z; TEuler.setX(atan2f(2.0f * (Y * Z + X * W), -XSquared - YSquared + ZSquared + WSquared)); TEuler.setY(asinf(-2.0f * (X * Z - Y * W))); TEuler.setZ(atan2f(2.0f * (X * Y + Z * W), XSquared - YSquared - ZSquared + WSquared)); TEuler *= RADTODEG; } I seem to have issues with the cubes not colliding with each other and colliding strangely with the ground. I have tried to capture the effect with the attached image. I would appreciate any help in understanding what I have done wrong. Thanks. EDIT : Solution The following code shows the changes I made to get accurate physics. void GameState::createScene() { m_pSceneMgr->createLight("Light")->setPosition(75,75,75); // Physics // As a test we want a floor plane for things to collide with Ogre::Entity *ent; Ogre::Plane p; p.normal = Ogre::Vector3(0,1,0); p.d = 0; Ogre::MeshManager::getSingleton().createPlane( "FloorPlane", Ogre::ResourceGroupManager::DEFAULT_RESOURCE_GROUP_NAME, p, 200000, 200000, 20, 20, true, 1, 9000,9000,Ogre::Vector3::UNIT_Z); ent = m_pSceneMgr->createEntity("floor", "FloorPlane"); ent->setMaterialName("Test/Floor"); Ogre::SceneNode *node = m_pSceneMgr->getRootSceneNode()->createChildSceneNode(); node->attachObject(ent); btTransform Transform; Transform.setIdentity(); // Fixed the transform vector here for y back to 0 to stop the objects sinking into the ground. Transform.setOrigin(btVector3(0,0,0)); // Give it to the motion state btDefaultMotionState *MotionState = new btDefaultMotionState(Transform); btCollisionShape *Shape = new btStaticPlaneShape(btVector3(0,1,0),0); // Add Mass btVector3 LocalInertia; Shape->calculateLocalInertia(0, LocalInertia); // CReate the rigid body object btRigidBody *RigidBody = new btRigidBody(0, MotionState, Shape, LocalInertia); // Store a pointer to the Ogre Node so we can update it later RigidBody->setUserPointer((void *) (node)); // Add it to the physics world World->addRigidBody(RigidBody); Objects.push_back(RigidBody); m_pNumEntities++; // End Physics } void GameState::CreateBox(const btVector3 &TPosition, const btVector3 &TScale, btScalar TMass) { Ogre::Vector3 size = Ogre::Vector3::ZERO; Ogre::Vector3 pos = Ogre::Vector3::ZERO; Ogre::Vector3 scale = Ogre::Vector3::ZERO; pos.x = TPosition.getX(); pos.y = TPosition.getY(); pos.z = TPosition.getZ(); scale.x = TScale.getX(); scale.y = TScale.getY(); scale.z = TScale.getZ(); Ogre::Entity *entity = m_pSceneMgr->createEntity( "Box" + Ogre::StringConverter::toString(m_pNumEntities), "cube.mesh"); entity->setCastShadows(true); Ogre::AxisAlignedBox boundingB = entity->getBoundingBox(); // The ogre bounding box is slightly bigger so I am reducing it for // use with the rigid body. size = boundingB.getSize()*0.95f; entity->setMaterialName("Test/Cube"); Ogre::SceneNode *node = m_pSceneMgr->getRootSceneNode()->createChildSceneNode(); node->attachObject(entity); node->setPosition(pos); node->showBoundingBox(true); //node->scale(scale); // Physics btTransform Transform; Transform.setIdentity(); Transform.setOrigin(TPosition); // Give it to the motion state btDefaultMotionState *MotionState = new btDefaultMotionState(Transform); // I got the size of the bounding box above but wasn't using it to set // the size for the rigid body. This now does. btVector3 HalfExtents(size.x*0.5f,size.y*0.5f,size.z*0.5f); btCollisionShape *Shape = new btBoxShape(HalfExtents); // Add Mass btVector3 LocalInertia; Shape->calculateLocalInertia(TMass, LocalInertia); // CReate the rigid body object btRigidBody *RigidBody = new btRigidBody(TMass, MotionState, Shape, LocalInertia); // Store a pointer to the Ogre Node so we can update it later RigidBody->setUserPointer((void *) (node)); // Add it to the physics world World->addRigidBody(RigidBody); Objects.push_back(RigidBody); m_pNumEntities++; } void GameState::UpdatePhysics(unsigned int TDeltaTime) { World->stepSimulation(TDeltaTime * 0.001f, 60); btRigidBody *TObject; for(std::vector<btRigidBody *>::iterator it = Objects.begin(); it != Objects.end(); ++it) { // Update renderer Ogre::SceneNode *node = static_cast<Ogre::SceneNode *>((*it)->getUserPointer()); TObject = *it; // Set position btVector3 Point = TObject->getCenterOfMassPosition(); node->setPosition(Ogre::Vector3((float)Point[0], (float)Point[1], (float)Point[2])); // Convert the bullet Quaternion to an Ogre quaternion btQuaternion btq = TObject->getOrientation(); Ogre::Quaternion quart = Ogre::Quaternion(btq.w(),btq.x(),btq.y(),btq.z()); // use the quaternion with setOrientation node->setOrientation(quart); } } The QuaternionToEuler function isn't needed so that was removed from code and header files. The objects now collide with the ground and each other appropriately.

    Read the article

  • Stairway to SQL Server Indexes: Step 1, Introduction to Indexes

    Indexes are the database objects that enable SQL Server to satisfy each data access request from a client application with the minimum amount of effort, resulting in the maximum performance of individual requests while also reducing the impact of one request upon another. Prerequisites: Familiarity with the following relational database concepts: Table, row, primary key, foreign key Join SQL Backup’s 35,000+ customers to compress and strengthen your backups "SQL Backup will be a REAL boost to any DBA lucky enough to use it." Jonathan Allen. Download a free trial now.

    Read the article

  • Investigation: Can different combinations of components effect Dataflow performance?

    - by jamiet
    Introduction The Dataflow task is one of the core components (if not the core component) of SQL Server Integration Services (SSIS) and often the most misunderstood. This is not surprising, its an incredibly complicated beast and we’re abstracted away from that complexity via some boxes that go yellow red or green and that have some lines drawn between them. Example dataflow In this blog post I intend to look under that facade and get into some of the nuts and bolts of the Dataflow Task by investigating how the decisions we make when building our packages can affect performance. I will do this by comparing the performance of three dataflows that all have the same input, all produce the same output, but which all operate slightly differently by way of having different transformation components. I also want to use this blog post to challenge a common held opinion that I see perpetuated over and over again on the SSIS forum. That is, that people assume adding components to a dataflow will be detrimental to overall performance. Its not surprising that people think this –it is intuitive to think that more components means more work- however this is not a view that I share. I have always been of the opinion that there are many factors affecting dataflow duration and the number of components is actually one of the less important ones; having said that I have never proven that assertion and that is one reason for this investigation. I have actually seen evidence that some people think dataflow duration is simply a function of number of rows and number of components. I’ll happily call that one out as a myth even without any investigation!  The Setup I have a 2GB datafile which is a list of 4731904 (~4.7million) customer records with various attributes against them and it contains 2 columns that I am going to use for categorisation: [YearlyIncome] [BirthDate] The data file is a SSIS raw format file which I chose to use because it is the quickest way of getting data into a dataflow and given that I am testing the transformations, not the source or destination adapters, I want to minimise external influences as much as possible. In the test I will split the customers according to month of birth (12 of those) and whether or not their yearly income is above or below 50000 (2 of those); in other words I will be splitting them into 24 discrete categories and in order to do it I shall be using different combinations of SSIS’ Conditional Split and Derived Column transformation components. The 24 datapaths that occur will each input to a rowcount component, again because this is the least resource intensive means of terminating a datapath. The test is being carried out on a Dell XPS Studio laptop with a quad core (8 logical Procs) Intel Core i7 at 1.73GHz and Samsung SSD hard drive. Its running SQL Server 2008 R2 on Windows 7. The Variables Here are the three combinations of components that I am going to test:     One Conditional Split - A single Conditional Split component CSPL Split by Month of Birth and income category that will use expressions on [YearlyIncome] & [BirthDate] to send each row to one of 24 outputs. This next screenshot displays the expression logic in use: Derived Column & Conditional Split - A Derived Column component DER Income Category that adds a new column [IncomeCategory] which will contain one of two possible text values {“LessThan50000”,”GreaterThan50000”} and uses [YearlyIncome] to determine which value each row should get. A Conditional Split component CSPL Split by Month of Birth and Income Category then uses that new column in conjunction with [BirthDate] to determine which of the same 24 outputs to send each row to. Put more simply, I am separating the Conditional Split of #1 into a Derived Column and a Conditional Split. The next screenshots display the expression logic in use: DER Income Category         CSPL Split by Month of Birth and Income Category       Three Conditional Splits - A Conditional Split component that produces two outputs based on [YearlyIncome], one for each Income Category. Each of those outputs will go to a further Conditional Split that splits the input into 12 outputs, one for each month of birth (identical logic in each). In this case then I am separating the single Conditional Split of #1 into three Conditional Split components. The next screenshots display the expression logic in use: CSPL Split by Income Category         CSPL Split by Month of Birth 1& 2       Each of these combinations will provide an input to one of the 24 rowcount components, just the same as before. For illustration here is a screenshot of the dataflow containing three Conditional Split components: As you can these dataflows have a fair bit of work to do and remember that they’re doing that work for 4.7million rows. I will execute each dataflow 10 times and use the average for comparison. I foresee three possible outcomes: The dataflow containing just one Conditional Split (i.e. #1) will be quicker There is no significant difference between any of them One of the two dataflows containing multiple transformation components will be quicker Regardless of which of those outcomes come to pass we will have learnt something and that makes this an interesting test to carry out. Note that I will be executing the dataflows using dtexec.exe rather than hitting F5 within BIDS. The Results and Analysis The table below shows all of the executions, 10 for each dataflow. It also shows the average for each along with a standard deviation. All durations are in seconds. I’m pasting a screenshot because I frankly can’t be bothered with the faffing about needed to make a presentable HTML table. It is plain to see from the average that the dataflow containing three conditional splits is significantly faster, the other two taking 43% and 52% longer respectively. This seems strange though, right? Why does the dataflow containing the most components outperform the other two by such a big margin? The answer is actually quite logical when you put some thought into it and I’ll explain that below. Before progressing, a side note. The standard deviation for the “Three Conditional Splits” dataflow is orders of magnitude smaller – indicating that performance for this dataflow can be predicted with much greater confidence too. The Explanation I refer you to the screenshot above that shows how CSPL Split by Month of Birth and salary category in the first dataflow is setup. Observe that there is a case for each combination of Month Of Date and Income Category – 24 in total. These expressions get evaluated in the order that they appear and hence if we assume that Month of Date and Income Category are uniformly distributed in the dataset we can deduce that the expected number of expression evaluations for each row is 12.5 i.e. 1 (the minimum) + 24 (the maximum) divided by 2 = 12.5. Now take a look at the screenshots for the second dataflow. We are doing one expression evaluation in DER Income Category and we have the same 24 cases in CSPL Split by Month of Birth and Income Category as we had before, only the expression differs slightly. In this case then we have 1 + 12.5 = 13.5 expected evaluations for each row – that would account for the slightly longer average execution time for this dataflow. Now onto the third dataflow, the quick one. CSPL Split by Income Category does a maximum of 2 expression evaluations thus the expected number of evaluations per row is 1.5. CSPL Split by Month of Birth 1 & CSPL Split by Month of Birth 2 both have less work to do than the previous Conditional Split components because they only have 12 cases to test for thus the expected number of expression evaluations is 6.5 There are two of them so total expected number of expression evaluations for this dataflow is 6.5 + 6.5 + 1.5 = 14.5. 14.5 is still more than 12.5 & 13.5 though so why is the third dataflow so much quicker? Simple, the conditional expressions in the first two dataflows have two boolean predicates to evaluate – one for Income Category and one for Month of Birth; the expressions in the Conditional Split in the third dataflow however only have one predicate thus they are doing a lot less work. To sum up, the difference in execution times can be attributed to the difference between: MONTH(BirthDate) == 1 && YearlyIncome <= 50000 and MONTH(BirthDate) == 1 In the first two dataflows YearlyIncome <= 50000 gets evaluated an average of 12.5 times for every row whereas in the third dataflow it is evaluated once and once only. Multiply those 11.5 extra operations by 4.7million rows and you get a significant amount of extra CPU cycles – that’s where our duration difference comes from. The Wrap-up The obvious point here is that adding new components to a dataflow isn’t necessarily going to make it go any slower, moreover you may be able to achieve significant improvements by splitting logic over multiple components rather than one. Performance tuning is all about reducing the amount of work that needs to be done and that doesn’t necessarily mean use less components, indeed sometimes you may be able to reduce workload in ways that aren’t immediately obvious as I think I have proven here. Of course there are many variables in play here and your mileage will most definitely vary. I encourage you to download the package and see if you get similar results – let me know in the comments. The package contains all three dataflows plus a fourth dataflow that will create the 2GB raw file for you (you will also need the [AdventureWorksDW2008] sample database from which to source the data); simply disable all dataflows except the one you want to test before executing the package and remember, execute using dtexec, not within BIDS. If you want to explore dataflow performance tuning in more detail then here are some links you might want to check out: Inequality joins, Asynchronous transformations and Lookups Destination Adapter Comparison Don’t turn the dataflow into a cursor SSIS Dataflow – Designing for performance (webinar) Any comments? Let me know! @Jamiet

    Read the article

< Previous Page | 205 206 207 208 209 210 211 212 213 214 215 216  | Next Page >