Search Results

Search found 30097 results on 1204 pages for 'jenson button event'.

Page 1/1204 | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >

  • Super Joybox 5 HID 0925:8884 not recognized as joystick in Ubuntu 12.04 LTS

    - by Tim Evans
    Problem: When using the "Super JoyBox 5" 4 port playstation 2 to USB adapter, the device is not recognized as a joystick. there is no js0 created, but instead another input eventX and mouseX are created in /dev/input. When using the directional buttons (up down left right) on a Playstation 1 controller attached to the device, the mouse cursor moves to the top, bottom, left, and right edges of the screen respectively. Buttons are unresponsive. The joypads attached to the device cannot be used in any games or other programs. Attempted remedies: Creating a symlink from the eventX to js0 does not solve the problem. Addl Info: joydev is loaded and running peroperly according to LSMOD. evtest can be run on the created eventX (sudo evtest /dev/input/event14 in my case) and the buttons and axes all register inputs. Here is a paste of EVTEST's diagnostic and the first couple button events. [code] sudo evtest /dev/input/event14 Input driver version is 1.0.1 Input device ID: bus 0x3 vendor 0x925 product 0x8884 version 0x100 Input device name: "HID 0925:8884" Supported events: Event type 0 (EV_SYN) Event type 1 (EV_KEY) Event code 288 (BTN_TRIGGER) Event code 289 (BTN_THUMB) Event code 290 (BTN_THUMB2) Event code 291 (BTN_TOP) Event code 292 (BTN_TOP2) Event code 293 (BTN_PINKIE) Event code 294 (BTN_BASE) Event code 295 (BTN_BASE2) Event code 296 (BTN_BASE3) Event code 297 (BTN_BASE4) Event code 298 (BTN_BASE5) Event code 299 (BTN_BASE6) Event code 300 (?) Event code 301 (?) Event code 302 (?) Event code 303 (BTN_DEAD) Event code 304 (BTN_A) Event code 305 (BTN_B) Event code 306 (BTN_C) Event code 307 (BTN_X) Event code 308 (BTN_Y) Event code 309 (BTN_Z) Event code 310 (BTN_TL) Event code 311 (BTN_TR) Event code 312 (BTN_TL2) Event code 313 (BTN_TR2) Event code 314 (BTN_SELECT) Event code 315 (BTN_START) Event code 316 (BTN_MODE) Event code 317 (BTN_THUMBL) Event code 318 (BTN_THUMBR) Event code 319 (?) Event code 320 (BTN_TOOL_PEN) Event code 321 (BTN_TOOL_RUBBER) Event code 322 (BTN_TOOL_BRUSH) Event code 323 (BTN_TOOL_PENCIL) Event code 324 (BTN_TOOL_AIRBRUSH) Event code 325 (BTN_TOOL_FINGER) Event code 326 (BTN_TOOL_MOUSE) Event code 327 (BTN_TOOL_LENS) Event code 328 (?) Event code 329 (?) Event code 330 (BTN_TOUCH) Event code 331 (BTN_STYLUS) Event code 332 (BTN_STYLUS2) Event code 333 (BTN_TOOL_DOUBLETAP) Event code 334 (BTN_TOOL_TRIPLETAP) Event code 335 (BTN_TOOL_QUADTAP) Event type 3 (EV_ABS) Event code 0 (ABS_X) Value 127 Min 0 Max 255 Flat 15 Event code 1 (ABS_Y) Value 127 Min 0 Max 255 Flat 15 Event code 2 (ABS_Z) Value 127 Min 0 Max 255 Flat 15 Event code 3 (ABS_RX) Value 127 Min 0 Max 255 Flat 15 Event code 4 (ABS_RY) Value 127 Min 0 Max 255 Flat 15 Event code 5 (ABS_RZ) Value 127 Min 0 Max 255 Flat 15 Event code 6 (ABS_THROTTLE) Value 127 Min 0 Max 255 Flat 15 Event code 7 (ABS_RUDDER) Value 127 Min 0 Max 255 Flat 15 Event code 8 (ABS_WHEEL) Value 127 Min 0 Max 255 Flat 15 Event code 9 (ABS_GAS) Value 127 Min 0 Max 255 Flat 15 Event code 10 (ABS_BRAKE) Value 127 Min 0 Max 255 Flat 15 Event code 11 (?) Value 127 Min 0 Max 255 Flat 15 Event code 12 (?) Value 127 Min 0 Max 255 Flat 15 Event code 13 (?) Value 127 Min 0 Max 255 Flat 15 Event code 14 (?) Value 127 Min 0 Max 255 Flat 15 Event code 15 (?) Value 127 Min 0 Max 255 Flat 15 Event code 16 (ABS_HAT0X) Value 0 Min -1 Max 1 Event code 17 (ABS_HAT0Y) Value 0 Min -1 Max 1 Event code 18 (ABS_HAT1X) Value 0 Min -1 Max 1 Event code 19 (ABS_HAT1Y) Value 0 Min -1 Max 1 Event code 20 (ABS_HAT2X) Value 0 Min -1 Max 1 Event code 21 (ABS_HAT2Y) Value 0 Min -1 Max 1 Event code 22 (ABS_HAT3X) Value 0 Min -1 Max 1 Event code 23 (ABS_HAT3Y) Value 0 Min -1 Max 1 Event type 4 (EV_MSC) Event code 4 (MSC_SCAN) Testing ... (interrupt to exit) Event: time 1351223176.126127, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90001 Event: time 1351223176.126130, type 1 (EV_KEY), code 288 (BTN_TRIGGER), value 1 Event: time 1351223176.126166, -------------- SYN_REPORT ------------ Event: time 1351223178.238127, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90001 Event: time 1351223178.238130, type 1 (EV_KEY), code 288 (BTN_TRIGGER), value 0 Event: time 1351223178.238167, -------------- SYN_REPORT ------------ Event: time 1351223180.422127, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90002 Event: time 1351223180.422129, type 1 (EV_KEY), code 289 (BTN_THUMB), value 1 Event: time 1351223180.422163, -------------- SYN_REPORT ------------ Event: time 1351223181.558099, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90002 Event: time 1351223181.558102, type 1 (EV_KEY), code 289 (BTN_THUMB), value 0 Event: time 1351223181.558137, -------------- SYN_REPORT ------------ Event: time 1351223182.486137, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90003 Event: time 1351223182.486140, type 1 (EV_KEY), code 290 (BTN_THUMB2), value 1 Event: time 1351223182.486172, -------------- SYN_REPORT ------------ Event: time 1351223183.302130, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90003 Event: time 1351223183.302132, type 1 (EV_KEY), code 290 (BTN_THUMB2), value 0 Event: time 1351223183.302165, -------------- SYN_REPORT ------------ Event: time 1351223184.030133, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90004 Event: time 1351223184.030136, type 1 (EV_KEY), code 291 (BTN_TOP), value 1 Event: time 1351223184.030166, -------------- SYN_REPORT ------------ Event: time 1351223184.558135, type 4 (EV_MSC), code 4 (MSC_SCAN), value 90004 Event: time 1351223184.558138, type 1 (EV_KEY), code 291 (BTN_TOP), value 0 Event: time 1351223184.558168, -------------- SYN_REPORT ------------ [/code] The directional buttons on the pad are being identified as HAT0Y and HAT0X axes, thats zero, not the letter O. Aparently, this device used to work flawlessly on kernel 2.4.x systems, and even as late as ubunto 10.04. Perhaps the Joydev rules for identifying joypads has changed? Currently, this kind of bug is affecting a few different type of controller adapters, but since this is the one that i PERSONALLY have (and has been driving me my own special brand of crazy), its the one im documenting. What i think should be happening instead: The device should be registering js0 through js3, one for each port, or JS0 that will handle all of the connected devices with different numbered axes for each connected joypad. Either way, it should work as a joystick and stop controlling the mouse cursor. Please help!

    Read the article

  • Have Button re-appear immediately after clicking button in ListView row

    - by Soeren
    I have 4 buttons on a page. Each button opens a modal window and let’s the user input data in a form. When the user hits the save button in the modal, a ListView appears on the page with the submitted data. The button the user clicked to open the modal window is set to visible=false, so it’s gone when the row is added to the ListView. Now there are 3 buttons and the same goes for those; when the user hits a button, a modal appears, and when the modal form is submitted, the button disappears and a row is added to the ListView. In the ListView row, there is a delete button. When this button is clicked, the row is deleted and the button that was initially clicked to add this row (and open the modal), SHOULD reappear, but it doesn’t. The row disappears, but I have to refresh the page before the button comes back. There is a ScriptManager on the masterpage, so I guess this is an AJAX partial refresh issue. I tried adding different events, but I can’t find the one that fires at the right time. I use an ObjectDataSource to fill the ListView, and the data comes from a database, wrapped in a business object. This code loads a business object in a List< and checks if the user inserted an item of a specific type. If he did, the button he used to open the modal is hidden. This works fine (maybe not the most elegant) _goals = GoalManager.GetGoalsByUser(UserID); if (_goals != null) { foreach (Goal _goalinlist in _goals) { if (_goalinlist.GoalType == 1) { Button1.Visible = false; goalid1 = true; } if (_goalinlist.GoalType == 2) { Button2.Visible = false; goalid2 = true; } if (_goalinlist.GoalType == 3) { Button3.Visible = false; goalid3 = true; } if (_goalinlist.GoalType == 4) { Button4.Visible = false; goalid4 = true; } } } As you can see, I tried setting a boolean, and then check it when the page is re-loaded. But the problem (I guess) is that the whole page isn't refreshed when the delete button is clicked in the ListView. This is the delete button in the ListView: <asp:ImageButton ID="ImageButton2" runat="server" CommandName="Delete" CausesValidation="false" ToolTip="Delete" CommandArgument='<%#Eval("GoalID")%>' ImageUrl="delete.gif" OnClientClick="return confirm('Delete this post?');" CssClass="button"/> I guess the question is, how do I make the button re-appear right after the ListView button is clicked?

    Read the article

  • In Javascript event handling, why "return false" or "event.preventDefault()" and "stopping the event

    - by Jian Lin
    It is said that when we handle a "click event", returning false or calling event.preventDefault() makes a difference, in which the difference is that preventDefault will only prevent the default event action to occur, i.e. a page redirect on a link click, a form submission, etc. and return false will also stop the event flow. Does that mean, if the click event is registered several times for several actions, using $('#clickme').click(function() { … }) returning false will stop the other handlers from running? I am on a Mac now and so can only use Firefox and Chrome but not IE, which has a different event model, and tested it on FF and Chrome and all 3 handlers ran without any stopping…. so what is the real difference, or, is there a situation where "stopping the event flow" is not desirable? this is related to http://stackoverflow.com/questions/3042036/using-jquerys-animate-if-the-clicked-on-element-is-a-href-a and http://stackoverflow.com/questions/2017755/whats-the-difference-between-e-preventdefault-and-return-false

    Read the article

  • Query specific logs from event log using nxlog

    - by user170899
    Below is my nxlog configuration define ROOT C:\Program Files (x86)\nxlog Moduledir %ROOT%\modules CacheDir %ROOT%\data Pidfile %ROOT%\data\nxlog.pid SpoolDir %ROOT%\data LogFile %ROOT%\data\nxlog.log <Extension json> Module xm_json </Extension> <Input internal> Module im_internal </Input> <Input eventlog> Module im_msvistalog Query <QueryList>\ <Query Id="0">\ <Select Path="Security">*</Select>\ </Query>\ </QueryList> </Input> <Output out> Module om_tcp Host localhost Port 3515 Exec $EventReceivedTime = integer($EventReceivedTime) / 1000000; \ to_json(); </Output> <Route 1> Path eventlog, internal => out </Route> <Select Path="Security">*</Select>\ - * gets everything from the Security log, but my requirement is to get specific logs starting with EventId - 4663. How do i do this? Please help. Thanks.

    Read the article

  • Attached event triggering if event is same as current event

    - by Psytronic
    I have a button which start/stops a timer. When I click it, I remove the current event handler, and attach the opposite one. Ie, click "Stop", and the "Start" function is attached for the next click. However in IE (7), not sure about 8. The newly attached event is also triggering after the original function is finished: Lifeline in IE7: Button pushed -- "Stop" function begins -- "Stop" function removed -- "Start" function attached -- "Stop" function finishes -- "Start" function is called. I know it's to do with the attaching event bit, because If I remove it, the Start event doesn't fire. This seems a bit odd, as I have the same situation in the "Start" function, ie, "Start" removed, "Stop" called. But this doesn't seem to cause an infinite loop, thankfully. if(btn.attachEvent){ btn.detachEvent("onclick", stop); btn.attachEvent("onclick", start); }else if(btn.addEventListener){ btn.removeEventListener("click", stop, true); btn.addEventListener("click", start , true); } It all works fine in FF and Chrome. I have switched the order of attach/detach, just to see if it makes a difference, but it doesn't. Addendum Sorry, and no answers involving jQuery, Protoype et al. Its not something I want to integrate for this project.

    Read the article

  • Call a function when "event.GetFrom(m_cpVoice)==S_OK" (ergo when event happens) [SAPI 5.1 and C++]

    - by Jesuskiewicz
    Hello, I'm doing a project with a 3D model, that speaks. So, I'm using SAPI 5.1, and I want to call a function asynchronously when there's a Viseme event (in order to play the animation related to). How could I do it? Thank you very much. Note: I use : hRes = m_cpVoice-Speak(L"All I want is to solve this problem", SPF_ASYNC , NULL); And I know the CspEvent, event.eEventId . All I want is how to call a function when Sapi event happens

    Read the article

  • How can I change mouse keymapping

    - by zuberuber
    I have Razer DeathAdder(left handed edition) and A4Tech wireless mouse. My problem is I don't know how to change wireless mouse keymapping(swaping left/right click). Can somebody guide me how to do such thing? List of my devices: ? Virtual core pointer id=2 [master pointer (3)] ? ? Virtual core XTEST pointer id=4 [slave pointer (2)] ? ? Logitech Unifying Device. Wireless PID:4004 id=8 [slave pointer (2)] ? ? Razer Razer DeathAdder id=11 [slave pointer (2)] ? ? A4TECH USB Device id=12 [slave pointer (2)] ? ? A4TECH USB Device id=13 [slave pointer (2)] ? Virtual core keyboard id=3 [master keyboard (2)] ? Virtual core XTEST keyboard id=5 [slave keyboard (3)] ? Power Button id=6 [slave keyboard (3)] ? Power Button id=7 [slave keyboard (3)] ? Logitech USB Keyboard id=9 [slave keyboard (3)] ? Logitech USB Keyboard id=10 [slave keyboard (3)] This is my Razer xinput: Device 'Razer Razer DeathAdder': Device Enabled (121): 1 Coordinate Transformation Matrix (123): 1.000000, 0.000000, 0.000000, 0.000000, 1.000000, 0.000000, 0.000000, 0.000000, 1.000000 Device Accel Profile (246): 0 Device Accel Constant Deceleration (247): 5.000000 Device Accel Adaptive Deceleration (248): 1.000000 Device Accel Velocity Scaling (249): 10.000000 Device Product ID (240): 5426, 22 Device Node (241): "/dev/input/event4" Evdev Axis Inversion (250): 0, 0 Evdev Axes Swap (252): 0 Axis Labels (253): "Rel X" (131), "Rel Y" (132), "Rel Vert Wheel" (274) Button Labels (254): "Button Left" (124), "Button Middle" (125), "Button Right" (126), "Button Wheel Up" (127), "Button Wheel Down" (128), "Button Horiz Wheel Left" (129), "Button Horiz Wheel Right" (130), "Button Side" (269), "Button Extra" (270), "Button Forward" (271), "Button Back" (272), "Button Task" (273), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243) Evdev Middle Button Emulation (255): 0 Evdev Middle Button Timeout (256): 50 Evdev Third Button Emulation (257): 0 Evdev Third Button Emulation Timeout (258): 1000 Evdev Third Button Emulation Button (259): 3 Evdev Third Button Emulation Threshold (260): 20 Evdev Wheel Emulation (261): 0 Evdev Wheel Emulation Axes (262): 0, 0, 4, 5 Evdev Wheel Emulation Inertia (263): 10 Evdev Wheel Emulation Timeout (264): 200 Evdev Wheel Emulation Button (265): 4 Evdev Drag Lock Buttons (266): 0 And this is my wireless mouse xinput: Device 'A4TECH USB Device': Device Enabled (121): 1 Coordinate Transformation Matrix (123): 1.000000, 0.000000, 0.000000, 0.000000, 1.000000, 0.000000, 0.000000, 0.000000, 1.000000 Device Accel Profile (246): 0 Device Accel Constant Deceleration (247): 1.000000 Device Accel Adaptive Deceleration (248): 1.000000 Device Accel Velocity Scaling (249): 10.000000 Device Product ID (240): 2522, 1359 Device Node (241): "/dev/input/event16" Evdev Axis Inversion (250): 0, 0 Evdev Axes Swap (252): 0 Axis Labels (253): "Rel X" (131), "Rel Y" (132), "Rel Horiz Wheel" (245), "Rel Vert Wheel" (274) Button Labels (254): "Button Left" (124), "Button Middle" (125), "Button Right" (126), "Button Wheel Up" (127), "Button Wheel Down" (128), "Button Horiz Wheel Left" (129), "Button Horiz Wheel Right" (130), "Button Side" (269), "Button Extra" (270), "Button Forward" (271), "Button Back" (272), "Button Task" (273), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243), "Button Unknown" (243) Evdev Middle Button Emulation (255): 0 Evdev Middle Button Timeout (256): 50 Evdev Third Button Emulation (257): 0 Evdev Third Button Emulation Timeout (258): 1000 Evdev Third Button Emulation Button (259): 3 Evdev Third Button Emulation Threshold (260): 20 Evdev Wheel Emulation (261): 0 Evdev Wheel Emulation Axes (262): 0, 0, 4, 5 Evdev Wheel Emulation Inertia (263): 10 Evdev Wheel Emulation Timeout (264): 200 Evdev Wheel Emulation Button (265): 4 Evdev Drag Lock Buttons (266): 0

    Read the article

  • Reassign the Right button functionality of touchpad

    - by Guandalino
    My notebook has a touchpad with two buttons. The touchpad works pretty well and I use it for vertical/horizontal scroll, two fingers scrolling etc. The Left button acts like the left button of a mouse. The Right button too, when clicked it makes appear the classic contextual menu (create new folder, move to trash etc...). The problem is that Right button doesn't work anymore for mechanical reasons, so pressing it has no effect. To recover the missing functionality, I'd like that pressing CTRL + Left button I get the same thing of pressing the Right button alone. Other combinations (CTRL + SHIFT + Left button) are also acceptable. Any way to do that?

    Read the article

  • SQLAuthority News – Book Signing Event – SQLPASS 2011 Event Log

    - by pinaldave
    I have been dreaming of writing book for really long time, and I finally got the chance – in fact, two chances!  I recently wrote two books: SQL Programming Joes 2 Pros: Programming and Development for Microsoft SQL Server 2008 [Amazon] | [Flipkart] | [Kindle] and SQL Wait Stats Joes 2 Pros: SQL Performance Tuning Techniques Using Wait Statistics, Types & Queues [Amazon] | [Flipkart] | [Kindle].  I had a lot of fun writing these two books, even though sometimes I had to sacrifice some family time and time for other personal development to write the books. The good side of writing book is that when the efforts put in writing books are recognize by books readers and kind organizations like expressor studio. Book Signing Event Book writing is a complex process.  Even after you spend months, maybe years, writing the material you still have to go through the editing and fact checking processes.  And, once the book is out there, there is no way to take back all the copies to change mistakes or add something you forgot.  Most of the time it is a one-way street. Book Signing Event Just like every author, I had a dream that after the books were written, they would be loved by people and gain acceptance by an audience. My first book, SQL Programming Joes 2 Pros: Programming and Development for Microsoft SQL Server 2008, is extremely popular because it helps lots of people learn various fundamental topics. My second book covers beginning to learn SQL Server Wait Stats, which is a relatively new subject. This book has had very good acceptance in the community. Book Signing Event Helping my community is my primary focus, so I was happy to see this year’s SQLPASS tag line: ‘This is a Community.‘ At the event, the expressor studio guys came up with a very novel idea. They had previously used my books and they had found them very useful. They got 100 copies of the book and decided to give it away to community folks. They invited me and my co-author Rick Morelan to hold a book signing event. We did a book signing on Thursday between 1 pm and 2 pm. Book Signing Event This event was one of the best events for me. This was my first book signing event outside of India. I reached the book signing location around 20 minutes before the scheduled time and what I saw was a big line for the book signing event. I felt very honored looking at the crowd and all the people around the event location. I felt very humbled when I saw some of my very close friends standing in the line to get my signature. It was really heartwarming to see so many enthusiasts waiting for more than an hour to get my signature. While standing in line I had the chance to have a conversation with every single person who showed up for the signature. I made sure that I repeated every single name and wrote it in every book with my signature. There is saying that if we write a name once we will remember it forever. I want to remember all of you who saw me at the book signing. Your comments were wonderful, your feedback was amazing and you were all very supportive. Book Signing Event I have made a note of every conversation I had with all of you when I was signing the books. Once again, I just want to express my thanks for coming to my book signing event. The whole experience was very humbling. On the top of it, I want to thank the expressor studio people who made it possible, who organized the whole signing event. I am so thankful to them for facilitating the whole experience, which is going to be hard to beat by any future experience. My books Reference: Pinal Dave (http://blog.SQLAuthority.com) Filed under: Pinal Dave, PostADay, SQL, SQL Authority, SQL PASS, SQL Query, SQL Server, SQL Tips and Tricks, SQLAuthority Author Visit, SQLAuthority News, T SQL, Technology

    Read the article

  • User is trying to leave! Set at-least confirm alert on browser(tab) close event!!

    - by kaushalparik27
    This is something that might be annoying or irritating for end user. Obviously, It's impossible to prevent end user from closing the/any browser. Just think of this if it becomes possible!!!. That will be a horrible web world where everytime you will be attacked by sites and they will not allow to close your browser until you confirm your shopping cart and do the payment. LOL:) You need to open the task manager and might have to kill the running browser exe processes.Anyways; Jokes apart, but I have one situation where I need to alert/confirm from the user in any anyway when they try to close the browser or change the url. Think of this: You are creating a single page intranet asp.net application where your employee can enter/select their TDS/Investment Declarations and you wish to at-least ALERT/CONFIRM them if they are attempting to:[1] Close the Browser[2] Close the Browser Tab[3] Attempt to go some other site by Changing the urlwithout completing/freezing their declaration.So, Finally requirement is clear. I need to alert/confirm the user what he is going to do on above bulleted events. I am going to use window.onbeforeunload event to set the javascript confirm alert box to appear.    <script language="JavaScript" type="text/javascript">        window.onbeforeunload = confirmExit;        function confirmExit() {            return "You are about to exit the system before freezing your declaration! If you leave now and never return to freeze your declaration; then they will not go into effect and you may lose tax deduction, Are you sure you want to leave now?";        }    </script>See! you are halfway done!. So, every time browser unloads the page, above confirm alert causes to appear on front of user like below:By saying here "every time browser unloads the page"; I mean to say that whenever page loads or postback happens the browser onbeforeunload event will be executed. So, event a button submit or a link submit which causes page to postback would tend to execute the browser onbeforeunload event to fire!So, now the hurdle is how can we prevent the alert "Not to show when page is being postback" via any button/link submit? Answer is JQuery :)Idea is, you just need to set the script reference src to jQuery library and Set the window.onbeforeunload event to null when any input/link causes a page to postback.Below will be the complete code:<head runat="server">    <title></title>    <script src="jquery.min.js" type="text/javascript"></script>    <script language="JavaScript" type="text/javascript">        window.onbeforeunload = confirmExit;        function confirmExit() {            return "You are about to exit the system before freezing your declaration! If you leave now and never return to freeze your declaration; then they will not go into effect and you may lose tax deduction, Are you sure you want to leave now?";        }        $(function() {            $("a").click(function() {                window.onbeforeunload = null;            });            $("input").click(function() {                window.onbeforeunload = null;            });        });    </script></head><body>    <form id="form1" runat="server">    <div></div>    </form></body></html>So, By this post I have tried to set the confirm alert if user try to close the browser/tab or try leave the site by changing the url. I have attached a working example with this post here. I hope someone might find it helpful.

    Read the article

  • Javascript Back Button - Stop the initial load of back button from working

    - by Evan
    Hi, I'm using a javascript back button link and forward button link to control the user's history inside a modal/lightbox window. The challenge I have is when the modal window is launched, and the "back" and "forward" buttons are present for the user to click, if the initial javascript back button is clicked when the window opens, it actually closes the modal window, because the javascript history is taking the user back to the page PRIOR to the opening of modal window. So, in essence, I'm trying to disable the "back" button from working on the initial load of the modal/light box. <a href="javascript:history.go(-1)">Back Button</a> <a href="javascript:history.go(1)">Foward Button</a> Is this possible?

    Read the article

  • custom button: properties change, button should be redrawn

    - by Rutger
    Hi I'm developing an Android app. I have class derived from button to represent a special type of button. This special type has some properties (integers) and according to these one or more circles have to be drawn on top of the button. So I overrode the onDraw function, which looks the values up and accordingly draws the circles. But the class has a function to set new values for its properties. So new values are set but the changes are not reflected in the UI. It seems like the onDraw function is not called. When later I click the button or show a pop up message above my interface the onDraw function is called and the button is drawn correctly. So my question: when changing the properties how can I say that the button has to be redrawn? Thanks a lot!

    Read the article

  • Button (using <button> tag) does not display in Opera

    - by Tanveer Dewan
    On my website I am using the html 'button' tag but the button does not display when page is opened in Opera version 12.15. Works fine in chrome, firefox and IE. If you go to my cdn test site CDN Comparison u see cloudflare, incapsula and page speed images. below the three cdn images there's a button which does not display. It's supposted to be right above the 'Test and monitor image load speed of the top three Free CDN Service providers' sentence. please help. <button id="button" onclick="changeLink()">Test Another Image</button> I have not added any style to this tag using CSS

    Read the article

  • Event sourcing: Write event before or after updating the model

    - by Magnus
    I'm reasoning about event sourcing and often I arrive at a chicken and egg problem. Would be grateful for some hints on how to reason around this. If I execute all I/O-bound processing async (ie writing to the event log) then how do I handle, or sometimes even detect, failures? I'm using Akka Actors so processing is sequential for each event/message. I do not have any database at this time, instead I would persist all the events in an event log and then keep an aggregated state of all the events in a model stored in memory. Queries are all against this model, you can consider it to be a cache. Example Creating a new user: Validate that the user does not exist in model Persist event to journal Update model (in memory) If step 3 breaks I still have persisted my event so I can replay it at a later date. If step 2 breaks I can handle that as well gracefully. This is fine, but since step 2 is I/O-bound I figured that I should do I/O in a separate actor to free up the first actor for queries: Updating a user while allowing queries (A0 = Front end/GUI actor, A1 = Processor Actor, A2 = IO-actor, E = event bus). (A0-E-A1) Event is published to update user 'U1'. Validate that the user 'U1' exists in model (A1-A2) Persist event to journal (separate actor) (A0-E-A1-A0) Query for user 'U1' profile (A2-A1) Event is now persisted continue to update model (A0-E-A1-A0) Query for user 'U1' profile (now returns fresh data) This is appealing since queries can be processed while I/O-is churning along at it's own pace. But now I can cause myself all kinds of problems where I could have two incompatible commands (delete and then update) be persisted to the event log and crash on me when replayed up at a later date, since I do the validation before persisting the event and then update the model. My aim is to have a simple reasoning around my model (since Actor processes messages sequentially single threaded) but not be waiting for I/O-bound updates when Querying. I get the feeling I'm modeling a database which in itself is might be a problem. If things are unclear please write a comment.

    Read the article

  • Backward compatibility with event-sourcing

    - by Tomas Jansson
    How do you stay backward compatible with event-sourcing? Let say you release a version that has one kind of event, let call it X. You know how to handle that event in all the systems that extracts the events from the event source. In a later release you make a change to event X or delete it, how do you stay backward compatible with that? To have a fully functional system you need to be able to handle the old event as the same time as you need to handle the updated version. Or if you delete that event type, then you will be stuck with code that is only there to handle legacy events which in my head can be a little bit messy in the long run.

    Read the article

  • Disable the asp button on click of the same button till the process ends

    - by Rupa
    Hi, I need help in the following scenario: On click of a SAVE button in aspx page,some processing will take place. The processing can be of 10-15 minutes.I want to disable the SAVE button till process ends and once process completes, the SAVE button should be enabled. I tried the following but couldnot solve the problem: 1)Client Side Scripting: I called a java script on client clik of button and set the disable property to true.How ever, server side saveButton_Click event is not called because of disabling the button on client side. 2)Disabling in saveButton_Click method on server side: In this case, as the call of saveButton_Click already happend and requested is sent to server,the client side is as it without any difference until the request is processed. Please help me in this asap. Its very URGENT. Thanks Rupa

    Read the article

  • Doing two Declarative Operations with One Button

    - by shay.shmeltzer
    You can file the below video under "things that get asked on OTN a lot". With ADF it is very easy to drag an operation to a page to create a button that activate it. But what if you want a single button to invoke two operations? For example have a button that does a "Delete" as well as a "Commit". The way to do it is to add an action binding, and then overwrite the button function in a backing bean to call the additional action. The nice thing is that JDeveloper will create all the binding code for you in the backing bean - all you need to do is duplicate it. Here is a quick demo:

    Read the article

  • Text limit on analytics event code

    - by Theo G
    I am just about to add the event code a button that downloads the pdf. Event code fields: _trackEvent(category, action, opt_label, opt_value, opt_noninteraction) Example of event code: onClick="_gaq.push(['_trackEvent', 'Videos', 'Play', 'Baby\'s First Birthday']);" I was just wondering if anyone knows if there is a text limit on the opt_value? Do you think the following would be too long 'Elmhurst School says IPC has made all the difference'?

    Read the article

  • Intercepting touch events on activity and button on Android

    - by hgpc
    I have an Android activity with an ImageButton. I would like to execute some logic when the button is clicked and show a different image for the pressed state, but also receive the touch event on the activity. By default only the button receives the touch event. If I set the clickable attribute of the button to false then only the activity receives the touch event. What's the best way to receive the touch event in both the activity and the button?

    Read the article

  • wxPython - Running built in event prior to custom bound event

    - by jhaukur
    I'm trying to figure out how I can specify that the mouse_down event in wxPython (StyledTextCtrl) is first handled by the built in event listener, which changes the caret position, and then handled by my own custom event handler. To bind the custom event handler I use wx.EVT_LEFT_DOWN(self.styCtrl, self.OnMouseClick) def OnMouseClick(self, evt): evt.Skip() foo() I want the built in event handler to fire and complete prior to foo().

    Read the article

  • Cutom event dispatchment location

    - by Martino Wullems
    Hello, I've been looking into custom event (listeners) for quite some time, but never succeeded in making one. There are so many different mehods, extending the Event class, but also Extending the EventDispatcher class, very confusing! I want to settle with this once and for all and learn the appriopate technique. package{ import flash.events.Event; public class CustomEvent extends Event{ public static const TEST:String = 'test'; //what exac is the purpose of the value in the string? public var data:Object; public function CustomEvent(type:String, bubbles:Boolean = false, cancelable:Boolean = false, data:Object = null):void { this.data = data; super(); } } } As far as I know a custom class where you set the requirements for the event to be dispatched has to be made: package { import flash.display.MovieClip; public class TestClass extends MovieClip { public function TestClass():void { if (ConditionForHoldToComplete == true) { dispatchEvent(new Event(CustomEvent.TEST)); } } } } I'm not sure if this is correct, but it should be something along the lines of this. Now What I want is something like a mouseevent, which can be applied to a target and does not require a specific class. It would have to work something like this: package com.op_pad._events{ import flash.events.MouseEvent; import flash.utils.Timer; import flash.events.TimerEvent; import flash.events.EventDispatcher; import flash.events.Event; public class HoldEvent extends Event { public static const HOLD_COMPLETE:String = "hold completed"; var timer:Timer; public function SpriteEvent(type:String, bubbles:Boolean=true, cancelable:Boolean=false) { super( type, bubbles, cancelable ); timer = new Timer(1000, 1); //somehow find the target where is event is placed upon -> target.addEventlistener target.addEventListener(MouseEvent.MOUSE_DOWN, startTimer); target.addEventListener(MouseEvent.MOUSE_UP, stopTimer); } public override function clone():Event { return new SpriteEvent(type, bubbles, cancelable); } public override function toString():String { return formatToString("MovieEvent", "type", "bubbles", "cancelable", "eventPhase"); } ////////////////////////////////// ///// c o n d i t i o n s ///// ////////////////////////////////// private function startTimer(e:MouseEvent):void { timer.start(); timer.addEventListener(TimerEvent.TIMER_COMPLETE, complete); } private function stopTimer(e:MouseEvent):void { timer.stop() } public function complete(e:TimerEvent):void { dispatchEvent(new HoldEvent(HoldEvent.HOLD_COMPLETE)); } } } This obviously won't work, but should give you an idea of what I want to achieve. This should be possible because mouseevent can be applied to about everything.The main problem is that I don't know where I should set the requirements for the event to be executed to be able to apply it to movieclips and sprites. Thanks in advance

    Read the article

  • Option Trading: Getting the most out of the event session options

    - by extended_events
    You can control different aspects of how an event session behaves by setting the event session options as part of the CREATE EVENT SESSION DDL. The default settings for the event session options are designed to handle most of the common event collection situations so I generally recommend that you just use the defaults. Like everything in the real world though, there are going to be a handful of “special cases” that require something different. This post focuses on identifying the special cases and the correct use of the options to accommodate those cases. There is a reason it’s called Default The default session options specify a total event buffer size of 4 MB with a 30 second latency. Translating this into human terms; this means that our default behavior is that the system will start processing events from the event buffer when we reach about 1.3 MB of events or after 30 seconds, which ever comes first. Aside: What’s up with the 1.3 MB, I thought you said the buffer was 4 MB?The Extended Events engine takes the total buffer size specified by MAX_MEMORY (4MB by default) and divides it into 3 equally sized buffers. This is done so that a session can be publishing events to one buffer while other buffers are being processed. There are always at least three buffers; how to get more than three is covered later. Using this configuration, the Extended Events engine can “keep up” with most event sessions on standard workloads. Why is this? The fact is that most events are small, really small; on the order of a couple hundred bytes. Even when you start considering events that carry dynamically sized data (eg. binary, text, etc.) or adding actions that collect additional data, the total size of the event is still likely to be pretty small. This means that each buffer can likely hold thousands of events before it has to be processed. When the event buffers are finally processed there is an economy of scale achieved since most targets support bulk processing of the events so they are processed at the buffer level rather than the individual event level. When all this is working together it’s more likely that a full buffer will be processed and put back into the ready queue before the remaining buffers (remember, there are at least three) are full. I know what you’re going to say: “My server is exceptional! My workload is so massive it defies categorization!” OK, maybe you weren’t going to say that exactly, but you were probably thinking it. The point is that there are situations that won’t be covered by the Default, but that’s a good place to start and this post assumes you’ve started there so that you have something to look at in order to determine if you do have a special case that needs different settings. So let’s get to the special cases… What event just fired?! How about now?! Now?! If you believe the commercial adage from Heinz Ketchup (Heinz Slow Good Ketchup ad on You Tube), some things are worth the wait. This is not a belief held by most DBAs, particularly DBAs who are looking for an answer to a troubleshooting question fast. If you’re one of these anxious DBAs, or maybe just a Program Manager doing a demo, then 30 seconds might be longer than you’re comfortable waiting. If you find yourself in this situation then consider changing the MAX_DISPATCH_LATENCY option for your event session. This option will force the event buffers to be processed based on your time schedule. This option only makes sense for the asynchronous targets since those are the ones where we allow events to build up in the event buffer – if you’re using one of the synchronous targets this option isn’t relevant. Avoid forgotten events by increasing your memory Have you ever had one of those days where you keep forgetting things? That can happen in Extended Events too; we call it dropped events. In order to optimizes for server performance and help ensure that the Extended Events doesn’t block the server if to drop events that can’t be published to a buffer because the buffer is full. You can determine if events are being dropped from a session by querying the dm_xe_sessions DMV and looking at the dropped_event_count field. Aside: Should you care if you’re dropping events?Maybe not – think about why you’re collecting data in the first place and whether you’re really going to miss a few dropped events. For example, if you’re collecting query duration stats over thousands of executions of a query it won’t make a huge difference to miss a couple executions. Use your best judgment. If you find that your session is dropping events it means that the event buffer is not large enough to handle the volume of events that are being published. There are two ways to address this problem. First, you could collect fewer events – examine you session to see if you are over collecting. Do you need all the actions you’ve specified? Could you apply a predicate to be more specific about when you fire the event? Assuming the session is defined correctly, the next option is to change the MAX_MEMORY option to a larger number. Picking the right event buffer size might take some trial and error, but a good place to start is with the number of dropped events compared to the number you’ve collected. Aside: There are three different behaviors for dropping events that you specify using the EVENT_RETENTION_MODE option. The default is to allow single event loss and you should stick with this setting since it is the best choice for keeping the impact on server performance low.You’ll be tempted to use the setting to not lose any events (NO_EVENT_LOSS) – resist this urge since it can result in blocking on the server. If you’re worried that you’re losing events you should be increasing your event buffer memory as described in this section. Some events are too big to fail A less common reason for dropping an event is when an event is so large that it can’t fit into the event buffer. Even though most events are going to be small, you might find a condition that occasionally generates a very large event. You can determine if your session is dropping large events by looking at the dm_xe_sessions DMV once again, this time check the largest_event_dropped_size. If this value is larger than the size of your event buffer [remember, the size of your event buffer, by default, is max_memory / 3] then you need a large event buffer. To specify a large event buffer you set the MAX_EVENT_SIZE option to a value large enough to fit the largest event dropped based on data from the DMV. When you set this option the Extended Events engine will create two buffers of this size to accommodate these large events. As an added bonus (no extra charge) the large event buffer will also be used to store normal events in the cases where the normal event buffers are all full and waiting to be processed. (Note: This is just a side-effect, not the intended use. If you’re dropping many normal events then you should increase your normal event buffer size.) Partitioning: moving your events to a sub-division Earlier I alluded to the fact that you can configure your event session to use more than the standard three event buffers – this is called partitioning and is controlled by the MEMORY_PARTITION_MODE option. The result of setting this option is fairly easy to explain, but knowing when to use it is a bit more art than science. First the science… You can configure partitioning in three ways: None, Per NUMA Node & Per CPU. This specifies the location where sets of event buffers are created with fairly obvious implication. There are rules we follow for sub-dividing the total memory (specified by MAX_MEMORY) between all the event buffers that are specific to the mode used: None: 3 buffers (fixed)Node: 3 * number_of_nodesCPU: 2.5 * number_of_cpus Here are some examples of what this means for different Node/CPU counts: Configuration None Node CPU 2 CPUs, 1 Node 3 buffers 3 buffers 5 buffers 6 CPUs, 2 Node 3 buffers 6 buffers 15 buffers 40 CPUs, 5 Nodes 3 buffers 15 buffers 100 buffers   Aside: Buffer size on multi-processor computersAs the number of Nodes or CPUs increases, the size of the event buffer gets smaller because the total memory is sub-divided into more pieces. The defaults will hold up to this for a while since each buffer set is holding events only from the Node or CPU that it is associated with, but at some point the buffers will get too small and you’ll either see events being dropped or you’ll get an error when you create your session because you’re below the minimum buffer size. Increase the MAX_MEMORY setting to an appropriate number for the configuration. The most likely reason to start partitioning is going to be related to performance. If you notice that running an event session is impacting the performance of your server beyond a reasonably expected level [Yes, there is a reasonably expected level of work required to collect events.] then partitioning might be an answer. Before you partition you might want to check a few other things: Is your event retention set to NO_EVENT_LOSS and causing blocking? (I told you not to do this.) Consider changing your event loss mode or increasing memory. Are you over collecting and causing more work than necessary? Consider adding predicates to events or removing unnecessary events and actions from your session. Are you writing the file target to the same slow disk that you use for TempDB and your other high activity databases? <kidding> <not really> It’s always worth considering the end to end picture – if you’re writing events to a file you can be impacted by I/O, network; all the usual stuff. Assuming you’ve ruled out the obvious (and not so obvious) issues, there are performance conditions that will be addressed by partitioning. For example, it’s possible to have a successful event session (eg. no dropped events) but still see a performance impact because you have many CPUs all attempting to write to the same free buffer and having to wait in line to finish their work. This is a case where partitioning would relieve the contention between the different CPUs and likely reduce the performance impact cause by the event session. There is no DMV you can check to find these conditions – sorry – that’s where the art comes in. This is  largely a matter of experimentation. On the bright side you probably won’t need to to worry about this level of detail all that often. The performance impact of Extended Events is significantly lower than what you may be used to with SQL Trace. You will likely only care about the impact if you are trying to set up a long running event session that will be part of your everyday workload – sessions used for short term troubleshooting will likely fall into the “reasonably expected impact” category. Hey buddy – I think you forgot something OK, there are two options I didn’t cover: STARTUP_STATE & TRACK_CAUSALITY. If you want your event sessions to start automatically when the server starts, set the STARTUP_STATE option to ON. (Now there is only one option I didn’t cover.) I’m going to leave causality for another post since it’s not really related to session behavior, it’s more about event analysis. - Mike Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

  • Option Trading: Getting the most out of the event session options

    - by extended_events
    You can control different aspects of how an event session behaves by setting the event session options as part of the CREATE EVENT SESSION DDL. The default settings for the event session options are designed to handle most of the common event collection situations so I generally recommend that you just use the defaults. Like everything in the real world though, there are going to be a handful of “special cases” that require something different. This post focuses on identifying the special cases and the correct use of the options to accommodate those cases. There is a reason it’s called Default The default session options specify a total event buffer size of 4 MB with a 30 second latency. Translating this into human terms; this means that our default behavior is that the system will start processing events from the event buffer when we reach about 1.3 MB of events or after 30 seconds, which ever comes first. Aside: What’s up with the 1.3 MB, I thought you said the buffer was 4 MB?The Extended Events engine takes the total buffer size specified by MAX_MEMORY (4MB by default) and divides it into 3 equally sized buffers. This is done so that a session can be publishing events to one buffer while other buffers are being processed. There are always at least three buffers; how to get more than three is covered later. Using this configuration, the Extended Events engine can “keep up” with most event sessions on standard workloads. Why is this? The fact is that most events are small, really small; on the order of a couple hundred bytes. Even when you start considering events that carry dynamically sized data (eg. binary, text, etc.) or adding actions that collect additional data, the total size of the event is still likely to be pretty small. This means that each buffer can likely hold thousands of events before it has to be processed. When the event buffers are finally processed there is an economy of scale achieved since most targets support bulk processing of the events so they are processed at the buffer level rather than the individual event level. When all this is working together it’s more likely that a full buffer will be processed and put back into the ready queue before the remaining buffers (remember, there are at least three) are full. I know what you’re going to say: “My server is exceptional! My workload is so massive it defies categorization!” OK, maybe you weren’t going to say that exactly, but you were probably thinking it. The point is that there are situations that won’t be covered by the Default, but that’s a good place to start and this post assumes you’ve started there so that you have something to look at in order to determine if you do have a special case that needs different settings. So let’s get to the special cases… What event just fired?! How about now?! Now?! If you believe the commercial adage from Heinz Ketchup (Heinz Slow Good Ketchup ad on You Tube), some things are worth the wait. This is not a belief held by most DBAs, particularly DBAs who are looking for an answer to a troubleshooting question fast. If you’re one of these anxious DBAs, or maybe just a Program Manager doing a demo, then 30 seconds might be longer than you’re comfortable waiting. If you find yourself in this situation then consider changing the MAX_DISPATCH_LATENCY option for your event session. This option will force the event buffers to be processed based on your time schedule. This option only makes sense for the asynchronous targets since those are the ones where we allow events to build up in the event buffer – if you’re using one of the synchronous targets this option isn’t relevant. Avoid forgotten events by increasing your memory Have you ever had one of those days where you keep forgetting things? That can happen in Extended Events too; we call it dropped events. In order to optimizes for server performance and help ensure that the Extended Events doesn’t block the server if to drop events that can’t be published to a buffer because the buffer is full. You can determine if events are being dropped from a session by querying the dm_xe_sessions DMV and looking at the dropped_event_count field. Aside: Should you care if you’re dropping events?Maybe not – think about why you’re collecting data in the first place and whether you’re really going to miss a few dropped events. For example, if you’re collecting query duration stats over thousands of executions of a query it won’t make a huge difference to miss a couple executions. Use your best judgment. If you find that your session is dropping events it means that the event buffer is not large enough to handle the volume of events that are being published. There are two ways to address this problem. First, you could collect fewer events – examine you session to see if you are over collecting. Do you need all the actions you’ve specified? Could you apply a predicate to be more specific about when you fire the event? Assuming the session is defined correctly, the next option is to change the MAX_MEMORY option to a larger number. Picking the right event buffer size might take some trial and error, but a good place to start is with the number of dropped events compared to the number you’ve collected. Aside: There are three different behaviors for dropping events that you specify using the EVENT_RETENTION_MODE option. The default is to allow single event loss and you should stick with this setting since it is the best choice for keeping the impact on server performance low.You’ll be tempted to use the setting to not lose any events (NO_EVENT_LOSS) – resist this urge since it can result in blocking on the server. If you’re worried that you’re losing events you should be increasing your event buffer memory as described in this section. Some events are too big to fail A less common reason for dropping an event is when an event is so large that it can’t fit into the event buffer. Even though most events are going to be small, you might find a condition that occasionally generates a very large event. You can determine if your session is dropping large events by looking at the dm_xe_sessions DMV once again, this time check the largest_event_dropped_size. If this value is larger than the size of your event buffer [remember, the size of your event buffer, by default, is max_memory / 3] then you need a large event buffer. To specify a large event buffer you set the MAX_EVENT_SIZE option to a value large enough to fit the largest event dropped based on data from the DMV. When you set this option the Extended Events engine will create two buffers of this size to accommodate these large events. As an added bonus (no extra charge) the large event buffer will also be used to store normal events in the cases where the normal event buffers are all full and waiting to be processed. (Note: This is just a side-effect, not the intended use. If you’re dropping many normal events then you should increase your normal event buffer size.) Partitioning: moving your events to a sub-division Earlier I alluded to the fact that you can configure your event session to use more than the standard three event buffers – this is called partitioning and is controlled by the MEMORY_PARTITION_MODE option. The result of setting this option is fairly easy to explain, but knowing when to use it is a bit more art than science. First the science… You can configure partitioning in three ways: None, Per NUMA Node & Per CPU. This specifies the location where sets of event buffers are created with fairly obvious implication. There are rules we follow for sub-dividing the total memory (specified by MAX_MEMORY) between all the event buffers that are specific to the mode used: None: 3 buffers (fixed)Node: 3 * number_of_nodesCPU: 2.5 * number_of_cpus Here are some examples of what this means for different Node/CPU counts: Configuration None Node CPU 2 CPUs, 1 Node 3 buffers 3 buffers 5 buffers 6 CPUs, 2 Node 3 buffers 6 buffers 15 buffers 40 CPUs, 5 Nodes 3 buffers 15 buffers 100 buffers   Aside: Buffer size on multi-processor computersAs the number of Nodes or CPUs increases, the size of the event buffer gets smaller because the total memory is sub-divided into more pieces. The defaults will hold up to this for a while since each buffer set is holding events only from the Node or CPU that it is associated with, but at some point the buffers will get too small and you’ll either see events being dropped or you’ll get an error when you create your session because you’re below the minimum buffer size. Increase the MAX_MEMORY setting to an appropriate number for the configuration. The most likely reason to start partitioning is going to be related to performance. If you notice that running an event session is impacting the performance of your server beyond a reasonably expected level [Yes, there is a reasonably expected level of work required to collect events.] then partitioning might be an answer. Before you partition you might want to check a few other things: Is your event retention set to NO_EVENT_LOSS and causing blocking? (I told you not to do this.) Consider changing your event loss mode or increasing memory. Are you over collecting and causing more work than necessary? Consider adding predicates to events or removing unnecessary events and actions from your session. Are you writing the file target to the same slow disk that you use for TempDB and your other high activity databases? <kidding> <not really> It’s always worth considering the end to end picture – if you’re writing events to a file you can be impacted by I/O, network; all the usual stuff. Assuming you’ve ruled out the obvious (and not so obvious) issues, there are performance conditions that will be addressed by partitioning. For example, it’s possible to have a successful event session (eg. no dropped events) but still see a performance impact because you have many CPUs all attempting to write to the same free buffer and having to wait in line to finish their work. This is a case where partitioning would relieve the contention between the different CPUs and likely reduce the performance impact cause by the event session. There is no DMV you can check to find these conditions – sorry – that’s where the art comes in. This is  largely a matter of experimentation. On the bright side you probably won’t need to to worry about this level of detail all that often. The performance impact of Extended Events is significantly lower than what you may be used to with SQL Trace. You will likely only care about the impact if you are trying to set up a long running event session that will be part of your everyday workload – sessions used for short term troubleshooting will likely fall into the “reasonably expected impact” category. Hey buddy – I think you forgot something OK, there are two options I didn’t cover: STARTUP_STATE & TRACK_CAUSALITY. If you want your event sessions to start automatically when the server starts, set the STARTUP_STATE option to ON. (Now there is only one option I didn’t cover.) I’m going to leave causality for another post since it’s not really related to session behavior, it’s more about event analysis. - Mike Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!

    Read the article

  • Python ttk.Button -command, runs without button being pressed

    - by Markus Kothe
    I'm making a small script in python with ttk and I have a problem where a function runs where it shouldn't. The button code looks as follows: btReload = ttk.Button(treeBottomUI, text="Reload", width=17, command=loadModelTree(treeModel)) btReload.pack(side="left") and the function is as this: def loadModelTree(tree): print ("Loading models...") allModels = os.listdir(confModPath) for chunk in allModels: ... For some reason, the function runs without the button being pressed. Why?

    Read the article

  • Event ID: 861 - The Windows Firewall has detected an application listening for incoming traffic

    - by Chris Marisic
    Firstly, my machines aren't compromised any person suggesting such will be DV'd. The security logs on some of my networks client machines (all Windows Xp Sp3) get filled with these useless error messages. Security Failure Audit Detailed Tracking Event ID: 861 User: NT AUTHORITY\NETWORK SERVICE The Windows Firewall has detected an application listening for incoming traffic. Name: - Path: C:\WINDOWS\system32\svchost.exe Process identifier: 976 User account: NETWORK SERVICE User domain: NT AUTHORITY Service: Yes RPC server: No IP version: IPv4 IP protocol: UDP Port number: 55035 Allowed: No User notified: No It's always on various random ports of UDP so setting up a port exception isn't really an option. It's always from svchost or lsass both of which are running services from DLLs. One of the most offending processes seems to the be DnsCache. I have in my global policy under AT < Network < Network Connection < Widnows Firewall < Domain Profile (I haven't changed any standard profile options do both need configured? To allow remote administration and desktop exceptions and have a custom program exception list that has %SystemRoot%\system32\svchost.exe:*:enabled:svchost (Windows won't allow you to add this exception on a local machine but it let me have it on here in the global policy it just doesn't seem to do anything) %SystemRoot%\system32\lsass.exe:*enabled:lsass (I think this one ended all of my LSASS messages) %SystemRoot%\system32\dnsrslvr.dll:*:enabled:dnscache (I tried adding the dll itself to the exception list, this didn't seem to do anything) Is there really any other options left other than disabling the Windows Firewall entirely, disabling auditing entirely or just changing the event viewer to just auto overwrite when needed? I'd much rather fix the problem and get rid of these entries ever being created instead of just trying to cover up the problem.

    Read the article

1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >