Search Results

Search found 341 results on 14 pages for 'overwriting'.

Page 13/14 | < Previous Page | 9 10 11 12 13 14  | Next Page >

  • xterm not wrapping text properly

    - by mulllhausen
    I'm configuring both my gnome-terminal and xterm columns (i still haven't picked which of these I will be using) and I have a couple of issues I would like to fix: the typing area seems to be smaller (fewer columns) than the display area the typed text is not wrapping to the next line when it reaches the end - it just continues back around on the same line, overwriting the prompt (i have set a custom bash prompt with PS1 in case this is relevant) $ lsb_release -a No LSB modules are available. Distributor ID: Debian Description: Debian GNU/Linux 7.1 (wheezy) Release: 7.1 Codename: wheezy $ echo $TERM xterm $ stty -a [peter@pc ~] $ stty -a speed 38400 baud; rows 52; columns 126; line = 0; intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = M-^?; eol2 = M-^?; swtch = M-^?; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R; werase = ^W; lnext = ^V; flush = ^O; min = 1; time = 0; -parenb -parodd cs8 hupcl -cstopb cread -clocal -crtscts -ignbrk brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon -ixoff -iuclc ixany imaxbel iutf8 opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0 isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop -echoprt echoctl echoke $[peter@mine ~] $ # the column width only goes up to here ------------------------------------------------> the results are identical in both the xterm and in gnome-terminal 3.4.1.1 and as you can see, the output of the stty -a command goes right up to the edge of the screen, while the typing does not go that far. I have found that I can get the desired result by setting the columns to a very large number, eg: $ stty cols 1800 this fixes both problems. Is it the right way to go about solving this problem? Will this "break" any of the output from programs? So far I have tried top and stty -a and these seem OK. more info as requested in the comments i found that if i cat some input into a file then the columns actually strech the full width of the terminal window: [peter@mine applications] $ cat > /tmp/asd aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaasssssssssssssssssssssssssssssssssssssssssssssssssssssssssssqqqqqqqqqqqqqqqq qqqq does this imply that it is actually bash that is restricting the number of columns and not the terminal? if so then how to alter the number of columns in bash?

    Read the article

  • Clonezilla restore from Samba - no 'restoredisk' option

    - by MT_Head
    I used a CloneZilla LiveCD to back up a couple of Windows machines to a Samba share. Now I'm trying to restore those images, and CloneZilla won't even give me the 'restoredisk' or 'restorepart' options on the menu. I'm guessing that this is because CZ isn't recognizing a valid image... but why? Here's a listing of the folder on the Samba share: -rwxrwxrwx 1 marc users 319 May 31 03:45 blkdev.list -rwxrwxrwx 1 marc users 5307 May 31 04:41 clonezilla-img -rwxrwxrwx 1 marc users 4 May 31 04:31 disk -rwxrwxrwx 1 marc users 16091 May 31 04:31 Info-dmi.txt -rwxrwxrwx 1 marc users 11029 May 31 04:31 Info-lshw.txt -rwxrwxrwx 1 marc users 1502 May 31 04:31 Info-lspci.txt -rwxrwxrwx 1 marc users 170 May 31 04:31 Info-packages.txt -rwxrwxrwx 1 marc users 80 May 31 04:41 Info-saved-by-cmd.txt -rwxrwxrwx 1 marc users 10 May 31 04:31 parts -rwxrwxrwx 1 marc users 2097152000 May 31 04:06 sda1.ntfs-ptcl-img.gz.aa -rwxrwxrwx 1 marc users 247361656 May 31 04:08 sda1.ntfs-ptcl-img.gz.ab -rwxrwxrwx 1 marc users 823182034 May 31 04:31 sda2.ntfs-ptcl-img.gz.aa -rwxrwxrwx 1 marc users 36 May 31 03:45 sda-chs.sf -rwxrwxrwx 1 marc users 31744 May 31 03:45 sda-hidden-data-after-mbr -rwxrwxrwx 1 marc users 512 May 31 03:45 sda-mbr -rwxrwxrwx 1 marc users 315 May 31 03:45 sda-pt.parted -rwxrwxrwx 1 marc users 285 May 31 03:45 sda-pt.parted.compact -rwxrwxrwx 1 marc users 259 May 31 03:45 sda-pt.sf (I've been experimenting with various permissions trying to get this to work; that's why they're currently all "rwxrwxrwx"...) I've got my CZ LiveCD stuck in a (different) machine with a 160GB SATA disk that I'm fine with overwriting; although CZ doesn't show a directory listing, it does show that the correct folder is mounted as /home/partimag. But a moment later, after selecting either Beginner or Expert, I'm only presented with the "savedisk", "saveparts", and "exit" options. What am I doing wrong? I am confident that the initial backup was successful; I can post the log if desired, or any other information that might be germane. Edit: I've copied the contents of the folder onto a 16GB USB stick and set THAT as /home/partimag. Still nothing. What the hell is CZ looking for?

    Read the article

  • Running an rsync sweep before initializing lsyncd for synchronizing instances on EC2

    - by chrisallenlane
    My company uses several EC2 servers that will scale up and down according to the load we're receiving on our sites at any given moment. For the sake of our discussion here, we're running four instances: master.ourdomain.com - the file syncing "hub" of the webservers www1/www2/www3.ourdomain.com - three webservers which turn on or off as dictated by load I'm using lsyncd to keep all of the webservers in sync, and for the most part, it's working quite well. We're using a two-way syncing scheme, such that each webserver syncs against master, and master syncs against each webserver. Thus, the webservers are kept in sync, even though they aren't syncing against each other directly. I'm having one problem that I'm having a hard time solving,though. It occurs under these circumstances: When changes are made on master (perhaps after we've pushed new code), while some of the redundant webservers are sleeping And then a sleeping webserver wakes-up to absorb load Under that circumstance, I would like the following to happen: First, the newly-awoken webserver should sync its file structure - one way - against master, to bring its web application code up-to-date. Then, and only then, should it begin pushing changes in its file structure back to master. Unfortunately, currently, when a sleeping server is started, when lsyncd starts up, it pushes changes back to master before updating its own codebase, thus overwriting new code with old. Thus, before lsyncd starts, I'd like to be able to synchronize the webservers code against master's, perhaps by running a simple one-way rsync against the two machines. We're running lsyncd v.2, and I've tried to make this happen by using the "bash" configuration options documented in the lsyncd manual. My configuration file looks like this: settings = { logfile = "/home/user/log/lsyncd/log.txt", statusFile = "/home/user/log/lsyncd/status.txt", maxProcesses = 2, nodaemon = false, } bash = { onStartup = "rsync [email protected]:/home/user/www /home/user/www" } sync{ default.rsyncssh, source="/home/user/www/", host="[email protected]", targetdir="/home/user/www/", rsyncOpts="-ltus", excludeFrom="/home/user/conf/lsyncd/exclude" } (I've obviously redacted that file somewhat to protect the identities of the guilty.) Simply put, though, this just isn't working. How else might I approach this problem? I was looking at the --delete-after option in man rsync, but I don't think that does what I'm looking for. Are there any suggestions about how I should approach this problem? Thanks for lending your time and expertise. Chris

    Read the article

  • How to prevent dual booted OSes from damaging each other?

    - by user1252434
    For better compatibility and performance in games I'm thinking about installing Windows additionally to Linux. I have security concerns about this, though. Note: "Windows" in the remaining text includes not only the OS but also any software running on it. Regardless of whether it comes included or is additionally installed, whether it is started intentionally or unintentionally (virus, malware). Is there an easy way to achieve the following requirements: Windows MUST NOT be able to kill my linux partition or my data disk neither single files (virus infection) nor overwriting the whole disk Windows MUST NOT be able to read data disk (- extra protection against spyware) Linux may or may not have access to the windows partition both Linux and Windows should have full access to the graphics card this rules out desktop VM solutions for gaming I want the manufacturer's windows graphics card driver Regarding Windows to be unable to destroy my linux install: this is not just the usual paranoia, that has happened to me in the past. So I don't accept "no ext4 driver" as an argument. Once bitten, twice shy. And even if destruction targeted at specific (linux) files is nearly impossible, there should be no way to shred the whole partition. I may accept the risk of malware breaking out of a barrier (e.g. VM) around the whole windows box, though. Currently I have a system disk (SSD) and a data disk (HDD), both SATA. I expect I have to add another disk. If i don't: even better. My CPU is a Intel Core i5, with VT-x and VT-d available, though untested. Ideas I've had so far: deactivate or hide other HDs until reboot at low level possible? can the boot loader (grub) do this for me? tiny VM layer: load windows in a VM that provides access to almost all hardware, except the HDs any ready made software solution for this? Preferably free. as I said: the main problem seems to be to provide full access to the graphics card hardware switch to cut power to disks commercial products expensive and lots of warnings against cheap home built solutions preferably all three hard disks with one switch (one push) mobile racks - won't wear of daily swapping be a problem?

    Read the article

  • Edit ePub eBooks with Your Favorite HTML Editor

    - by Matthew Guay
    ePub eBooks are increasingly popular today, but often they’ve been made by converting other file formats. Here’s how you can edit ePub books to remove irregularities and make them better for reading on your devices. ePub’s are actually a zip file containing images, XHTML files with your text, and more with the .epub extension. You can make them better by editing the XHTML files directly.  Code gurus can edit the code directly, but even if you’ve never edited HTML, you can still quickly make changes with a WYSIWYG editor. Extract the Files from your ePub eBook As mentioned before, ePub files are actually renamed zip files.  So first let’s get all of the files in your ePub eBook accessible.  Find an eBook you want to edit and then change the file extension to .zip. If you don’t see the file extensions, click Organize in the menu bar and select Folder and search options. Select the View tab, and then uncheck the box beside Hide extensions for known file types.  Click Ok, and then change the file type as above. Windows will warn you about changing the file type; click Yes to proceed. Now you can browse the files of the ePub file.  Notice that it contains mostly HTML or XHTML files and images.  Click Extract all files to save them all in a folder so you can easily edit them. Alternately, you can open the ePub file directly in your favorite file archival program such as 7-zip.  Browse to the location of your ePub file, double-click it, and it’ll automatically open even if you don’t change the file extension to zip.  Now you can extract the folder, or extract individual files as before.   Edit Your eBook in KompoZer The actual ebook contents are stored in HTML or XHTML files.  These may be stored on the top folder of you ePub file’s directory, or they may be stored in \OEBPS\text in the file. To change the contents of your eBook, you’ll want to edit these files.  Often there may be separate files for each chapter, so you may have to use trial and error to find the one you need to edit.  You could edit them by hand in Windows using Notepad if you don’t have an HTML editor installed. A better option would be to use an HTML editor.  Here we’ll use the free KompoZer program to edit the files just like we’d edit a document in Word. Download KompoZer (link below), and unzip the files.  Then open the new folder and launch kompozer.exe; you don’t even need to install it.  In fact, you could even store KompoZer on a flash drive so you could edit HTML files from any computer. In KompoZer, open the HTML or XHTML file from your eBook that you want to edit. Now you can edit the file just like you would edit a document in Word.  Remove extra and unneeded text, make titles stand out, correct misspellings … anything you want!  This is especially helpful if your ePub file was created by converting a PDF as these often have many small errors. Or, if you’d rather edit the code itself, select the Source tab and edit as you wish. When you’re done making the changes, make sure to save the file in the same location with the same file name. Recreate Your Edited ePub eBook Once you’ve made all the changes you wanted, it’s time to turn this folder of files back into ePub.  Make sure you change the name of the folder if it still has the same name as the original ePub or zip file so you don’t mix them up or have trouble with overwriting the old files. Zip the folder using Windows Explorer or your favorite archival utility.  If you are using another archival program, make sure to compress it as a zip folder; other compression methods will render the ePub unreadable by your eReader app. Now change the file extension again, this time back to .epub. Now you can read your eBook with your changes in your favorite reader program or app on your mobile device. Conclusion Whether you need to remove an odd, misplaced character or need to do fine editing, using an HTML editor is a great way to make your ePub eBooks look just like you want.  Also, with an editor like KompoZer it’s not even difficult. Download KompoZer Similar Articles Productive Geek Tips Change the Default Editor From Nano on Ubuntu LinuxConvert a PDF eBook to ePub Format for Your iPad, iPhone, or eReaderRead Mobi eBooks on Kindle for PCEdit Your Firefox Bookmarks Easier with Flat Bookmark EditorChange the Default Editor for Batch Files in Vista TouchFreeze Alternative in AutoHotkey The Icy Undertow Desktop Windows Home Server – Backup to LAN The Clear & Clean Desktop Use This Bookmarklet to Easily Get Albums Use AutoHotkey to Assign a Hotkey to a Specific Window Latest Software Reviews Tinyhacker Random Tips HippoRemote Pro 2.2 Xobni Plus for Outlook All My Movies 5.9 CloudBerry Online Backup 1.5 for Windows Home Server XPS file format & XPS Viewer Explained Microsoft Office Web Apps Guide Know if Someone Accessed Your Facebook Account Shop for Music with Windows Media Player 12 Access Free Documentaries at BBC Documentaries Rent Cameras In Bulk At CameraRenter

    Read the article

  • MySQL Connector/Net 6.6.3 Beta 2 has been released

    - by fernando
    MySQL Connector/Net 6.6.3, a new version of the all-managed .NET driver for MySQL has been released.  This is the second of two beta releases intended to introduce users to the new features in the release. This release is feature complete it should be stable enough for users to understand the new features and how we expect them to work.  As is the case with all non-GA releases, it should not be used in any production environment.  It is appropriate for use with MySQL server versions 5.0-5.6. It is now available in source and binary form from http://dev.mysql.com/downloads/connector/net/#downloads and mirror sites (note that not all mirror sites may be up to date at this point-if you can't find this version on some mirror, please try again later or choose another download site.) The 6.6 version of MySQL Connector/Net brings the following new features:   * Stored routine debugging   * Entity Framework 4.3 Code First support   * Pluggable authentication (now third parties can plug new authentications mechanisms into the driver).   * Full Visual Studio 2012 support: everything from Server Explorer to Intellisense&   the Stored Routine debugger. Stored Procedure Debugging ------------------------------------------- We are very excited to introduce stored procedure debugging into our Visual Studio integration.  It works in a very intuitive manner by simply clicking 'Debug Routine' from Server Explorer. You can debug stored routines, functions&   triggers. These release contains fixes specific of the debugger as well as other fixes specific of other areas of Connector/NET:   * Added feature to define initial values for InOut stored procedure arguments.   * Debugger: Fixed Visual Studio locked connection after debugging a routine.   * Fix for bug Cannot Create an Entity with a Key of Type String (MySQL bug #65289, Oracle bug #14540202).   * Fix for bug "CacheServerProperties can cause 'Packet too large' error". MySQL Bug #66578 Orabug #14593547.   * Fix for handling unnamed parameter in MySQLCommand. This fix allows the mysqlcommand to handle parameters without requiring naming (e.g. INSERT INTO Test (id,name) VALUES (?, ?) ) (MySQL Bug #66060, Oracle bug #14499549).   * Fixed end of line issue when debugging a routine.   * Added validation to avoid overwriting a routine backup file when it hasn't changed.   * Fixed inheritance on Entity Framework Code First scenarios. (MySql bug #63920 and Oracle bug #13582335).   * Fixed "Trying to customize column precision in Code First does not work" (MySql bug #65001, Oracle bug #14469048).   * Fixed bug ASP.NET Membership database fails on MySql database UTF32 (MySQL bug #65144, Oracle bug #14495292).   * Fix for MySqlCommand.LastInsertedId holding only 32 bit values (MySql bug #65452, Oracle bug #14171960).   * Fixed "Decimal type should have digits at right of decimal point", now default is 2, and user's changes in     EDM designer are recognized (MySql bug #65127, Oracle bug #14474342).   * Fix for NullReferenceException when saving an uninitialized row in Entity Framework (MySql bug #66066, Oracle bug #14479715).   * Fix for error when calling RoleProvider.RemoveUserFromRole(): causes an exception due to a wrong table being used (MySql bug #65805, Oracle bug #14405338).   * Fix for "Memory Leak on MySql.Data.MySqlClient.MySqlCommand", too many MemoryStream's instances created (MySql bug #65696, Oracle bug #14468204).   * Added ANTLR attribution notice (Oracle bug #14379162).   * Fix for debugger failing when having a routine with an if-elseif-else.   * Also the programming interface for authentication plugins has been redefined. Some limitations remains, due to the current debugger architecture:   * Some MySQL functions cannot be debugged currently (get_lock, release_lock, begin, commit, rollback, set transaction level)..   * Only one debug session may be active on a given server. The Debugger is feature complete at this point. We look forward to your feedback. Documentation ------------------------------------- You can view current Connector/Net documentation at http://dev.mysql.com/doc/refman/5.5/en/connector-net.html You can find our team blog at http://blogs.oracle.com/MySQLOnWindows. You can also post questions on our forums at http://forums.mysql.com/. Enjoy and thanks for the support!

    Read the article

  • Problems integrating nServiceBus with StructureMap

    - by SimonB
    I'm trying to use StructureMap with nServiceBus. The Project: Uses a GenericHost Endpoint to send command messages Configures nServiceBus using the StructMapBuilder. Uses a simple StructureMap registry config Uses a start up class TestServer supporting IWantToRunAtStartup The TestServer class has ctor dependency on a TestManager class The TestManager class has ctor dependency on IBus ObjectFactory.WhatDoIHave() shows StructureMap knows how to construct the classes. When run I get buildup errors. nServiceBus seems to be overwriting the config? Note that when I add a IBus ctor depenendency to my event handlers without any other config all appears fine. Error: Exception when starting endpoint, error has been logged. Reason: Error creating object with name 'nSeviceBusStructureMapTest.TestServer' : Unsatisfied dependency expressed through constructor argument with index 0 of type [nSeviceBusStructureMapTest.ITestManager] : No unique object of type [nSeviceBusStructureMapTest.ITestManager] is defined : Unsatisfied dependency of type [nSeviceBusStructureMapTest.ITestManager]: expected at least 1 matching object to wire the [miningServiceManage] parameter on the constructor of object [nSeviceBusStructureMapTest.TestServer] Source: using System; using System.Diagnostics; using NServiceBus; using StructureMap; using StructureMap.Configuration.DSL; namespace nSeviceBusStructureMapTest { public class TestSmRegistry : Registry { public TestSmRegistry() { For<ITestManager>().Use<TestManager>(); For<TestServer>().Use<TestServer>(); } } public class TestEndPoint : AsA_Server, IConfigureThisEndpoint, IWantCustomLogging { public void Init() { Configure.With().StructureMapBuilder(ObjectFactory.Container); ObjectFactory.Configure(c => c.AddRegistry<TestSmRegistry>()); Debug.WriteLine(ObjectFactory.WhatDoIHave()); } } public class TestServer : IWantToRunAtStartup { public TestServer(ITestManager miningServiceManage) { _miningServiceManage = miningServiceManage; } private readonly ITestManager _miningServiceManage; public void Run() { _miningServiceManage.Run(); } public void Stop() { } } public interface ITestManager { void Run(); } public class TestManager : ITestManager { public TestManager(IBus bus) { _bus = bus; } private readonly IBus _bus; public void Run() { if (_bus == null) Debug.WriteLine("Error no bus"); // Send messages on bus; } } } <MsmqTransportConfig InputQueue="test" ErrorQueue="error" NumberOfWorkerThreads="1" MaxRetries="5" /> <UnicastBusConfig> <MessageEndpointMappings> </MessageEndpointMappings> </UnicastBusConfig> Any ideas?

    Read the article

  • NSObject release destroys local copy of object's data

    - by Spider-Paddy
    I know this is something stupid on my part but I don't get what's happening. I create an object that fetches data & puts it into an array in a specific format, since it fetches asynchronously (has to download & parse data) I put a delegate method into the object that needs the data so that the data fetching object copies it's formatted array into an array in the calling object. The problem is that when the data fetching object is released, the copy it created in the caller is being erased, code is: In .h file @property (nonatomic, retain) NSArray *imagesDataSource; In .m file // Fetch item details ImagesParser *imagesParserObject = [[ImagesParser alloc] init:self]; [imagesParserObject getArticleImagesOfArticleId:(NSInteger)currentArticleId]; [imagesParserObject release] <-- problematic release // Called by parser when images parsing is finished -(void)imagesDataTransferComplete:(ImagesParser *)imagesParserObject { self.imagesDataSource = [ImagesParserObject.returnedArray copy]; // copy array to local variable // If there are more pics, they must be assembled in an array for possible UIImageView animation NSInteger picCount = [imagesDataSource count]; if(picCount > 1) // 1 image is assumed to be the pic already displayed { // Build image array NSMutableArray *tempPicArray = [[NSMutableArray alloc] init]; // Temp space to hold images while building for(int i = 0; i < picCount; i++) { // Get Nr from only article in detailDataSource & pic name (Small) from each item in imagesDataSource NSString *picAddress = [NSString stringWithFormat:@"http://some.url.com/shopdata/image/article/%@/%@", [[detailDataSource objectAtIndex:0] objectForKey:@"Nr"], [[imagesDataSource objectAtIndex:i] objectForKey:@"Small"]]; NSURL *picURL = [NSURL URLWithString:picAddress]; NSData *picData = [NSData dataWithContentsOfURL:picURL]; [tempPicArray addObject:[UIImage imageWithData:picData]]; } imagesArray = [tempPicArray copy]; // copy makes immutable copy of array [tempPicArray release]; currentPicIndex = 0; // Assume first pic is pic already being shown } else imagesArray = nil; // No need for a needless pic array // Remove please wait message [pleaseWaitViewControllerObject.view removeFromSuperview]; } I put in tons of NSLog lines to keep track of what was going on & self.imagesDataSource is populated with the returned array but when the parser object is released self.imagesDataSource becomes empty. I thought self.imagesDataSource = [ImagesParserObject.returnedArray copy]; is supposed to make an independant object, like as if it was alloc, init'ed, so that self.imagesDataSource is not just a pointer to the parser's array but is it's own array. So why does the release of the parser object clear the copy of the array. (I checked & double checked that it's not something overwriting self.imagesDataSource, commenting out [imagesParserObject release] consistently fixes the problem) Also, I have exactly the same problem with self.detailDataSource which is declared & populated in the exact same way as self.imagesDataSource I thought that once I call the parser I could release it because the caller no longer needs to refer to it, all further activity is carried out by the parser object through it's delegate method, what am I doing wrong?

    Read the article

  • Searches (and general querying) with HBase and/or Cassandra (best practices?)

    - by alexeypro
    I have User model object with quite few fields (properties, if you wish) in it. Say "firstname", "lastname", "city" and "year-of-birth". Each user also gets "unique id". I want to be able to search by them. How do I do that properly? How to do that at all? My understanding (will work for pretty much any key-value storage -- first goes key, then value) u:123456789 = serialized_json_object ("u" as a simple prefix for user's keys, 123456789 is "unique id"). Now, thinking that I want to be able to search by firstname and lastname, I can save in: f:Steve = u:384734807,u:2398248764,u:23276263 f:Alex = u:12324355,u:121324334 so key is "f" - which is prefix for firstnames, and "Steve" is actual firstname. For "u:Steve" we save as value all user id's who are "Steve's". That makes every search very-very easy. Querying by few fields (properties) -- say by firstname (i.e. "Steve") and lastname (i.e. "l:Anything") is still easy - first get list of user ids from "f:Steve", then list from "l:Anything", find crossing user ids, an here you go. Problems (and there are quite a few): Saving, updating, deleting user is a pain. It has to be atomic and consistent operation. Also, if we have size of value limited to some value - then we are in (potential) trouble. And really not of an answer here. Only zipping the list of user ids? Not too cool, though. What id we want to add new field to search by. Eventually. Say by "city". We certainly can do the same way "c:Los Angeles" = ..., "c:Chicago" = ..., but if we didn't foresee all those "search choices" from the very beginning, then we will have to be able to create some night job or something to go by all existing User records and update those "c:CITY" for them... Quite a big job! Problems with locking. User "u:123" updates his name "Alex", and user "u:456" updates his name "Alex". They both have to update "f:Alex" with their id's. That means either we get into overwriting problem, or one update will wait for another (and imaging if there are many of them?!). What's the best way of doing that? Keeping in mind that I want to search by many fields? P.S. Please, the question is about HBase/Cassandra/NoSQL/Key-Value storages. Please please - no advices to use MySQL and "read about" SELECTs; and worry about scaling problems "later". There is a reason why I asked MY question exactly the way I did. :-)

    Read the article

  • How can I capture the keystroke that triggers "CellEndEdit" on a DataGridView in C#?

    - by Andy Stampor
    I have a DataGridView that is set to EditOnF2. I do some special processing of data in the CellEndEdit eventhandler that sets the value of the cell. I still want the functionality of the EditOnKeystrokeOrF2 of reverting to the original value when the Esc key is pressed. Unfortunately, at the CellEndEdit eventhandler, I don't see a way to tell what caused the CellEndEdit event to be fired. I only want to change the value of the cell if the Esc key is not pressed. How can I tell if it was or not? Edit: It is worth noting that the KeyDown event does not get fired when the cell is being edited, nor for the final ending keystroke. Edit2: I have tried the KeyPreview suggestion, but the form still does not capture the Escape key being pressed. Edit3: I've been experimenting with trying to get this working. I originally posted some of the following as a separate post, but feel it might be more relevant to include it here. I have a cell in a DataGridView that is now set to EditProgrammatically. To capture the keystroke that starts an edit, I am setting the cell.Value equal to the keystroke. However, this ruins the "Escape" functionality of the cell - when you press escape, instead of reverting to the original value, it reverts to the keystroke that I programmatically inserted into the cell. I believe that if I could set the "EditedFormattedValue" on a cell, this would be where I want to put my keystroke value, however this appears to be read only. How can I accomplish what I am attempting? An example to clarify: If the cell has a value of "54.3" in it, and I press the "9" key, it begins editing the cell and places a "9" there. If I hit Escape, instead of reverting to "54.3" it reverts to "9". What I want is for it to return to its original value of "54.3". So, I am trying to tackle this issue from both the beginning and the end. I think the real problem is that I am overwriting the original value and have no way to determine if I should revert it or not. Edit4: It looks like CellValidating might be worth using, but I am seeing strange behavior when I experiment with it. In a new project, I create the DataGridView and register for the various events and see that CellValidating is fired before the CellEndEdit. However, in my project where I am trying to get this to work, CellEndEdit is firing BEFORE CellValidating. Any ideas on what the difference might be?

    Read the article

  • Self updating app, wont overwrite existing app, using Android packagemanager?

    - by LokiSinclair
    I know there are plenty of questions about this on here, but I've tried everything (but the correct 'thing', obviously!) and nothing seems to shine any light on the problem I'm having. I've written an app (for a customer), which is designed to be hosted on their own server. The app references a simple text file with the latest version code in it and checks it against it's own version. If it's out of date it goes off and downloads the update. Everything is working as intended up to this point. I use the: Intent i = new Intent(Intent.ACTION_VIEW); i.setDataAndType(Uri.fromFile(outputFile), "application/vnd.android.package-archive"); i.setFlags(Intent.FLAG_ACTIVITY_NEW_TASK); startActivity(i); ...code to start the install process of the newly downloaded .apk file. And that all starts as I would expect. I click on "Install" - when I'm prompted to confirm the overwriting of the current app, with the new. It starts, and then displays: App not installed. And existing package by the same name with a conflicting signature is already installed. Now I'm aware that Android can't have multiple applications sharing the same package name, which is fine, but nothing comes up in LogCat and I can only assume that the OS is annoyed at me attempting to 'update' my app, even though I'm going through all the correct channels and using the inbuilt package manager to do it for me! Can anyone tell me what the OS is moaning about? I'm not attempting to install two apps side by side, I want it to update it, which it starts to do, and then gets really confused. Is it something to do with me using the same keystore for signing the packages? I highly doubt it as I've used the same keystores previously to handle updates to games and the like, but I just can't figure out what it's complaining about. Hopefully someone out there has had this issue and solved it, and can point me in the right direction. I'm flying a bit blind with the limited information it's giving me :( Cheers.

    Read the article

  • close file with fclose() but file still in use

    - by Marco
    Hi all, I've got a problem with deleting/overwriting a file using my program which is also being used(read) by my program. The problem seems to be that because of the fact my program is reading data from the file (output.txt) it puts the file in a 'in use' state which makes it impossible to delete or overwrite the file. I don't understand why the file stays 'in use' because I close the file after use with fclose(); this is my code: bool bBool = true while(bBool){ //Run myprogram.exe tot generate (a new) output.txt //Create file pointer and open file FILE* pInputFile = NULL; pInputFile = fopen("output.txt", "r"); // //then I do some reading using fscanf() // //And when I'm done reading I close the file using fclose() fclose(pInputFile); //The next step is deleting the output.txt if( remove( "output.txt" ) == -1 ){ //ERROR }else{ //Succesfull } } I use fclose() to close the file but the file remains in use by my program until my program is totally shut down. What is the solution to free the file so it can be deleted/overwrited? In reality my code isn't a loop without an end ; ) Thanks in advance! Marco Update Like ask a part of my code which also generates the file 'in use'. This is not a loop and this function is being called from the main(); Here is a piece of code: int iShapeNr = 0; void firstRun() { //Run program that generates output.txt runProgram(); //Open Shape data file FILE* pInputFile = NULL; int iNumber = 0; pInputFile = fopen("output.txt", "r"); //Put all orientations of al detected shapes in an array int iShapeNr = 0; int iRotationBuffer[1024];//1024 is maximum detectable shapes, can be changed in RoboRealm int iXMinBuffer[1024]; int iXMaxBuffer[1024]; int iYMinBuffer[1024]; int iYMaxBuffer[1024]; while(feof(pInputFile) == 0){ for(int i=0;i<9;i++){ fscanf(pInputFile, "%d", &iNumber); fscanf(pInputFile, ","); if(i == 1) { iRotationBuffer[iShapeNr] = iNumber; } if(i == 3){//xmin iXMinBuffer[iShapeNr] = iNumber; } if(i == 4){//xmax iXMaxBuffer[iShapeNr] = iNumber; } if(i == 5){//ymin iYMinBuffer[iShapeNr] = iNumber; } if(i == 6){//ymax iYMaxBuffer[iShapeNr] = iNumber; } } iShapeNr++; } fflush(pInputFile); fclose(pInputFile); } The while loop parses the file. The output.txt contains sets of 9 variables, the number of sets is unknown but always in sets of 9. output.txt could contain for example: 0,1,2,3,4,5,6,7,8,8,7,6,5,4,1,2,3,0

    Read the article

  • Using hidden values with jQuery (and ASP.NET MVC) -- not working?

    - by SlackerCoder
    Im using a couple of JSON calls to render data, etc etc. In order to keep the proper key value, I am storing it in a tag. I have this in several places in my code, none of which cause an issue like this one is. Here is the jQuery: The call that "sets" the value: $("a[id^='planSetupAddNewPlan']").live('click', function() { var x = $(this).attr('id'); x = x.substring(19); $("#hidPlanSetupCurrentGroupKey").val(x); $.getJSON("/GroupSetup/PlanSetupAddNewList", { GroupKey: x }, function(data) { $("#planSetupAddNew").html('' + data.TableResult + ''); alert('First Inside 2 ' + x); $.blockUI({ message: $("#planSetupAddNew") }); }); }); The call that "gets" the value: $("#ddlPlanSetupAddNewProduct").live('change', function() { var a = $("#hidPlanSetupCurrentGroupKey").val(); var prod = $(this).val(); alert(a); $.getJSON("/GroupSetup/PlanSetupChangePlanList", { GroupKey: a, Product: prod }, function(data) { if (data.Message == "Success") { $("#planSetupAddNewPlan").html('' + data.TableResult + ''); } else if (data.Message == "Error") { //Do something } }); }); Here is the html in question: <div id="planSetupAddNew" style="display:none; cursor: default;"> <input type="hidden" id="hidPlanSetupCurrentGroupKey" /> <div id="planSetupAddNewData"> </div> </div> In the first section, the alert ('First Inside 2 ' + x) returns what I expect (where x = the key value), and if I add a line to display the contents of the hidden field, that works as well: ie. var key = $("#hidPlanSetupCurrentGroupKey").val(); alert(key); In the "alert(a);" call, I am getting "undefined". I have looked at the other code in the same view and it is the same and it works. I must be missing something, or have some sort of mistype that I havent caught. Just an overview of the controller events: The first call (/GroupSetup/PlanSetupAddNewList) will return an html string building a "form" for users to enter information into. The second call (/GroupSetup/PlanSetupChangePlanList) just changes a second dropdown based on the first dropdown selection (overwriting the html in the div). If you need more info, let me know! Any thoughts/tips/pointers/suggestions?!?! Thanks for all your help :)

    Read the article

  • Randomised objects are assigning themselves to more than one array location

    - by Thaddeus Aid
    this.size = 9; this.populationSize = 10; Random rand = new Random(); Integer[][] easy1 = new Integer[size][size]; easy1 = this.initializeEasy1(easy1); this.sudokuArray = new Sudoku[this.populationSize]; for (int i = 0; i < this.sudokuArray.length; i++){ long seed = rand.nextLong(); System.out.println("" + seed); this.sudokuArray[i] = new Sudoku(easy1, this.size, seed); } I am building an evolutionary sudoku solver and I am having a problem where the last Sudoku object is overwriting all the other objects in the array. Where in the code did I mess up? /edit here is the constructor of the class public Sudoku(Integer[][] givensGrid, int s, long seed){ this.size = s; this.givens = givensGrid; this.grid = this.givens.clone(); Random rand = new Random(seed); System.out.println("Random " + rand.nextInt()); // step though each row of the grid for (int i = 0; i < size; i++){ ArrayList<Integer> numbers = new ArrayList<Integer>(); numbers = this.makeNumbers(numbers); // step through each column to find the givens and remove from numbers for (int j = 0; j < size; j++){ if (this.grid[i][j] != 0){ numbers.remove(this.grid[i][j]); } } // go back through the row and assign the numbers randomly for (int j = 0; j < size; j++){ if (this.grid[i][j] == 0){ int r = rand.nextInt(numbers.size()); this.grid[i][j] = numbers.get(r); numbers.remove(r); } } } System.out.println("============="); System.out.println(this.toString()); }

    Read the article

  • why isn't the input text file being read into the ArrayList<String>, diary?

    - by hmg
    Here's my code: public void readFile() throws IOException { System.out.println("Input file name: "); Scanner in = new Scanner(System.in); String readName = in.nextLine(); Scanner reader = new Scanner(new File(readName)); System.out.println("Reading file..."); while (reader.hasNextLine()) { System.out.println(reader.nextLine()); } System.out.println("File read."); Scanner inAgain = new Scanner(System.in); System.out.println("Would you like to read this into the current Diary?"); System.out.println("Warning: This will overwrite your current Diary!"); String readFileConfirm = inAgain.nextLine(); ArrayList<String> readData = new ArrayList<String>(); if (readFileConfirm.equalsIgnoreCase("yes")) { for (int i = 0; i < readData.size(); i++) { readData.add(reader.nextLine()); } System.out.println("Data added to extra array..."); System.out.println("Here are the contents."); for (int i = 0; i < readData.size(); i++) { System.out.println(readData.get(i)); } System.out.println("Contents read."); System.out.println("Now overwriting current Diary with read file..."); diary.clear(); for (int i = 0; i < diary.size(); i++) { diary.add(readData.get(i)); } System.out.println("New Diary created!"); System.out.println("Use 'recall' to see your new Diary!"); } else { System.out.println("Very well, returning to first command entry."); } } Thanks in advance! -h

    Read the article

  • Bulk update & occasional insert (coredata) - Too slow

    - by Andrew
    Update: Currently looking into NSSET's minusSet links: http://stackoverflow.com/questions/1475636/comparing-two-arrays Hi guys, Could benefit from your wisdom here.. I'm using Coredata in my app, on first launch I download a data file and insert over 500 objects (each with 60 attributes) - fast, no problem. Each subsequent launch I download an updated version of the file, from which I need to update all existing objects' attributes (except maybe 5 attributes) and create new ones for items which have been added to the downloaded file. So, first launch I get 500 objects.. say a week later my file now contains 507 items.. I create two arrays, one for existing and one for downloaded. NSArray *peopleArrayDownloaded = [CoreDataHelper getObjectsFromContext:@"person" :@"person_id" :YES :managedObjectContextPeopleTemp]; NSArray *peopleArrayExisting = [CoreDataHelper getObjectsFromContext:@"person" :@"person_id" :YES :managedObjectContextPeople]; If the count of each array is equal then I just do this: NSUInteger index = 0; if ([peopleArrayExisting count] == [peopleArrayDownloaded count]) { NSLog(@"Number of people downloaded is same as the number of people existing"); for (person *existingPerson in peopleArrayExisting) { person *tempPerson = [peopleArrayDownloaded objectAtIndex:index]; // NSLog(@"Updating id: %@ with id: %@",existingPerson.person_id,tempPerson.person_id); // I have 60 attributes which I to update on each object, is there a quicker way other than overwriting existing? index++; } } else { NSLog(@"Number of people downloaded is different to number of players existing"); So now comes the slow part. I end up using this (which is tooooo slow): NSLog(@"Need people added to the league"); for (person *tempPerson in peopeArrayDownloaded) { NSPredicate *predicate = [NSPredicate predicateWithFormat:@"person_id = %@",tempPerson.person_id]; // NSLog(@"Searching for existing person, person_id: %@",existingPerson.person_id); NSArray *filteredArray = [peopleArrayExisting filteredArrayUsingPredicate:predicate]; if ([filteredArray count] == 0) { NSLog(@"Couldn't find an existing person in the downloaded file. Adding.."); person *newPerson = [NSEntityDescription insertNewObjectForEntityForName:@"person" inManagedObjectContext:managedObjectContextPeople]; Is there a way to generate a new array of index items referring to the additional items in my downloaded file? Incidentally, on my tableViews I'm using NSFetchedResultsController so updating attributes will call [cell setNeedsDisplay]; .. about 60 times per cell, not a good thing and it can crash the app. Thanks for reading :)

    Read the article

  • Hot to implement grails server-side-triggered dialog, or how to break out of update region after AJA

    - by werner5471
    In grails, I use the mechanism below in order to implement what I'd call a conditional server-side-triggered dialog: When a form is submitted, data must first be processed by a controller. Based on the outcome, there must either be a) a modal Yes/No confirmation in front of the "old" screen or b) a redirect to a new controller/view replacing the "old" screen (no confirmation required). So here's my current approach: In the originating view, I have a <g:formRemote name="requestForm" url="[controller:'test', action:'testRequest']", update:"dummyRegion"> and a <span id="dummyRegion"> which is hidden by CSS When submitting the form, the test controller checks if a confirmation is necessary and if so, renders a template with a yui-based dialog including Yes No buttons in front of the old screen (which works fine because the dialog "comes from" the dummyRegion, not overwriting the page). When Yes is pressed, the right other controller & action is called and the old screen is replaced, if No is pressed, the dialog is cancelled and the "old" screen is shown again without the dialog. Works well until here. When submitting the form and test controller sees that NO confirmation is necessary, I would usually directly redirect to the right other controller & action. But the problem is that the corresponding view of that controller does not appear because it is rendered in the invisble dummyRegion as well. So I currently use a GSP template including a javascript redirect which I render instead. However a javascript redirect is often not allowed by the browser and I think it's not a clean solution. So (finally ;-) my question is: How do I get a controller redirect to cause the corresponding view to "break out" of my AJAX dummyRegion, replacing the whole screen again? Or: Do you have a better approach for what I have in mind? But please note that I cannot check on the client side whether the confirmation is necessary, there needs to be a server call! Also I'd like to avoid that the whole page has to be refreshed just for the confirmation dialog to pop up (which would also be possible without AJAX). Thanks for any hints!

    Read the article

  • MEF C# Service - DLL Updating

    - by connerb
    Currently, I have a C# service that runs off of many .dll's and has modules/plugins that it imports at startup. I would like to create an update system that basically stops the service, deletes any files it is told to delete (old versions), downloads new versions from a server, and starts the service. I believe I have coded this right except for the delete part, because as long as I am not overwriting anything, the file will download. If I try to overwrite something, it won't work, which is why I am trying to delete it before hand. However, when I do File.Delete() to the path that I want to do, it gives me access to the path is denied. Here is my code: new Thread(new ThreadStart(() => { ServiceController controller = new ServiceController("client"); controller.Stop(); controller.WaitForStatus(ServiceControllerStatus.Stopped); try { if (um.FilesUpdated != null) { foreach (FilesUpdated file in um.FilesUpdated) { if (file.OldFile != null) { File.Delete(Path.Combine(Utility.AssemblyDirectory, file.OldFile)); } if (file.NewFile != null) { wc.DownloadFile(cs.UpdateUrl + "/updates/client/" + file.NewFile, Path.Combine(Utility.AssemblyDirectory, file.NewFile)); } } } if (um.ModulesUpdated != null) { foreach (ModulesUpdated module in um.ModulesUpdated) { if (module.OldModule != null) { File.Delete(Path.Combine(cs.ModulePath, module.OldModule)); } if (module.NewModule != null) { wc.DownloadFile(cs.UpdateUrl + "/updates/client/modules/" + module.NewModule, Path.Combine(cs.ModulePath, module.NewModule)); } } } } catch (Exception ex) { Logger.log(ex); } controller.Start(); })).Start(); I believe it is because the files are in use, but I can't seem to unload them. I though stopping the service would work, but apparently not. I have also checked the files and they are not read-only (but the folder is, which is located in Program Files, however I couldn't seem to get it to not be read-only programmatically or manually). The service is also being run as an administrator (NT AUTHORITY\SYSTEM). I've read about unloading the AppDomain but AppDomain.Unload(AppDomain.CurrentDomain); returned an exception as well. Not too sure even if this is a problem with MEF or my program just not having the correct permissions...I would assume that it's mainly because the file is in use.

    Read the article

  • NetBackup with VSS and Instant Recovery - Failing to delete old snapshots

    - by Jonathan Bourke
    We are attempting to implement Microsoft VSS for snap-shotting in our NetBackup 6.5.3.1 environment. The clients are both 32 & 64 bit Windows 2003 Server. Snapshot parameters are: Instant recovery is enabled Maximum snapshots = 1 Provider type = 1 (System) Snapshot attribute = 1 (Differential) All backups successfully complete, and VSS shadows are successfully created both for the snapshot backup and for the open files (shadow copy components). The Issue: NetBackup is not clearing or overwriting old snapshots with each successive backup. When we list shadows, and shadow storage, it is increasing and increasing. IT is not honouring the Maximum Snapshot setting. The Logs: The bpfis log doesn’t really appear to show any errors other than for methods which we are not employing (VxVM, Flashsnap, etc.). A section is as follows: 11:54:10.744 [348.4724] <2> logparams: D:\Program Files\Veritas\NetBackup\bin\bpfis.exe delete -nbu -id htpststr001.san.mgmt.det_1248918143 -bpstart_to 300 -bpend_to 300 -clnt htpststr001.san.mgmt.det 11:54:10.744 [348.4724] <4> bpfis: INF - BACKUP START 348 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: VfMS error 10; see following messages: 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: Non-fatal method error was reported 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: vfm_configure_fi_one: method: FlashSnap, type: FIM, function: FlashSnap_init 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: VfMS method error 3; see following message: 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: FlashSnap_init: Veritas Volume Manager not installed. 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: VfMS error 10; see following messages: 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: Non-fatal method error was reported 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: vfm_configure_fi_one: method: vxvm, type: FIM, function: vxvm_init 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: VfMS method error 3; see following message: 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: vxvm_init: Veritas Volume Manager not installed. 11:54:11.713 [348.4724] <4> onlfi_thaw: Thawing C:\ using snapshot method VSS. 11:54:11.713 [348.4724] <2> onlfi_vfms_logf: vfm_thaw: delete snapshot ... 11:54:11.744 [348.4724] <2> onlfi_vfms_logf: snapshot services: emcclariionfi:Thu Jul 30 2009 11:54:11.744000 <Thread id - 4724> Unable to import any login credentials for any appliances. 11:54:11.760 [348.4724] <2> onlfi_vfms_logf: snapshot services: hpevafi:Thu Jul 30 2009 11:54:11.760000 <Thread id - 4724> CHpEvaPlugin::init: CLI tool is not installed. 11:54:11.760 [348.4724] <2> onlfi_vfms_logf: snapshot services: hpmsafi:Thu Jul 30 2009 11:54:11.760000 <Thread id - 4724> No array mangement credentials are available in configuration file. 11:54:13.806 [348.4724] <4> onlfi_thaw: do_thaw return value: 0 11:54:13.806 [348.4724] <4> onlfi_thaw: Thawing D:\ using snapshot method VSS. 11:54:15.806 [348.4724] <4> onlfi_thaw: do_thaw return value: 0 11:54:19.806 [348.4724] <2> fis_delete_id: removing D:\Program Files\Veritas\NetBackup\online_util\fi_cntl\bpfis.fim.htpststr001.san.mgmt.det_1248918143.0 11:54:19.806 [348.4724] <2> fis_delete_id: removing D:\Program Files\Veritas\NetBackup\online_util\fi_cntl\bpfis.fim.htpststr001.san.mgmt.det_1248918143.0.fiid 11:54:19.853 [348.4724] <4> bpfis: INF - EXIT STATUS 0: the requested operation was successfully completed The Question: Has anyone any experience of NetBackup / VSS not clearing snapshots after backups? We will ultimately be using a HP EVA for the snapshots, but we want to ensure correct functioning at a VSS level before we go further. Regards, Jonathan (PS: Question previously posted by my colleague on entsupport.symantec.com)

    Read the article

  • I cut-to-move DCIM folder to ext SD when an auto android OS update popped up b4 I could choose target - Cannot recover 200+ photos

    - by ZeroG
    I was downloading my Exhibit II's DCIM camera folder (with month's of photos inside) to its external SD card, in order to transfer them into my laptop. In my overconfidence, I hurriedly chose cut-to-move (rather than copy-to-move) when KABOOM! —an automatic Android OS update popped up before I could choose the target!!! I figured everything was in cache & calmly tried to go through with the update. But that was not a typically seamless event. It showed downloading icon but hmm… since I rooted the phone it brought the command line up & recovery sequence. But neither Android nor I had yet downloaded any alternate custom ROM Files to internal SD to update from! So were they trying to make me unroot my phone by giving me some bogus update on the fly or just give me a hard time in trying to hand me down an unrooted ROM that I'd have to figure out how to root again? Yes, I know there was that blurb about overwriting a file of the same name but I was trying to shake the darn stubborn update being forced on my phone during this precarious moment. I thought I had frozen or turned off all those auto-updates previously. Anyway, phones are small & fingers are big (sigh)... I tried to reboot into safe mode but the resultant photo file was partially overwritten (200 files had names but Zero bytes in them). I thought maybe it was still hung in cache or deposited somewhere else but I have searched everywhere with file managers. Since I did not have Titanium backing up camera, photo folder or gallery, I cannot recover 200+ photos. Dumb. You can understand my dilemma as I am involved in the arts & although just a camera phone, most of these photos were historic & aesthetic or at least as to subject matter. Photo-ops don't reoccur. I have tried a couple of recovery apps from the market like Search Duplicates & Recover to no avail. I was only able to salvage stuff I'd sent out in messages. I've got several decades in computers & this is such a miserable beginner's piece of bad luck I can't believe it happened to me. They were precious photos! Yes, I turned on Titanium since & yes I even tried USB to laptop recoveries. Being on a MacBookPro I'm trying androidfiletransfer.dmg, but I'd have to upgrade to Peach Sunrise to get above Android 3.0 for that App to recognize the phone via USB & the programmer says installation zeros your data, so that pretty much toasts any secret hidden places where these photos may have been deposited. Don't want to do that & am still trying to find them. They certainly didn't make it to my external SD Card. If any of you techies out there know anything, please help & thanks. Despite decades of being in computing, unfamiliar & ever-changing hard or software can humble even the most seasoned veterans.

    Read the article

  • NetBackup with VSS and Instant Recovery - Failing to delete old snapshots

    - by Jonathan Bourke
    We are attempting to implement Microsoft VSS for snap-shotting in our NetBackup 6.5.3.1 environment. The clients are both 32 & 64 bit Windows 2003 Server. Snapshot parameters are: Instant recovery is enabled Maximum snapshots = 1 Provider type = 1 (System) Snapshot attribute = 1 (Differential) All backups successfully complete, and VSS shadows are successfully created both for the snapshot backup and for the open files (shadow copy components). The Issue: NetBackup is not clearing or overwriting old snapshots with each successive backup. When we list shadows, and shadow storage, it is increasing and increasing. IT is not honouring the Maximum Snapshot setting. The Logs: The bpfis log doesn’t really appear to show any errors other than for methods which we are not employing (VxVM, Flashsnap, etc.). A section is as follows: 11:54:10.744 [348.4724] <2> logparams: D:\Program Files\Veritas\NetBackup\bin\bpfis.exe delete -nbu -id htpststr001.san.mgmt.det_1248918143 -bpstart_to 300 -bpend_to 300 -clnt htpststr001.san.mgmt.det 11:54:10.744 [348.4724] <4> bpfis: INF - BACKUP START 348 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: VfMS error 10; see following messages: 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: Non-fatal method error was reported 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: vfm_configure_fi_one: method: FlashSnap, type: FIM, function: FlashSnap_init 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: VfMS method error 3; see following message: 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: FlashSnap_init: Veritas Volume Manager not installed. 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: VfMS error 10; see following messages: 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: Non-fatal method error was reported 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: vfm_configure_fi_one: method: vxvm, type: FIM, function: vxvm_init 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: VfMS method error 3; see following message: 11:54:11.713 [348.4724] <8> onlfi_initialize_vfms: vxvm_init: Veritas Volume Manager not installed. 11:54:11.713 [348.4724] <4> onlfi_thaw: Thawing C:\ using snapshot method VSS. 11:54:11.713 [348.4724] <2> onlfi_vfms_logf: vfm_thaw: delete snapshot ... 11:54:11.744 [348.4724] <2> onlfi_vfms_logf: snapshot services: emcclariionfi:Thu Jul 30 2009 11:54:11.744000 <Thread id - 4724> Unable to import any login credentials for any appliances. 11:54:11.760 [348.4724] <2> onlfi_vfms_logf: snapshot services: hpevafi:Thu Jul 30 2009 11:54:11.760000 <Thread id - 4724> CHpEvaPlugin::init: CLI tool is not installed. 11:54:11.760 [348.4724] <2> onlfi_vfms_logf: snapshot services: hpmsafi:Thu Jul 30 2009 11:54:11.760000 <Thread id - 4724> No array mangement credentials are available in configuration file. 11:54:13.806 [348.4724] <4> onlfi_thaw: do_thaw return value: 0 11:54:13.806 [348.4724] <4> onlfi_thaw: Thawing D:\ using snapshot method VSS. 11:54:15.806 [348.4724] <4> onlfi_thaw: do_thaw return value: 0 11:54:19.806 [348.4724] <2> fis_delete_id: removing D:\Program Files\Veritas\NetBackup\online_util\fi_cntl\bpfis.fim.htpststr001.san.mgmt.det_1248918143.0 11:54:19.806 [348.4724] <2> fis_delete_id: removing D:\Program Files\Veritas\NetBackup\online_util\fi_cntl\bpfis.fim.htpststr001.san.mgmt.det_1248918143.0.fiid 11:54:19.853 [348.4724] <4> bpfis: INF - EXIT STATUS 0: the requested operation was successfully completed The Question: Has anyone any experience of NetBackup / VSS not clearing snapshots after backups? We will ultimately be using a HP EVA for the snapshots, but we want to ensure correct functioning at a VSS level before we go further. Regards, Jonathan (PS: Question previously posted by my colleague on entsupport.symantec.com)

    Read the article

  • SQL SERVER – How to Recover SQL Database Data Deleted by Accident

    - by Pinal Dave
    In Repair a SQL Server database using a transaction log explorer, I showed how to use ApexSQL Log, a SQL Server transaction log viewer, to recover a SQL Server database after a disaster. In this blog, I’ll show you how to use another SQL Server disaster recovery tool from ApexSQL in a situation when data is accidentally deleted. You can download ApexSQL Recover here, install, and play along. With a good SQL Server disaster recovery strategy, data recovery is not a problem. You have a reliable full database backup with valid data, a full database backup and subsequent differential database backups, or a full database backup and a chain of transaction log backups. But not all situations are ideal. Here we’ll address some sub-optimal scenarios, where you can still successfully recover data. If you have only a full database backup This is the least optimal SQL Server disaster recovery strategy, as it doesn’t ensure minimal data loss. For example, data was deleted on Wednesday. Your last full database backup was created on Sunday, three days before the records were deleted. By using the full database backup created on Sunday, you will be able to recover SQL database records that existed in the table on Sunday. If there were any records inserted into the table on Monday or Tuesday, they will be lost forever. The same goes for records modified in this period. This method will not bring back modified records, only the old records that existed on Sunday. If you restore this full database backup, all your changes (intentional and accidental) will be lost and the database will be reverted to the state it had on Sunday. What you have to do is compare the records that were in the table on Sunday to the records on Wednesday, create a synchronization script, and execute it against the Wednesday database. If you have a full database backup followed by differential database backups Let’s say the situation is the same as in the example above, only you create a differential database backup every night. Use the full database backup created on Sunday, and the last differential database backup (created on Tuesday). In this scenario, you will lose only the data inserted and updated after the differential backup created on Tuesday. If you have a full database backup and a chain of transaction log backups This is the SQL Server disaster recovery strategy that provides minimal data loss. With a full chain of transaction logs, you can recover the SQL database to an exact point in time. To provide optimal results, you have to know exactly when the records were deleted, because restoring to a later point will not bring back the records. This method requires restoring the full database backup first. If you have any differential log backup created after the last full database backup, restore the most recent one. Then, restore transaction log backups, one by one, it the order they were created starting with the first created after the restored differential database backup. Now, the table will be in the state before the records were deleted. You have to identify the deleted records, script them and run the script against the original database. Although this method is reliable, it is time-consuming and requires a lot of space on disk. How to easily recover deleted records? The following solution enables you to recover SQL database records even if you have no full or differential database backups and no transaction log backups. To understand how ApexSQL Recover works, I’ll explain what happens when table data is deleted. Table data is stored in data pages. When you delete table records, they are not immediately deleted from the data pages, but marked to be overwritten by new records. Such records are not shown as existing anymore, but ApexSQL Recover can read them and create undo script for them. How long will deleted records stay in the MDF file? It depends on many factors, as time passes it’s less likely that the records will not be overwritten. The more transactions occur after the deletion, the more chances the records will be overwritten and permanently lost. Therefore, it’s recommended to create a copy of the database MDF and LDF files immediately (if you cannot take your database offline until the issue is solved) and run ApexSQL Recover on them. Note that a full database backup will not help here, as the records marked for overwriting are not included in the backup. First, I’ll delete some records from the Person.EmailAddress table in the AdventureWorks database.   I can delete these records in SQL Server Management Studio, or execute a script such as DELETE FROM Person.EmailAddress WHERE BusinessEntityID BETWEEN 70 AND 80 Then, I’ll start ApexSQL Recover and select From DELETE operation in the Recovery tab.   In the Select the database to recover step, first select the SQL Server instance. If it’s not shown in the drop-down list, click the Server icon right to the Server drop-down list and browse for the SQL Server instance, or type the instance name manually. Specify the authentication type and select the database in the Database drop-down list.   In the next step, you’re prompted to add additional data sources. As this can be a tricky step, especially for new users, ApexSQL Recover offers help via the Help me decide option.   The Help me decide option guides you through a series of questions about the database transaction log and advises what files to add. If you know that you have no transaction log backups or detached transaction logs, or the online transaction log file has been truncated after the data was deleted, select No additional transaction logs are available. If you know that you have transaction log backups that contain the delete transactions you want to recover, click Add transaction logs. The online transaction log is listed and selected automatically.   Click Add if to add transaction log backups. It would be best if you have a full transaction log chain, as explained above. The next step for this option is to specify the time range.   Selecting a small time range for the time of deletion will create the recovery script just for the accidentally deleted records. A wide time range might script the records deleted on purpose, and you don’t want that. If needed, you can check the script generated and manually remove such records. After that, for all data sources options, the next step is to select the tables. Be careful here, if you deleted some data from other tables on purpose, and don’t want to recover them, don’t select all tables, as ApexSQL Recover will create the INSERT script for them too.   The next step offers two options: to create a recovery script that will insert the deleted records back into the Person.EmailAddress table, or to create a new database, create the Person.EmailAddress table in it, and insert the deleted records. I’ll select the first one.   The recovery process is completed and 11 records are found and scripted, as expected.   To see the script, click View script. ApexSQL Recover has its own script editor, where you can review, modify, and execute the recovery script. The insert into statements look like: INSERT INTO Person.EmailAddress( BusinessEntityID, EmailAddressID, EmailAddress, rowguid, ModifiedDate) VALUES( 70, 70, N'[email protected]' COLLATE SQL_Latin1_General_CP1_CI_AS, 'd62c5b4e-c91f-403f-b630-7b7e0fda70ce', '20030109 00:00:00.000' ); To execute the script, click Execute in the menu.   If you want to check whether the records are really back, execute SELECT * FROM Person.EmailAddress WHERE BusinessEntityID BETWEEN 70 AND 80 As shown, ApexSQL Recover recovers SQL database data after accidental deletes even without the database backup that contains the deleted data and relevant transaction log backups. ApexSQL Recover reads the deleted data from the database data file, so this method can be used even for databases in the Simple recovery model. Besides recovering SQL database records from a DELETE statement, ApexSQL Recover can help when the records are lost due to a DROP TABLE, or TRUNCATE statement, as well as repair a corrupted MDF file that cannot be attached to as SQL Server instance. You can find more information about how to recover SQL database lost data and repair a SQL Server database on ApexSQL Solution center. There are solutions for various situations when data needs to be recovered. Reference: Pinal Dave (http://blog.sqlauthority.com)Filed under: PostADay, SQL, SQL Authority, SQL Backup and Restore, SQL Query, SQL Server, SQL Tips and Tricks, T SQL

    Read the article

  • Updating a SharePoint master page via a solution (WSP)

    - by Kelly Jones
    In my last blog post, I wrote how to deploy a SharePoint theme using Features and a solution package.  As promised in that post, here is how to update an already deployed master page. There are several ways to update a master page in SharePoint.  You could upload a new version to the master page gallery, or you could upload a new master page to the gallery, and then set the site to use this new page.  Manually uploading your master page to the master page gallery might be the best option, depending on your environment.  For my client, I did these steps in code, which is what they preferred. (Image courtesy of: http://www.joiningdots.net/blog/2007/08/sharepoint-and-quick-launch.html ) Before you decide which method you need to use, take a look at your existing pages.  Are they using the SharePoint dynamic token or the static token for the master page reference?  The wha, huh? SO, there are four ways to tell an .aspx page hosted in SharePoint which master page it should use: “~masterurl/default.master” – tells the page to use the default.master property of the site “~masterurl/custom.master” – tells the page to use the custom.master property of the site “~site/default.master” – tells the page to use the file named “default.master” in the site’s master page gallery “~sitecollection/default.master” – tells the page to use the file named “default.master” in the site collection’s master page gallery For more information about these tokens, take a look at this article on MSDN. Once you determine which token your existing pages are pointed to, then you know which file you need to update.  So, if the ~masterurl tokens are used, then you upload a new master page, either replacing the existing one or adding another one to the gallery.  If you’ve uploaded a new file with a new name, you’ll just need to set it as the master page either through the UI (MOSS only) or through code (MOSS or WSS Feature receiver code – or using SharePoint Designer). If the ~site or ~sitecollection tokens were used, then you’re limited to either replacing the existing master page, or editing all of your existing pages to point to another master page.  In most cases, it probably makes sense to just replace the master page. For my project, I’m working with WSS and the existing pages are set to the ~sitecollection token.  Based on this, I decided to just upload a new version of the existing master page (and not modify the dozens of existing pages). Also, since my client prefers Features and solutions, I created a master page Feature and a corresponding Feature Receiver.  For information on creating the elements and feature files, check out this post: http://sharepointmagazine.net/technical/development/deploying-the-master-page . This works fine, unless you are overwriting an existing master page, which was my case.  You’ll run into errors because the master page file needs to be checked out, replaced, and then checked in.  I wrote code in my Feature Activated event handler to accomplish these steps. Here are the steps necessary in code: Get the file name from the elements file of the Feature Check out the file from the master page gallery Upload the file to the master page gallery Check in the file to the master page gallery Here’s the code in my Feature Receiver: 1: public override void FeatureActivated(SPFeatureReceiverProperties properties) 2: { 3: try 4: { 5:   6: SPElementDefinitionCollection col = properties.Definition.GetElementDefinitions(System.Globalization.CultureInfo.CurrentCulture); 7:   8: using (SPWeb curweb = GetCurWeb(properties)) 9: { 10: foreach (SPElementDefinition ele in col) 11: { 12: if (string.Compare(ele.ElementType, "Module", true) == 0) 13: { 14: // <Module Name="DefaultMasterPage" List="116" Url="_catalogs/masterpage" RootWebOnly="FALSE"> 15: // <File Url="myMaster.master" Type="GhostableInLibrary" IgnoreIfAlreadyExists="TRUE" 16: // Path="MasterPages/myMaster.master" /> 17: // </Module> 18: string Url = ele.XmlDefinition.Attributes["Url"].Value; 19: foreach (System.Xml.XmlNode file in ele.XmlDefinition.ChildNodes) 20: { 21: string Url2 = file.Attributes["Url"].Value; 22: string Path = file.Attributes["Path"].Value; 23: string fileType = file.Attributes["Type"].Value; 24:   25: if (string.Compare(fileType, "GhostableInLibrary", true) == 0) 26: { 27: //Check out file in document library 28: SPFile existingFile = curweb.GetFile(Url + "/" + Url2); 29:   30: if (existingFile != null) 31: { 32: if (existingFile.CheckOutStatus != SPFile.SPCheckOutStatus.None) 33: { 34: throw new Exception("The master page file is already checked out. Please make sure the master page file is checked in, before activating this feature."); 35: } 36: else 37: { 38: existingFile.CheckOut(); 39: existingFile.Update(); 40: } 41: } 42:   43: //Upload file to document library 44: string filePath = System.IO.Path.Combine(properties.Definition.RootDirectory, Path); 45: string fileName = System.IO.Path.GetFileName(filePath); 46: char slash = Convert.ToChar("/"); 47: string[] folders = existingFile.ParentFolder.Url.Split(slash); 48:   49: if (folders.Length > 2) 50: { 51: Logger.logMessage("More than two folders were detected in the library path for the master page. Only two are supported.", 52: Logger.LogEntryType.Information); //custom logging component 53: } 54:   55: SPFolder myLibrary = curweb.Folders[folders[0]].SubFolders[folders[1]]; 56:   57: FileStream fs = File.OpenRead(filePath); 58:   59: SPFile newFile = myLibrary.Files.Add(fileName, fs, true); 60:   61: myLibrary.Update(); 62: newFile.CheckIn("Updated by Feature", SPCheckinType.MajorCheckIn); 63: newFile.Update(); 64: } 65: } 66: } 67: } 68: } 69: } 70: catch (Exception ex) 71: { 72: string msg = "Error occurred during feature activation"; 73: Logger.logException(ex, msg, ""); 74: } 75:   76: } 77:   78: /// <summary> 79: /// Using a Feature's properties, get a reference to the Current Web 80: /// </summary> 81: /// <param name="properties"></param> 82: public SPWeb GetCurWeb(SPFeatureReceiverProperties properties) 83: { 84: SPWeb curweb; 85:   86: //Check if the parent of the web is a site or a web 87: if (properties != null && properties.Feature.Parent.GetType().ToString() == "Microsoft.SharePoint.SPWeb") 88: { 89:   90: //Get web from parent 91: curweb = (SPWeb)properties.Feature.Parent; 92: 93: } 94: else 95: { 96: //Get web from Site 97: using (SPSite cursite = (SPSite)properties.Feature.Parent) 98: { 99: curweb = (SPWeb)cursite.OpenWeb(); 100: } 101: } 102:   103: return curweb; 104: } This did the trick.  It allowed me to update my existing master page, through an easily repeatable process (which is great when you are working with more than one environment and what to do things like TEST it!).  I did run into what I would classify as a strange issue with one of my subsites, but that’s the topic for another blog post.

    Read the article

  • SQL SERVER – Using expressor Composite Types to Enforce Business Rules

    - by pinaldave
    One of the features that distinguish the expressor Data Integration Platform from other products in the data integration space is its concept of composite types, which provide an effective and easily reusable way to clearly define the structure and characteristics of data within your application.  An important feature of the composite type approach is that it allows you to easily adjust the content of a record to its ultimate purpose.  For example, a record used to update a row in a database table is easily defined to include only the minimum set of columns, that is, a value for the key column and values for only those columns that need to be updated. Much like a class in higher level programming languages, you can also use the composite type as a way to enforce business rules onto your data by encapsulating a datum’s name, data type, and constraints (for example, maximum, minimum, or acceptable values) as a single entity, which ensures that your data can not assume an invalid value.  To what extent you use this functionality is a decision you make when designing your application; the expressor design paradigm does not force this approach on you. Let’s take a look at how these features are used.  Suppose you want to create a group of applications that maintain the employee table in your human resources database. Your table might have a structure similar to the HumanResources.Employee table in the AdventureWorks database.  This table includes two columns, EmployeID and rowguid, that are maintained by the relational database management system; you cannot provide values for these columns when inserting new rows into the table. Additionally, there are columns such as VacationHours and SickLeaveHours that you might choose to update for all employees on a monthly basis, which justifies creation of a dedicated application. By creating distinct composite types for the read, insert and update operations against this table, you can more easily manage this table’s content. When developing this application within expressor Studio, your first task is to create a schema artifact for the database table.  This process is completely driven by a wizard, only requiring that you select the desired database schema and table.  The resulting schema artifact defines the mapping of result set records to a record within the expressor data integration application.  The structure of the record within the expressor application is a composite type that is given the default name CompositeType1.  As you can see in the following figure, all columns from the table are included in the result set and mapped to an identically named attribute in the default composite type. If you are developing an application that needs to read this table, perhaps to prepare a year-end report of employees by department, you would probably not be interested in the data in the rowguid and ModifiedDate columns.  A typical approach would be to drop this unwanted data in a downstream operator.  But using an alternative composite type provides a better approach in which the unwanted data never enters your application. While working in expressor  Studio’s schema editor, simply create a second composite type within the same schema artifact, which you could name ReadTable, and remove the attributes corresponding to the unwanted columns. The value of an alternative composite type is even more apparent when you want to insert into or update the table.  In the composite type used to insert rows, remove the attributes corresponding to the EmployeeID primary key and rowguid uniqueidentifier columns since these values are provided by the relational database management system. And to update just the VacationHours and SickLeaveHours columns, use a composite type that includes only the attributes corresponding to the EmployeeID, VacationHours, SickLeaveHours and ModifiedDate columns. By specifying this schema artifact and composite type in a Write Table operator, your upstream application need only deal with the four required attributes and there is no risk of unintentionally overwriting a value in a column that does not need to be updated. Now, what about the option to use the composite type to enforce business rules?  If you review the composition of the default composite type CompositeType1, you will note that the constraints defined for many of the attributes mirror the table column specifications.  For example, the maximum number of characters in the NationaIDNumber, LoginID and Title attributes is equivalent to the maximum width of the target column, and the size of the MaritalStatus and Gender attributes is limited to a single character as required by the table column definition.  If your application code leads to a violation of these constraints, an error will be raised.  The expressor design paradigm then allows you to handle the error in a way suitable for your application.  For example, a string value could be truncated or a numeric value could be rounded. Moreover, you have the option of specifying additional constraints that support business rules unrelated to the table definition. Let’s assume that the only acceptable values for marital status are S, M, and D.  Within the schema editor, double-click on the MaritalStatus attribute to open the Edit Attribute window.  Then click the Allowed Values checkbox and enter the acceptable values into the Constraint Value text box. The schema editor is updated accordingly. There is one more option that the expressor semantic type paradigm supports.  Since the MaritalStatus attribute now clearly specifies how this type of information should be represented (a single character limited to S, M or D), you can convert this attribute definition into a shared type, which will allow you to quickly incorporate this definition into another composite type or into the description of an output record from a transform operator. Again, double-click on the MaritalStatus attribute and in the Edit Attribute window, click Convert, which opens the Share Local Semantic Type window that you use to name this shared type.  There’s no requirement that you give the shared type the same name as the attribute from which it was derived.  You should supply a name that makes it obvious what the shared type represents. In this posting, I’ve overviewed the expressor semantic type paradigm and shown how it can be used to make your application development process more productive.  The beauty of this feature is that you choose when and to what extent you utilize the functionality, but I’m certain that if you opt to follow this approach your efforts will become more efficient and your work will progress more quickly.  As always, I encourage you to download and evaluate expressor Studio for your current and future data integration needs. Reference: Pinal Dave (http://blog.SQLAuthority.com) Filed under: CodeProject, Pinal Dave, PostADay, SQL, SQL Authority, SQL Documentation, SQL Query, SQL Server, SQL Tips and Tricks, SQLServer, T SQL, Technology

    Read the article

  • tile_static, tile_barrier, and tiled matrix multiplication with C++ AMP

    - by Daniel Moth
    We ended the previous post with a mechanical transformation of the C++ AMP matrix multiplication example to the tiled model and in the process introduced tiled_index and tiled_grid. This is part 2. tile_static memory You all know that in regular CPU code, static variables have the same value regardless of which thread accesses the static variable. This is in contrast with non-static local variables, where each thread has its own copy. Back to C++ AMP, the same rules apply and each thread has its own value for local variables in your lambda, whereas all threads see the same global memory, which is the data they have access to via the array and array_view. In addition, on an accelerator like the GPU, there is a programmable cache, a third kind of memory type if you'd like to think of it that way (some call it shared memory, others call it scratchpad memory). Variables stored in that memory share the same value for every thread in the same tile. So, when you use the tiled model, you can have variables where each thread in the same tile sees the same value for that variable, that threads from other tiles do not. The new storage class for local variables introduced for this purpose is called tile_static. You can only use tile_static in restrict(direct3d) functions, and only when explicitly using the tiled model. What this looks like in code should be no surprise, but here is a snippet to confirm your mental image, using a good old regular C array // each tile of threads has its own copy of locA, // shared among the threads of the tile tile_static float locA[16][16]; Note that tile_static variables are scoped and have the lifetime of the tile, and they cannot have constructors or destructors. tile_barrier In amp.h one of the types introduced is tile_barrier. You cannot construct this object yourself (although if you had one, you could use a copy constructor to create another one). So how do you get one of these? You get it, from a tiled_index object. Beyond the 4 properties returning index objects, tiled_index has another property, barrier, that returns a tile_barrier object. The tile_barrier class exposes a single member, the method wait. 15: // Given a tiled_index object named t_idx 16: t_idx.barrier.wait(); 17: // more code …in the code above, all threads in the tile will reach line 16 before a single one progresses to line 17. Note that all threads must be able to reach the barrier, i.e. if you had branchy code in such a way which meant that there is a chance that not all threads could reach line 16, then the code above would be illegal. Tiled Matrix Multiplication Example – part 2 So now that we added to our understanding the concepts of tile_static and tile_barrier, let me obfuscate rewrite the matrix multiplication code so that it takes advantage of tiling. Before you start reading this, I suggest you get a cup of your favorite non-alcoholic beverage to enjoy while you try to fully understand the code. 01: void MatrixMultiplyTiled(vector<float>& vC, const vector<float>& vA, const vector<float>& vB, int M, int N, int W) 02: { 03: static const int TS = 16; 04: array_view<const float,2> a(M, W, vA); 05: array_view<const float,2> b(W, N, vB); 06: array_view<writeonly<float>,2> c(M,N,vC); 07: parallel_for_each(c.grid.tile< TS, TS >(), 08: [=] (tiled_index< TS, TS> t_idx) restrict(direct3d) 09: { 10: int row = t_idx.local[0]; int col = t_idx.local[1]; 11: float sum = 0.0f; 12: for (int i = 0; i < W; i += TS) { 13: tile_static float locA[TS][TS], locB[TS][TS]; 14: locA[row][col] = a(t_idx.global[0], col + i); 15: locB[row][col] = b(row + i, t_idx.global[1]); 16: t_idx.barrier.wait(); 17: for (int k = 0; k < TS; k++) 18: sum += locA[row][k] * locB[k][col]; 19: t_idx.barrier.wait(); 20: } 21: c[t_idx.global] = sum; 22: }); 23: } Notice that all the code up to line 9 is the same as per the changes we made in part 1 of tiling introduction. If you squint, the body of the lambda itself preserves the original algorithm on lines 10, 11, and 17, 18, and 21. The difference being that those lines use new indexing and the tile_static arrays; the tile_static arrays are declared and initialized on the brand new lines 13-15. On those lines we copy from the global memory represented by the array_view objects (a and b), to the tile_static vanilla arrays (locA and locB) – we are copying enough to fit a tile. Because in the code that follows on line 18 we expect the data for this tile to be in the tile_static storage, we need to synchronize the threads within each tile with a barrier, which we do on line 16 (to avoid accessing uninitialized memory on line 18). We also need to synchronize the threads within a tile on line 19, again to avoid the race between lines 14, 15 (retrieving the next set of data for each tile and overwriting the previous set) and line 18 (not being done processing the previous set of data). Luckily, as part of the awesome C++ AMP debugger in Visual Studio there is an option that helps you find such races, but that is a story for another blog post another time. May I suggest reading the next section, and then coming back to re-read and walk through this code with pen and paper to really grok what is going on, if you haven't already? Cool. Why would I introduce this tiling complexity into my code? Funny you should ask that, I was just about to tell you. There is only one reason we tiled our extent, had to deal with finding a good tile size, ensure the number of threads we schedule are correctly divisible with the tile size, had to use a tiled_index instead of a normal index, and had to understand tile_barrier and to figure out where we need to use it, and double the size of our lambda in terms of lines of code: the reason is to be able to use tile_static memory. Why do we want to use tile_static memory? Because accessing tile_static memory is around 10 times faster than accessing the global memory on an accelerator like the GPU, e.g. in the code above, if you can get 150GB/second accessing data from the array_view a, you can get 1500GB/second accessing the tile_static array locA. And since by definition you are dealing with really large data sets, the savings really pay off. We have seen tiled implementations being twice as fast as their non-tiled counterparts. Now, some algorithms will not have performance benefits from tiling (and in fact may deteriorate), e.g. algorithms that require you to go only once to global memory will not benefit from tiling, since with tiling you already have to fetch the data once from global memory! Other algorithms may benefit, but you may decide that you are happy with your code being 150 times faster than the serial-version you had, and you do not need to invest to make it 250 times faster. Also algorithms with more than 3 dimensions, which C++ AMP supports in the non-tiled model, cannot be tiled. Also note that in future releases, we may invest in making the non-tiled model, which already uses tiling under the covers, go the extra step and use tile_static memory on your behalf, but it is obviously way to early to commit to anything like that, and we certainly don't do any of that today. Comments about this post by Daniel Moth welcome at the original blog.

    Read the article

< Previous Page | 9 10 11 12 13 14  | Next Page >