Search Results

Search found 45925 results on 1837 pages for 'process start'.

Page 179/1837 | < Previous Page | 175 176 177 178 179 180 181 182 183 184 185 186  | Next Page >

  • I would like to build a Java Web Application. What's the quickest start?

    - by Vish
    I am a reasonably competent programmer, with about 11 years experience. I particularly like Ruby/Rails and ActionScript. And Java is not too hard. I'd like to develop a web application in Java in order to leverage things like BlazeDS and learn some new tricks with regard to JBoss and the whole Java App thing. But I'm having trouble - and confused - looking for a smooth start. I don't want to use Grails, thanks, I want to use Pure Java, with whatever Framework you can think of. Spring, Hibernate, lalala, it's cool. Where do I start?

    Read the article

  • How to troubleshoot when one has no idea where to start?

    - by Chris Walton
    I am looking for hints, tips and answers on how to get started on troubleshooting when: The problem is intermittent The problem could lie literally anywhere - operating system; free source software; my own software developments; purchased software; crumbs on the keyboard; the specific combination of software I am currently running; Maxwell's demon; the little blue men actually running the machine have gone on strike; etc. I have expertise only in a few of the areas that are potential candidates for the cause of the problem. The specific problem I am having is detailed below as an example, but I am not seeking answers to my current problem, but rather where and how to start on tackling such problems. I am currently encountering a problem with my new machine. On a few occasions the machine has just frozen; not accepting keystrokes, mouseclicks, or anything except the power on/off switch. Invariably I have been merely browsing the web; I have had a few (<= 6 other applications) running. None of these applications are major; and represent a mix of commercial programs and open source programs, typically migrated from Unix of some variety. My machine is a Windows 7 I7 quad core laptop.

    Read the article

  • How to process AJAX requests more securely in PHP?

    - by animuson
    Ok, so I want to send AJAX requests to my website from my Flash games to process data, but I don't want people downloading them, decompiling them, then sending fake requests to be processed, so I'm trying to figure out the most secure way to process in the PHP files. My first idea was to use Apache's built in Authorization module to require a username and password to access the pages on a separate subdomain of my website, but then you'd have to include that username and password in the AJAX request anyway so that seems kind of pointless to even try. My current option looks pretty promising but I want to make sure it will work. Basically it just checks the IP address being sent using REMOTE_ADDR to make sure it's the IP address that my server runs on. <? $allowed = new Array("64.120.211.89", "64.120.211.90"); if (!in_array($_SERVER['REMOTE_ADDR'], $allowed)) header("HTTP/1.1 403 Forbidden"); ?> Both of those IP addresses point to my server. Things I'm worried about: 1) If I send a request from Flash/ActionScript, will that affect the IP address in any way? 2) Is it possible for malicious users to change the IP address that is being sent with REMOTE_ADDR to one of my IP addresses? Any other ways you would suggest that might be more secure?

    Read the article

  • Getting Hprof dump for other processes from application code

    - by Natarajan
    Hi, In my application , i have an option to capture the hprof dump. I used android.os.Debug.dumpHprofData (String fileName) Initially i though the hprof data generated by the method above is for the entire device , which is not so . The hprof data generated is only for my process. Now i am trying to generate hprof data for other process as well. I need to get the Hprof dump for all the running processes from application code. from adb shell i tried "kill -10 " , This command will generate the hprof file for the corresponding process in the data/misc folder. Now the problem is this command is working perfectly from the adb shell prompt , but i am not able to embed the command to mycode. My code is like Runtime.getRuntime().exec("chmod 777 /data/misc") Runtime.getRunTime().exec("kill -10 ") No exceptions are thrown , but somehow it is not working. The same code above is capturing Hprof dump for my process, when i give my process ID. I tried with "android.os.Process.sendSignal (int pid, android.os.Process.SIGNAL_USR1) ;" also.Getting the same problem.It is capturing Hprof dump for my process. For other processes it is not working. Do we need to have any special permission to kill other process from our process ? Or is it a built issue ? can you please suggest some possible way to get Hprof dump for other processes from application code? Thanks

    Read the article

  • CM and Agile validation process of merging to the Trunk?

    - by LoneCM
    Hello All, We are a new Agile shop and we are encountering an issue that I hope others have seen. In our process, the Trunk is considered an integration branch; it does not have to be releasable, but it does have to be stable and functional for others to branch off of. We create Feature branches of the Trunk for new development. All work and testing occurs in these branches. An individual branch pulls up as needed to stay integrated with the Trunk as other features that are accepted and are committed. But now we have numerous feature branches. Each are focused, have a short life cycle, and are pushed to the trunk as they are completed, so we not debating the need for the branches and trying very much to be Agile. My issue comes in here: I require that the branches pull up from the Trunk at the end of their life cycle and complete the validation, regression testing and handle all configuration issues before pushing to the trunk. Once reintegrated into the Trunk, I ask for at least a build and an automated smoke test. However, I am now getting push back on the Trunk validation. The argument is that the developers can merge the code and not need the QA validation steps because they already complete the work in the feature branch. Therefore, the extra testing is not needed. I have attempted to remind management of the numerous times "brainless" merges have failed. Thier solution is to instead of build and regression testing to have the developer diff the Feature branch and the newly merged Trunk. That process in thier mind would replace the regression testing I asked for. So what do you require when you reintegrate back to the Trunk? What are the issues that we will encounter if we remove this step and replace with the diff? Is the cost of staying Agile the additional work of the intergration of the branches? Thanks for any input. LoneCM

    Read the article

  • MSBuild Starter Kits - i.e. Copy and just start modding...

    - by vdh_ant
    Hi guys Just wondering if anyone knows if there are any MSBuild starter kits out there. What I mean by starter kits is that from the looks of it most builds to kinda the same sort of steps with minor changes here and there (i.e. most builds would run test, coverage, zip up the results, produce a report, deploy etc). Also what most people in general want from a CI build, test build, release build is mostly the same with minor changes here and there. Now don't get me wrong i think that most scripts are fairly different in the end. But I can't help but think that most start out life being fairly similar. Hence does anyone know of any "starter kits" that have like a dev/CI/test/release build with the common tasks that most people would want that you can just start changing and modifying? Cheers Anthony

    Read the article

  • Implementing events to communicate between two processes - C

    - by Jamie Keeling
    Hello all! I have an application consisting of two windows, one communicates to the other and sends it a struct constaining two integers (In this case two rolls of a dice). I will be using events for the following circumstances: Process a sends data to process b, process b displays data Process a closes, in turn closing process b Process b closes a, in turn closing process a I have noticed that if the second process is constantly waiting for the first process to send data then the program will be just sat waiting, which is where the idea of implementing threads on each process occurred and I have started to implement this already. The problem i'm having is that I don't exactly have a lot of experience with threads and events so I'm not sure of the best way to actually implement what I want to do. Following is a small snippet of what I have so far in the producer application; Create thread: case IDM_FILE_ROLLDICE: { hDiceRoll = CreateThread( NULL, // lpThreadAttributes (default) 0, // dwStackSize (default) ThreadFunc(hMainWindow), // lpStartAddress NULL, // lpParameter 0, // dwCreationFlags &hDiceID // lpThreadId (returned by function) ); } break; The data being sent to the other process: DWORD WINAPI ThreadFunc(LPVOID passedHandle) { HANDLE hMainHandle = *((HANDLE*)passedHandle); WCHAR buffer[256]; LPCTSTR pBuf; LPVOID lpMsgBuf; LPVOID lpDisplayBuf; struct diceData storage; HANDLE hMapFile; DWORD dw; //Roll dice and store results in variable storage = RollDice(); hMapFile = CreateFileMapping( (HANDLE)0xFFFFFFFF, // use paging file NULL, // default security PAGE_READWRITE, // read/write access 0, // maximum object size (high-order DWORD) BUF_SIZE, // maximum object size (low-order DWORD) szName); // name of mapping object if (hMapFile == NULL) { dw = GetLastError(); MessageBox(hMainHandle,L"Could not create file mapping object",L"Error",MB_OK); return 1; } pBuf = (LPTSTR) MapViewOfFile(hMapFile, // handle to map object FILE_MAP_ALL_ACCESS, // read/write permission 0, 0, BUF_SIZE); if (pBuf == NULL) { MessageBox(hMainHandle,L"Could not map view of file",L"Error",MB_OK); CloseHandle(hMapFile); return 1; } CopyMemory((PVOID)pBuf, &storage, (_tcslen(szMsg) * sizeof(TCHAR))); //_getch(); MessageBox(hMainHandle,L"Completed!",L"Success",MB_OK); UnmapViewOfFile(pBuf); return 0; } I'm trying to find out how I would integrate an event with the threaded code to signify to the other process that something has happened, I've seen an MSDN article on using events but it's just confused me if anything, I'm coming up on empty whilst searching on the internet too. Thanks for any help Edit: I can only use the Create/Set/Open methods for events, sorry for not mentioning it earlier.

    Read the article

  • How to ignore an error in Powershell and let it continue?

    - by Jake
    I am trying to see if a process is running on multiple servers and then format it into a table. get-process -ComputerName server1,server2,server3 -name explorer | Select-Object processname,machinename Thats the easy part - When the process does not exist or if the server is unavailable, powershell outputs a big ugly error, messes up the the table and doesn't continue. Example Get-Process : Couldn't connect to remote machine.At line:1 char:12 + get-process <<<< -ComputerName server1,server2,server3 -name explorer | format-table processname,machinename + CategoryInfo : NotSpecified: (:) [Get-Process], InvalidOperatio nException + FullyQualifiedErrorId : System.InvalidOperationException,Microsoft.Power Shell.Commands.GetProcessCommand How do I get around this? If the I would still like to get notified if the process isn't available or Running.

    Read the article

  • Using events in threads between processes - C

    - by Jamie Keeling
    Hello all! I have an application consisting of two windows, one communicates to the other and sends it a struct constaining two integers (In this case two rolls of a dice). I will be using events for the following circumstances: Process a sends data to process b, process b displays data Process a closes, in turn closing process b Process b closes a, in turn closing process a I have noticed that if the second process is constantly waiting for the first process to send data then the program will be just sat waiting, which is where the idea of implementing threads on each process occurred and I have started to implement this already. The problem i'm having is that I don't exactly have a lot of experience with threads and events so I'm not sure of the best way to actually implement what I want to do. Following is a small snippet of what I have so far in the producer application; Create thread: case IDM_FILE_ROLLDICE: { hDiceRoll = CreateThread( NULL, // lpThreadAttributes (default) 0, // dwStackSize (default) ThreadFunc(hMainWindow), // lpStartAddress NULL, // lpParameter 0, // dwCreationFlags &hDiceID // lpThreadId (returned by function) ); } break; The data being sent to the other process: DWORD WINAPI ThreadFunc(LPVOID passedHandle) { HANDLE hMainHandle = *((HANDLE*)passedHandle); WCHAR buffer[256]; LPCTSTR pBuf; LPVOID lpMsgBuf; LPVOID lpDisplayBuf; struct diceData storage; HANDLE hMapFile; DWORD dw; //Roll dice and store results in variable storage = RollDice(); hMapFile = CreateFileMapping( (HANDLE)0xFFFFFFFF, // use paging file NULL, // default security PAGE_READWRITE, // read/write access 0, // maximum object size (high-order DWORD) BUF_SIZE, // maximum object size (low-order DWORD) szName); // name of mapping object if (hMapFile == NULL) { dw = GetLastError(); MessageBox(hMainHandle,L"Could not create file mapping object",L"Error",MB_OK); return 1; } pBuf = (LPTSTR) MapViewOfFile(hMapFile, // handle to map object FILE_MAP_ALL_ACCESS, // read/write permission 0, 0, BUF_SIZE); if (pBuf == NULL) { MessageBox(hMainHandle,L"Could not map view of file",L"Error",MB_OK); CloseHandle(hMapFile); return 1; } CopyMemory((PVOID)pBuf, &storage, (_tcslen(szMsg) * sizeof(TCHAR))); //_getch(); MessageBox(hMainHandle,L"Completed!",L"Success",MB_OK); UnmapViewOfFile(pBuf); return 0; } I'm trying to find out how I would integrate an event with the threaded code to signify to the other process that something has happened, I've seen an MSDN article on using events but it's just confused me if anything, I'm coming up on empty whilst searching on the internet too. Thanks for any help Edit: I can only use the Create/Set/Open methods for events, sorry for not mentioning it earlier.

    Read the article

  • Adventures in Lab Management Configuration: CMMI Edition Part 1 of 3

    - by Enrique Lima
    I remember at one point someone telling me how close Migrate was to Migraine. This was a process that included an environment from TFS 2008 to TFS 2010, needed to be migrated too as far as the process template goes.  Here we are talking about CMMI v4.2 to CMMI v5.0.  Now, the process to migrate the TFS Infrastructure is one thing, migrating the Process Template is a different deal, not hard … just involved. Followed a combination of steps that came from a blog post as the main guidance and then MSDN (as suggested on the guidance post) to complement some tasks and steps. Again, the focus I have here is CMMI. The high level steps taken to enable the TFS 2008 CMMI v4.2 migrated to TFS 2010 Process Template are: 1)  Backup the Collection, Configuration and Warehouse Databases. 2)  Downloaded the Process Template using Visual Studio 2010. 3) Exported, modified and imported Bug Type Definition 4) Exported, modified and imported Scenario or Requirement Type Definition. 5) Created and imported bug field mappings. Now, we can attempt to connect using Test Manager, and you should be able to get this going. After that was done, it was time to enroll VMs that already existed in the environment.  This was a bit more challenging, but in the end it was a matter of just analyzing the changes that had been made to had a temporary work around from the time we migrated to the time we converted the Work Items and such and added fields to enable communication between the project and the Test and Lab Manager component. There are 2 more parts to this post, the second will describe the detailed steps taken to complete the Process Template update and the third will talk about the gotchas and fixes for the Lab Management portion.

    Read the article

  • Python Multiprocessing Question

    - by Avan
    My program has 2 parts divided into the core and downloader. The core handles all the app logic while the downloader just downloads urls. Right now, I am trying to use the python multiprocessing module to accomplish the task of the core as a process and the downloader as a process. The first problem I noticed was that if I spawn the downloader process from the core process so that the downloader is the child process and the core is the parent, the core process(parent) is blocked until the child is finished. I do not want this behaivor though. I would like to have a core process and a downloader process that are both able to execute their code and communicate between each other. example ... def main(): jobQueue = Queue() jobQueue.put("http://google.com) d = Downloader(jobQueue) p = Process(target=d.start()) p.start() if __name__ == '__main__': freeze_support() main() where Downloader's start() just takes the url out of the queue and downloads it. In order to have the 2 processes unblocked, would I need to create 2 processes from the parent process and then share something between them?

    Read the article

  • How to use IO.popen to write-to and read-from a child process?

    - by mackenir
    I am running net share from a ruby script to delete a windows network share. If files on the share are in use, net share will ask the user if they want to go ahead with the deletion, so my script needs to inspect the output from the command, and write out Y if it detects that net share is asking it for input. In order to be able to write out to the process I open it with access flags "r+". When attempting to write to the process with IO#puts, I get an error: Errno::EPIPE: Broken pipe What am I doing wrong here? (The error occurs on the line net_share.puts "Y") (The question text written out by net share is not followed by a newline, so I am using IO#readpartial to read in the output.) def delete_network_share share_path command = "net share #{share_path} /DELETE" net_share = IO.popen(command, "r+") text = "" while true begin line = net_share.readpartial 1000 #read all the input that's available rescue EOFError break end text += line if text.scan("Do you want to continue this operation? (Y/N)").size > 0 net_share.puts "Y" net_share.flush #probably not needed? end end net_share.close end

    Read the article

  • Find the closest palindrome number C#

    - by user294837
    Hi All, I am trying to create a console application that reads number from a file all underneath each other like so: 101 9 and then outputs into another file the closest palindrome number. So far what I have is not quite rightm i.e. I don't think I can put the class inside a method which is a bit more Java I was wandering if anyone could help at all? Thanks :) namespace PalidromeC { class Program { static public void Main(string[] args) { #region WriteAnswers try { StreamWriter sw = new StreamWriter("C://Temp/PalindromeAnswers.txt"); sw.WriteLine("Answers"); sw.Close(); }//try catch (Exception e) { Console.WriteLine("Exception: " + e.Message); }//catch #endregion #region ReadFile try { string numbers; StreamReader sr = new StreamReader("C://Temp/Palindrome.txt"); numbers = sr.ReadLine(); while (numbers != null) { Console.WriteLine(numbers); numbers = sr.ReadLine(); }//while sr.Close(); Console.ReadLine(); }//try catch (Exception e) { Console.WriteLine("Exception: " + e.Message); }//catch #endregion NearPalindromeFinder f = new NearPalindromeFinder(); int palindrome = f.findNearPalindrome(n); Console.WriteLine("Nearest Palindrome = " + palindrome); }//main public static void testFindNearestPalindrome(int n) { class NearPalindromeFinder { int findNearPalindrome(int start) { if (testPalindrome(start)) return start; else { int neg = start; int pos = start; for (int i = 0; i < start; i++) { if (testPalindrome(start-i)) { neg = start-i; break; }//if if (testPalindrome(start+i)) { pos = start+i; break; }//if }//for return (start == neg) ? pos : neg; }//else }//findNearPalindrome bool testPalindrome(int start) { if (start == 0 || start == 1) return true; String str = String.valueOf(start); String rev = new if (str.equals(rev)) return true; else return false; }//testPalindrome }//NearPalindromeFinder class }//testFindNearestPalindrome }//Program Class

    Read the article

  • Mouse Wheel Scroll - How can I capture the time interval between start and stop of scrolling?

    - by Rahat
    Is there any way to capture the time interval between mouse wheel scroll start and stop? Actually I want to capture the interval between the scrolling start and stop when I very quickly scroll the mouse wheel. I have already looked at MouseWheel event but it don't fulfill my requirement. In senes that it always gives a value of Delta 120 or -120 but i want to call a function depending on the speed of the mouse scroll for example when i scroll the mouse normally i want to perform function 1 and when i scrolled the mouse very quickly i want to perform the function 2. In other words is there any way to distinguish between the mouse scroll high and normal speed. Any advice will be appreciated.

    Read the article

  • Where to start openGL ES to create and rotate a cube in an iPhone?

    - by micropsari
    Hello. I've done some apps on iPhone using Objevtive-C and Cocos2d, and I'd like to start learning 3D. My first goal is to make a very simple app that: - Displays a 3D cube in the center of the screen. - And move the camera around the cube I want something very simple: no texture, no background. Where should I start to be able to do this app? I'm searching for some tutorials or exemples. Thanks!

    Read the article

  • How can you exclude folders from appearing in the Recent Items feature of Windows 7 start menu?

    - by Jordan Weinstein
    to be clear, I like the 'recent items' feature. I do not want to turn it off. I work at a law firm where we integrate Office with a document management system (DMS). If recent items are turned on, those DMS opened documents will show up in the recent items of a Windows 7 start menu when hovering over Word (or Excel\PPT etc). However the integration doesn't work correctly so if a user were to click on one of those, something wouldn't work right. In short, we've always needed to turn off Recent Items completely for a DMS integrated workstation. Curious if anyone knows of a way to exclude a directory from being "captured" so to speak. When you open a DMS document, the file gets copied to local directory where it saves it as you work, until you close and it checks it back in to the DMS. I'd like to be able to exclude that local directory from recent items. so local files in My Docs and Desktop would show up in recent items, but not DMS opened documents. Hope this makes sense.

    Read the article

  • Which options do I have for Java process communication?

    - by Dmitriy Matveev
    We have a place in a code of such form: void processParam(Object param) { wrapperForComplexNativeObject result = jniCallWhichMayCrash(param); processResult(result); } processParam - method which is called with many different arguments. jniCallWhichMayCrash - a native method which is intended to do some complex processing of it's parameter and to create some complex object. It can crash in some cases. wrapperForComplexNativeObject - wrapper type generated by SWIG processResult - a method written in pure Java which processes it's parameter by creation of several kinds (by the kinds I'm not meaning classes, maybe some like hierarchies) of objects: 1 - Some non-unique objects which are referencing each other (from the same hierarchy), these objects can have duplicates created from the invocations of processParam() method with different parameter values. Since it's costly to keep all the duplicates it's necessary to cache them. 2 - Some unique objects which are referencing each other (from the same hierarchy) and some of the objects of 1st kind. After processParam is executed for each of the arguments from some set the data created in processResult will be processed together. The problem is in fact that jniCallWhichMayCrash method may crash the entire JVM and this will be very bad. The reason of crash may be such that it can happen for one argument value and not for the other. We've decided that it's better to ignore crashes inside of JVM and just skip some chunks of data when such crashes occur. In order to do this we should run processParam function inside of separate process and pass the result somehow (HOW? HOW?! This is a question) to the main process and in case of any crashes we will only lose some part of data (It's ok) without lose of everything else. So for now the main problem is implementation of transport between different processes. Which options do I have? I can think about serialization and transmitting of binary data by the streams, but serialization may be not very fast due to object complexity. Maybe I have some other options of implementing this?

    Read the article

  • App Pool crashes before loading mscorsvr. How to troubleshoot?

    - by codepoke
    I have an app pool that recycles every 29 hours, per default. It recycles smoothly 9 times out of 10, and I'm pretty sure the recycle itself is good for the app. Once every couple weeks the recycle does not work. The old worker process dies cleanly and the new worker process starts, but will not serve up content. Recycling the app pool again manually works like a charm. The failed worker process stops and dies cleanly and a second new worker process fires up and serves content perfectly. I took a crash dump against the failed worker process prior to recycling it, and DebugDiag found nothing to complain about. I tried to dig a little deeper using WinDBG, but mscorsvr/mscorwks is not loaded yet 15 minutes after the new process started. There are 14 threads running (4 async) and 20 pending client connections, but .NET is not even loaded into the process yet. Any suggestions where to poke and prod to find a root cause on this?

    Read the article

  • Which work process in my company should I Improve first?

    - by shoren
    I've just started to work in a new place, and I see several things they do that I find really terrible, and I want to know if they are indeed so wrong, or I am just too strict. Please let me know if my criticism is in place, and your opinion on which problem is the worst and should be fixed first. The developement is all in Java. 1) Not using svnignore. This means svn stat can't be used, and developers forget to add files and break the build. 2) Generated files go to same folders as committed files. Can't use simple maven clean, have to find them one by one. Maven clean doesn't remove all of them. 3) Not fixing IDE analyze warnings. Analyze code returns about 5,000 warning, of many different kinds. 4) Not following conventions: spring beans names sometimes start with uppercase and sometimes not, ant properties sometimes with underline and sometimes with dots delimiter, etc. 5) Incremental build takes 6 minutes, even when nothing is changed. 6) Developers only use remote debug, and don't know how to run the Tomcat server internally from the IDE. 7) Developers always restart the server after every compilation, instead of dynamically reloading the class and saving the server's state. It takes them at least 10 minutes to start checking any change in the code. 8) Developers only compile from command line. When there are compilation errors, they manually open the file and go the the problematic line. 9) A complete mess in project dependencies. Over 200 open sources are depended on, and no one knows what is indeed needed and why. They do know that not all dependencies are necessary. 10) Mixing Maven and Ant in a way that disables the benefits of both. In one case, even dependency checks are not done by Maven. 11) Not using generics properly. 12) Developers don't use Subversion integration with IDE (Eclipse, Intellij Idea). What do you think? Where should I start? Is any of the things I mentioned not really a problem?

    Read the article

  • Exploring TCP throughput with DTrace (2)

    - by user12820842
    Last time, I described how we can use the overlap in distributions of unacknowledged byte counts and send window to determine whether the peer's receive window may be too small, limiting throughput. Let's combine that comparison with a comparison of congestion window and slow start threshold, all on a per-port/per-client basis. This will help us Identify whether the congestion window or the receive window are limiting factors on throughput by comparing the distributions of congestion window and send window values to the distribution of outstanding (unacked) bytes. This will allow us to get a visual sense for how often we are thwarted in our attempts to fill the pipe due to congestion control versus the peer not being able to receive any more data. Identify whether slow start or congestion avoidance predominate by comparing the overlap in the congestion window and slow start distributions. If the slow start threshold distribution overlaps with the congestion window, we know that we have switched between slow start and congestion avoidance, possibly multiple times. Identify whether the peer's receive window is too small by comparing the distribution of outstanding unacked bytes with the send window distribution (i.e. the peer's receive window). I discussed this here. # dtrace -s tcp_window.d dtrace: script 'tcp_window.d' matched 10 probes ^C cwnd 80 10.175.96.92 value ------------- Distribution ------------- count 1024 | 0 2048 | 4 4096 | 6 8192 | 18 16384 | 36 32768 |@ 79 65536 |@ 155 131072 |@ 199 262144 |@@@ 400 524288 |@@@@@@ 798 1048576 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 3848 2097152 | 0 ssthresh 80 10.175.96.92 value ------------- Distribution ------------- count 268435456 | 0 536870912 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 5543 1073741824 | 0 unacked 80 10.175.96.92 value ------------- Distribution ------------- count -1 | 0 0 | 1 1 | 0 2 | 0 4 | 0 8 | 0 16 | 0 32 | 0 64 | 0 128 | 0 256 | 3 512 | 0 1024 | 0 2048 | 4 4096 | 9 8192 | 21 16384 | 36 32768 |@ 78 65536 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 5391 131072 | 0 swnd 80 10.175.96.92 value ------------- Distribution ------------- count 32768 | 0 65536 |@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ 5543 131072 | 0 Here we are observing a large file transfer via http on the webserver. Comparing these distributions, we can observe: That slow start congestion control is in operation. The distribution of congestion window values lies below the range of slow start threshold values (which are in the 536870912+ range), so the connection is in slow start mode. Both the unacked byte count and the send window values peak in the 65536-131071 range, but the send window value distribution is narrower. This tells us that the peer TCP's receive window is not closing. The congestion window distribution peaks in the 1048576 - 2097152 range while the receive window distribution is confined to the 65536-131071 range. Since the cwnd distribution ranges as low as 2048-4095, we can see that for some of the time we have been observing the connection, congestion control has been a limiting factor on transfer, but for the majority of the time the receive window of the peer would more likely have been the limiting factor. However, we know the window has never closed as the distribution of swnd values stays within the 65536-131071 range. So all in all we have a connection that has been mildly constrained by congestion control, but for the bulk of the time we have been observing it neither congestion or peer receive window have limited throughput. Here's the script: #!/usr/sbin/dtrace -s tcp:::send / (args[4]-tcp_flags & (TH_SYN|TH_RST|TH_FIN)) == 0 / { @cwnd["cwnd", args[4]-tcp_sport, args[2]-ip_daddr] = quantize(args[3]-tcps_cwnd); @ssthresh["ssthresh", args[4]-tcp_sport, args[2]-ip_daddr] = quantize(args[3]-tcps_cwnd_ssthresh); @unacked["unacked", args[4]-tcp_sport, args[2]-ip_daddr] = quantize(args[3]-tcps_snxt - args[3]-tcps_suna); @swnd["swnd", args[4]-tcp_sport, args[2]-ip_daddr] = quantize((args[4]-tcp_window)*(1 tcps_snd_ws)); } One surprise here is that slow start is still in operation - one would assume that for a large file transfer, acknowledgements would push the congestion window up past the slow start threshold over time. The slow start threshold is in fact still close to it's initial (very high) value, so that would suggest we have not experienced any congestion (the slow start threshold is adjusted when congestion occurs). Also, the above measurements were taken early in the connection lifetime, so the congestion window did not get a changes to get bumped up to the level of the slow start threshold. A good strategy when examining these sorts of measurements for a given service (such as a webserver) would be start by examining the distributions above aggregated by port number only to get an overall feel for service performance, i.e. is congestion control or peer receive window size an issue, or are we unconstrained to fill the pipe? From there, the overlap of distributions will tell us whether to drill down into specific clients. For example if the send window distribution has multiple peaks, we may want to examine if particular clients show issues with their receive window.

    Read the article

< Previous Page | 175 176 177 178 179 180 181 182 183 184 185 186  | Next Page >