Search Results

Search found 35692 results on 1428 pages for 'build process'.

Page 384/1428 | < Previous Page | 380 381 382 383 384 385 386 387 388 389 390 391  | Next Page >

• Linking with Boost error

  - by drhorrible

  I just downloaded and ran the boost installer for version 1.42 (from boostpro.com), and set up my project according to the getting started guide. However, when I build the program, I get this linker error: LINK : fatal error LNK1104: cannot open file 'libboost_program_options-vc90-mt-gd-1_42.lib' The build log adds this (I've replaced project-specific paths with *'s): Creating temporary file "******\Debug\RSP00001252363252.rsp" with contents [ /OUT:"*********.exe" /INCREMENTAL /LIBPATH:"C:\Program Files\boost\boost_1_42_0\lib" /MANIFEST /MANIFESTFILE:"Debug\hw6.exe.intermediate.manifest" /MANIFESTUAC:"level='asInvoker' uiAccess='false'" /DEBUG /PDB:"********\Debug\***.pdb" /SUBSYSTEM:CONSOLE /DYNAMICBASE /NXCOMPAT /MACHINE:X86 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib ".\Debug\****.obj" ".\Debug\****.exe.embed.manifest.res" ] Creating command line "link.exe @********\Debug\RSP00001252363252.rsp /NOLOGO /ERRORREPORT:PROMPT" I've also emailed [email protected] (with a message very similar to this), but I thought maybe so would be faster.

  Read the article

• Trapping Error Status in MSBuild

  - by Ruben Bartelink

  As part of some build automation of running xUnit.net tests with MSBuild, I'm running into a case where I need to loop over a batch of items. Inside the loop, I need to detect whether an iteration failed, but I want to continue executing regardless. Then after the batched bit, I need to know whether one or more errors have occurred in order to report the outcome to TeamBuild. IOW, in pseudocode: Task Name=RunTests CreateItems ForEach item CallTarget Target=RunTest ContinueOnError=true CombineNUnitResults Report success/failure Task Name=RunTest XUnit item I'm hoping this can be achieved without a custom task (or hacking the xunit.net MSBuild task as Jonne did). (But willing to use MSBuild Community or Sdc tasks) And @BradWilson: I this is isnt possible to do cleanly, I'll be looking for Jonne's change a la the NUnit task to also make it into the xunit task See also: http://stackoverflow.com/questions/517560/how-do-i-get-team-build-to-show-test-results-and-coverage-for-xunit-net-test-suit

  Read the article

• Difference between Setting.settings and web.config?

  - by Muneeb

  This might sound a bit dumb. I always had this impression that web.config should store all settings which are suspect to change post-build and setting.settings should have the one which may change pre-build. but I have seen projects which had like connection string in setting.settings. Connection Strings should always been in web.config, shouldnt it? I am interested in a design perspective answer. Just a bit of background: My current scenario is that I am developing a web application with all the three tiers abstracted in three separate visual studio projects thus every tier has its own .settings and .config file.

  Read the article

• Programming pattern to flatten deeply nested ajax callbacks?

  - by chiborg

  I've inherited JavaScript code where the success callback of an Ajax handler initiates another Ajax call where the success callback may or may not initiate another Ajax call. This leads to deeply nested anonymous functions. Maybe there is a clever programming pattern that avoids the deep-nesting and is more DRY. jQuery.extend(Application.Model.prototype, { process: function() { jQuery.ajax({ url:myurl1, dataType:'json', success:function(data) { // process data, then send it back jQuery.ajax({ url:myurl2, dataType:'json', success:function(data) { if(!data.ok) { jQuery.ajax({ url:myurl2, dataType:'json', success:mycallback }); } else { mycallback(data); } } }); } }); } });

  Read the article

• HTTP Headers for Unknown Content-Length

  - by jocull

  I am currently trying to stream content out to the web after a trans-coding process. This usually works fine by writing binary out to my web stream, but some browsers (specifically IE7, IE8) do not like not having the Content-Length defined in the HTTP header. I believe that "valid" headers are supposed to have this set. What is the proper way to stream content to the web when you have an unknown Content-Length? The trans-coding process can take awhile, so I want to start streaming it out as it completes.

  Read the article

• error A2070: invalid instruction operands IN SSE MASM64

  - by Green

  when compiling this in ml64.exe 64bit (masm64) the SSE command give me an error what do i need to do to include the SSE commands in 64 bit? .code test PROC movlps [rdx], xmm7 ;;error A2070: invalid instruction operands ;//Inc in vec ptr add rsi, 16 movhlps xmm6, xmm7 movss [rdx+8], xmm6 ;;rror A2070: invalid instruction operands ret test ENDP end i get the error: 1>Performing Custom Build Step 1> Assembling: extasm.asm 1>extasm.asm(6) : error A2070: invalid instruction operands 1>extasm.asm(10) : error A2070: invalid instruction operands 1>Microsoft (R) Macro Assembler (x64) Version 8.00.50727.215 1>Copyright (C) Microsoft Corporation. All rights reserved. 1>Project : error PRJ0019: A tool returned an error code from "Performing Custom Build Step"

  Read the article

• An Object reference is required for the non-static field

  - by Muhammad Akhtar

  I have make my existing method to static method to get access in javascript, like.. [WebMethod(EnableSession = true), ScriptMethod()] public static void Build(String ID) { Control releaseControl = LoadControl("~/Controls/MyControl.ascx"); //An Object reference is required for the non-static field, mthod or property // 'System.Web.UI.TemplateControl.LoadControl(string)' plc.Controls.Add(releaseControl); // where plc is place holder control //object reference is required for the nonstatic field, method, or property '_Default.pl' } When I build I am getting error and I have posted these in comments below each line before converted it to static method, it working perfectly. Please suggest me the solution of my issue. Thanks

  Read the article

• Preserving DataRowState when serializing DataSet using DataContractSerializer

  - by user349453

  For various reasons I am having to send a typed dataset to a WCF service endpoint. This works fine except that upon Deserializing, the RowState of each row in each DataTable is set to 'Added', regardless of what they were on the client. If I write the serialized stream out to a file, I see that the RowState is not part of the Serialized data. How can I add this so that I can preserve the RowState across service boundaries? Not that I think it matters, but the client process is running .net 3.5 while the service process is running .net 4.0

  Read the article

• SSL cert install in Mongrel

  - by normalocity

  Running a RoR app. I've created a self-signed SSL cert for my test/dev environment, which consists of a single mongrel, and Mac OS X Leopard. Can I install the SSL cert in such a way that mongrel will use it, or do I have build a different server stack to handle SSL (e.g. Apache in front of a mongrel cluster), and install the cert inside of Apache? For my production server, I'll eventually build a server stack like this, so it's no big deal if I have to go that direction - it's just a question of now or later. Thanks!

  Read the article

• Permission denied: cannot call non-public or static methods remotely.

  - by rstat1

  Ok I've found a solution to this particular error message on here already. But my case is slightly different. There are no "non-public" or "static" methods in my code. All are public. What I'm trying to do is pass a FrameworkElement (more specifically a web browser control) that was created in one process over to another process for display and use. Also I'm not using (and would to avoid using) any of the framework 3.5 addin stuff. Fails at the following line everytime. fe = FrameworkElementAdapters.ContractToViewAdapter(tab.ReturnBrowserObject) tab.ReturnBrowserObject returns an INativeHandleContract which the above line is suppose to convert to a FrameworkElement.

  Read the article

• How can I turn the structure of an XML file into a folder structure using ANT

  - by 1ndivisible

  I would like to be able to pass an XML file to an ANT build script and have it create a folder structure mimicking the nodal structure of the XML, using the build files parent directory as the root. For Example using: <root> <folder1> <folder1-1/> </folder1> <folder2/> <folder3> <folder3-1/> </folder3> </root> ant would create: folder1 -folder1-1 folder2 folder3 -folder3-1 I know how to create a directory, but i'm not sure how to have ANT parse the XML.

  Read the article

• Rails User-Profile model challenges

  - by Craig

  I am attempting to create an enrollment process similar to SO's: route to an OpenID provider provider returns the user's information to the UsersController (a guess) UsersController creates user, then routes to the ProfilesController's new or edit action. For now, I'm simply trying to create the user, then route to the ProfilesController's new or edit action (not sure which I should be using). Here's what I have thus far: Models: class User < ActiveRecord::Base has_one :profile end class Profile < ActiveRecord::Base belongs_to :user end Routes: map.resources :users do |user| user.resource :profile end new_user_profile GET /users/:user_id/profile/new(.:format) {:controller=>"profiles", :action=>"new"} edit_user_profile GET /users/:user_id/profile/edit(.:format) {:controller=>"profiles", :action=>"edit"} user_profile GET /users/:user_id/profile(.:format) {:controller=>"profiles", :action=>"show"} PUT /users/:user_id/profile(.:format) {:controller=>"profiles", :action=>"update"} DELETE /users/:user_id/profile(.:format) {:controller=>"profiles", :action=>"destroy"} POST /users/:user_id/profile(.:format) {:controller=>"profiles", :action=>"create"} users GET /users(.:format) {:controller=>"users", :action=>"index"} POST /users(.:format) {:controller=>"users", :action=>"create"} new_user GET /users/new(.:format) {:controller=>"users", :action=>"new"} edit_user GET /users/:id/edit(.:format) {:controller=>"users", :action=>"edit"} user GET /users/:id(.:format) {:controller=>"users", :action=>"show"} PUT /users/:id(.:format) {:controller=>"users", :action=>"update"} DELETE /users/:id(.:format) {:controller=>"users", :action=>"destroy"} Controllers: class UsersController < ApplicationController # generate new-user form def new @user = User.new end # process new-user-form post def create @user = User.new(params[:user]) if @user.save redirect_to new_user_profile_path(@user) ... end end # generate edit-user form def edit @user = User.find(params[:id]) end # process edit-user-form post def update @user = User.find(params[:id]) respond_to do |format| if @user.update_attributes(params[:user]) flash[:notice] = 'User was successfully updated.' format.html { redirect_to(users_path) } format.xml { head :ok } ... end end end class ProfilesController < ApplicationController before_filter :get_user def get_user @user = User.find(params[:user_id]) end # generate new-profile form def new @user.profile = Profile.new @profile = @user.profile end # process new-profile-form post def create @user.profile = Profile.new(params[:profile]) @profile = @user.profile respond_to do |format| if @profile.save flash[:notice] = 'Profile was successfully created.' format.html { redirect_to(@profile) } format.xml { render :xml => @profile, :status => :created, :location => @profile } ... end end end # generate edit-profile form def edit @profile = @user.profile end # generate edit-profile-form post def update @profile = @user.profile respond_to do |format| if @profile.update_attributes(params[:profile]) flash[:notice] = 'Profile was successfully updated.' # format.html { redirect_to(@profile) } format.html { redirect_to(user_profile(@user)) } format.xml { head :ok } else format.html { render :action => "edit" } format.xml { render :xml => @profile.errors, :status => :unprocessable_entity } end end end Edit-User View: ... <% form_for(@user) do |f| %> ... New-Profile View: ... <% form_for([@user,@profile]) do |f| %> .. I'm having two problems: When saving an edit to the User model, the UsersController attempts to route to http://localhost:3000/users/1/profile.%23%3Cprofile:0x10438e3e8%3E, instead of http://localhost:3000/users/1/profile When the new-profile form is being rendered, it throws an error that reads: undefined method `user_profiles_path' for # Is it better to create a blank profile when the user is created (in the UsersController), then edit it OR follow the rest-ful convention of creating the profile in the ProfilesController (as I have done)? What am I missing? I did review Associating Two Models in Rails (user and profile), but it didn't address my needs. Thanks for your time.

  Read the article

• How do I create an Access 2003 MDE programmatically or by command line in Access 2007?

  - by Ned Ryerson

  I have a legacy Access 2003 database file that must remain in that format to preserve its menus and toolbars. I have recently moved to Access 2007 in my build environment and will be deploying the compiled Access 2003 program with the Access 2007 runtime. In Access 2003, I could script the process of creating an MDE with the Access Developer Extensions (WZADE.mde) using the command line and an .xml file of build preferences (without creating an install package). The Access 2007 developer extensions do not seem to offer a similar option. I can "Package a Solution", but it creates an accdr and buries it in a CD installer. I've tried programmatic options like Docmd.RunCommand acMakeMDEFILe and Syscmd(603, mdbpath, mdepath) but they no longer work in Access 2007. Of course, i can manually create an MDE using Database ToolsCreate MDE, but that is no scriptable as far as I can tell.

  Read the article

• CruiseControl.Net Publisher email failing to send e-mail

  - by CrimsonX

  So here's my problem: I cannot seem to be able to configure CruiseControl.NET to send out an e-mail to me when a build occurs. I copied the example from the documentation and filled it in with my own values. http://confluence.public.thoughtworks.org/display/CCNET/Email+Publisher Here's the relevant section of ccnet.config below <publishers> <merge> <files> <file>C:\Build\Temp\*.xml</file> </files> </merge> <xmllogger logDir="buildlogs" /> <statistics> <statisticList> <statistic /> </statisticList> </statistics> <email includeDetails="TRUE" mailhostUsername="user" mailhostPassword="password" useSSL="TRUE"> <from>[email protected]</from> <mailhost>mail.mycompanysmtpserver.com</mailhost> <users> <user name="MyName Lastname" group="buildmaster" address="[email protected]" /> </users> <groups> <group name="buildmaster"> <notifications> <notificationType>Always</notificationType> </notifications> </group> </groups> <modifierNotificationTypes> <NotificationType>Failed</NotificationType> <NotificationType>Fixed</NotificationType> </modifierNotificationTypes> <subjectSettings> <subject buildResult="StillBroken" value="Build is still broken for {CCNetProject}" /> </subjectSettings> </email> </publishers> I have had a successful CruiseControl.NET server configured for some time, and it successfully updates people through CCTray, but I need to add e-mail support as well. I've already looked at the relevant StackOverflow articles like this one and many more and tried my hand at googling the solution but I dont know what I could be doing wrong. The only other thing that I'd like to validate is that I can send/receive e-mails using my username/password with the SMTP server I received from IT (my mail works properly) but I just followed the steps on testing an SMTP server here and everything looks fine http://support.microsoft.com/kb/153119 Anyone have any ideas as to why I'm getting this problem, or how to troubleshoot it further?

  Read the article

• SSAS/SSRS remove parameter from cube report destroys report

  - by Jon

  Group, We built a data cube using SSAS and are now building SSRS reports off of that cube. Not sure if anyone has come across this, but when you build the report using the wizard and include parameters all looks fine. However if you are in the report after the wizard is compete, and you decide you want to remove one of the parameters you created it debunks the report and the only way to get it back is to re-create the whole report. Any way you can remove or add parameters after the initial build without destroying your report? Thanks in advance for the help! I love this forumn!

  Read the article

• Lambda Expression to be used in Select() query

  - by jameschinnock

  Hi, I am trying to build a lambda expression, containing two assignments (as shown further down), that I can then pass to a Queryable.Select() method. I am trying to pass a string variable into a method and then use that variable to build up the lambda expression so that I can use it in a LINQ Select query. My reasoning behind it is that I have a SQL Server datasource with many column names, I am creating a charting application that will allow the user to select, say by typing in the column name, the actual column of data they want to view in the y-axis of my chart, with the x-axis always being the DateTime. Therefore, they can essentially choose what data they chart against the DateTime value (it’s a data warehouse type app). I have, for example, a class to store the retrieved data in, and hence use as the chart source of: public class AnalysisChartSource { public DateTime Invoicedate { get; set; } public Decimal yValue { get; set; } } I have (purely experimentaly) built an expression tree for the Where clause using the String value and that works fine: public void GetData(String yAxis) { using (DataClasses1DataContext db = new DataClasses1DataContext()) { var data = this.FunctionOne().AsQueryable<AnalysisChartSource>(); //just to get some temp data in.... ParameterExpression pe = Expression.Parameter(typeof(AnalysisChartSource), "p"); Expression left = Expression.MakeMemberAccess(pe, typeof(AnalysisChartSource).GetProperty(yAxis)); Expression right = Expression.Constant((Decimal)16); Expression e2 = Expression.LessThan(left, right); Expression expNew = Expression.New(typeof(AnalysisChartSource)); LambdaExpression le = Expression.Lambda(left, pe); MethodCallExpression whereCall = Expression.Call( typeof(Queryable), "Where", new Type[] { data.ElementType }, data.Expression, Expression.Lambda<Func<AnalysisChartSource, bool>>(e2, new ParameterExpression[] { pe })); } } However……I have tried a similar approach for the Select statement, but just can’t get it to work as I need the Select() to populate both X and Y values of the AnalysisChartSource class, like this: .Select(c => new AnalysisChartSource { Invoicedate = c.Invoicedate, yValue = c.yValue}).AsEnumerable(); How on earth can I build such an expression tree….or….possibly more to the point…..is there an easier way that I have missed entirely?

  Read the article

• Implementing google dashboard type interface in asp.net

  - by Sam_Cogan

  I'm looking at implementing a Google IG type dashboard in a .net app. There are a number of options I've found to do this, and i'm trying to establish what is going to be the best to use, in terms of speed, versatility etc. So far the options I am looking at are either to use asp.net webparts and .net Ajax, this would make it quicker to build, but I'm concerned this is going to make the application bulky and slow, or using JQuery, and either .net MVC or Webforms, to custom build an interface. Does anyone have any thoughts on what the best option may be, or any options I may have missed? All I want to do here is to allow users to customise a dashboard with a number of components (which will be user controls). I do also have access to Telerik controls, but I'm not sure if they would be any use here.

  Read the article

• Using .net 3.5 assemblies in asp.net 2.0 web application

  - by masterik

  I have an .net assembly build against 3.5 framework. This assembly has a class Foo with two method overrides: public class Foo { public T Set<T>(T value); public T Set<T>(Func<T> getValueFunc); } I'm referencing this assembly in my asp.net 2.0 web application to use first override of the Set method (without Func). But on build I get an error saying that I should reference System.Core to use System.Func delegate... but I'm not using this type... Is there a workaround to solve this? PS: There is no option to convert my web application targeting 3.5 framework.

  Read the article

• Unity and Object Creation

  - by William

  I am using unity as my IoC container. I am trying to implement a type of IProviderRepository. The concrete implementation has a constructor that accepts a type of IRepository. When I remove the constructor parameter from the concrete implementation everything works fine. I am sure the container is wired correctly. When I try to create the concrete object with the constructor I receive the following error: "The current build operation (build key Build Key[EMRGen.Infrastructure.Data.IRepository1[EMRGen.Model.Provider.Provider], null]) failed: The current type, EMRGen.Infrastructure.Data.IRepository1[EMRGen.Model.Provider.Provider], is an interface and cannot be constructed. Are you missing a type mapping? (Strategy type BuildPlanStrategy, index 3)". Is it possible to achieve the above mention functionality with Unity? Namely have Unity infer a concrete type from the Interface and also inject the constructor of the concrete type with the appropriate concrete object based on constructor parameters. Below is sample of my types defined in Unity and a skeleton class listing for what I want to achieve. IProviderRepository is implemented by ProviderRepository which has a constructor that expects a type of IRepository. <typeAlias alias="ProviderRepositoryInterface" type="EMRGen.Model.Provider.IProviderRepository, EMRGen.Model" /> <typeAlias alias="ProviderRepositoryConcrete" type="EMRGen.Infrastructure.Repositories.Providers.ProviderRepository, EMRGen.Infrastructure.Repositories" /> <typeAlias alias="ProviderGenericRepositoryInterface" type="EMRGen.Infrastructure.Data.IRepository`1[[EMRGen.Model.Provider.IProvider, EMRGen.Model]], EMRGen.Infrastructure" /> <typeAlias alias="ProviderGenericRepositoryConcrete" type="EMRGen.Infrastructure.Repositories.EntityFramework.ApplicationRepository`1[[EMRGen.Model.Provider.Provider, EMRGen.Model]], EMRGen.Infrastructure.Repositories" /> <!-- Provider Mapping--> <typeAlias alias="ProviderInterface" type="EMRGen.Model.Provider.IProvider, EMRGen.Model" /> <typeAlias alias="ProviderConcrete" type="EMRGen.Model.Provider.Doctor, EMRGen.Model" /> //Illustrate the call being made inside my class public class PrescriptionService { PrescriptionService() { IUnityContainer uc = UnitySingleton.Instance.Container; UnityServiceLocator unityServiceLocator = new UnityServiceLocator(uc); ServiceLocator.SetLocatorProvider(() => unityServiceLocator); IProviderRepository pRepository = ServiceLocator.Current.GetInstance<IProviderRepository>(); } } public class GenericRepository<IProvider> : IRepository<IProvider> { } public class ProviderRepository : IProviderRepository { private IRepository<IProvider> _genericProviderRepository; //Explict public default constructor public ProviderRepository(IRepository<IProvider> genericProviderRepository) { _genericProviderRepository = genericProviderRepository; } }

  Read the article

• AJAX: how to get progress feedback in web apps, and to avoid timeouts on long requests?

  - by David Dombrowsky

  This is a general design question about how to make a web application that will receive a large amount of uploaded data, process it, and return a result, all without the dreaded spinning beach-ball for 5 minutes or a possible HTTP timeout. Here's the requirements: make a web form where you can upload a CSV file containing a list of URLs when the user clicks "submit", the server fetches the file, and checks each URL to see if its alive, and what the title tag of the page is. the result is a downloadable CSV file containing the URL, and the result HTTP code the input CSV can be very large ( 100000 rows), so the fetch process might take 5-30 minutes. My solution so far is to have a spinning javascript loop on the client site, which queries the server every second to determine the overall progress of the job. This seems kludgy to me, and I'm hesitant to accept this as the best solution. I'm using perl, template toolkit, and jquery, but any solution using any web technology would be acceptable.

  Read the article

• Writing an Iron Python debugger

  - by Kragen

  As a learning exercise I'm writing myself a simple extension / plugin / macro framework using IronPython - I've gotten the basics working but I'd like to add some basic debugging support to make my script editor easier to work with. I've been hunting around on the internet a bit and I've found a couple of good resources on writing managed debuggers (including Mike Stall's excellent .Net Debugging blog and the MSDN documentaiton on the CLR Debugging API) - I understand that IronPython is essentially IL however apart from that I'm a tad lost on how to get started, in particular: Are there any significant differences between debugging a dynamic language (such as IronPython) to a static one (such as C#)? Do I need to execute my script in a special way to get IronPython to output suitable debugging information? Is debugging a script running inside the current process going to cause deadlocks, or does IronPython execute my script in a child process? Am I better off looking into how to produce a simple C# debugger first to get the general idea? (I'm not interested in the GUI aspect of making a debugger for now - I've already got a pretty good idea of how this might work)

  Read the article

• MathML or OMML to PNG w/ .NET?

  - by charliedigital

  Are there any libraries which take MathML (or, even more preferably, OMML) and outputs a .PNG file? I am putting together an export process for .docx files and, as a part of this process, I'd like to extract equations and render them as .PNG files. Word 2007 does this natively when you save a document for the web, but so far, I have not been able to find a way to do this programmatically (if anyone has an answer for that, it would be even better). So the next best thing is to take the OMML and use the Microsoft provided XSL stylesheets and transform them to MathML. Unfortunately, I haven't been able to find any (working) rendering libraries for either MathML or OMML. If there aren't any pure .NET libraries for this, I'll settle for just about anything that I can call from a commandline to output a .PNG from either MathML or OMML.

  Read the article

• Silverlight 4.0 question - synchronous calls to asmx web service

  - by Anvar

  Hi, I have Silverlight problem. I have to deal with fairly large legacy application that has some web services exposed (regular asmx, not wcf). What I need to do is to build Silverlight app that consumes those web services. Due to business logic I need to have the same web method to be called from Silverlight app synchronously. I was able to build web services consumption but only asynchronously because that would be default Silverlight behavior. Is there a way to make asynchronous calls synchronous in Silverlight? I looked here and googled around but came across only wcf examples. I would appreciate if somebody would get me a code example for regular asmx web service. I use Silverlight 4.0. Thanks!

  Read the article

• qemu-kvm virtual machine virtio network freeze under load

  - by Rick Koshi

  I'm having a problem with my virtual machines, where the network will freeze under heavy load. I'm using CentOS 6.2 as both host and guest, not using libvirt, just running qemu-kvm directly as follows: /usr/libexec/qemu-kvm \ -drive file=/data2/vm/rb-dev2-www1-vm.img,index=0,media=disk,cache=none,if=virtio \ -boot order=c \ -m 2G \ -smp cores=1,threads=2 \ -vga std \ -name rb-dev2-www1-vm \ -vnc :84,password \ -net nic,vlan=0,macaddr=52:54:20:00:00:54,model=virtio \ -net tap,vlan=0,ifname=tap84,script=/etc/qemu-ifup \ -monitor unix:/var/run/vm/rb-dev2-www1-vm.mon,server,nowait \ -rtc base=utc \ -device piix3-usb-uhci \ -device usb-tablet /etc/qemu-ifup (used by the above command) is a very simple script, containing the following: #!/bin/sh sudo /sbin/ifconfig $1 0.0.0.0 promisc up sudo /usr/sbin/brctl addif br0 $1 sleep 2 And here's the info on br0 and other interfaces: avl-host3 14# brctl show bridge name bridge id STP enabled interfaces br0 8000.180373f5521a no bond0 tap84 virbr0 8000.525400858961 yes virbr0-nic avl-host3 15# ip addr show 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 scope host lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: em1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc mq master bond0 state UP qlen 1000 link/ether 18:03:73:f5:52:1a brd ff:ff:ff:ff:ff:ff 3: em2: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc mq master bond0 state UP qlen 1000 link/ether 18:03:73:f5:52:1a brd ff:ff:ff:ff:ff:ff 4: em3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 18:03:73:f5:52:1e brd ff:ff:ff:ff:ff:ff 5: em4: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 18:03:73:f5:52:20 brd ff:ff:ff:ff:ff:ff 6: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP link/ether 18:03:73:f5:52:1a brd ff:ff:ff:ff:ff:ff inet6 fe80::1a03:73ff:fef5:521a/64 scope link valid_lft forever preferred_lft forever 7: br0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN link/ether 18:03:73:f5:52:1a brd ff:ff:ff:ff:ff:ff inet 172.16.1.46/24 brd 172.16.1.255 scope global br0 inet6 fe80::1a03:73ff:fef5:521a/64 scope link valid_lft forever preferred_lft forever 8: virbr0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN link/ether 52:54:00:85:89:61 brd ff:ff:ff:ff:ff:ff inet 192.168.122.1/24 brd 192.168.122.255 scope global virbr0 9: virbr0-nic: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 500 link/ether 52:54:00:85:89:61 brd ff:ff:ff:ff:ff:ff 12: tap84: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 500 link/ether ba:e8:9b:2a:ff:48 brd ff:ff:ff:ff:ff:ff inet6 fe80::b8e8:9bff:fe2a:ff48/64 scope link valid_lft forever preferred_lft forever bond0 is a bond of em1 and em2. virbr0 and virbr0-nic are vestigial interfaces left over from CentOS's default installation. They are unused (as far as I know). The guest runs perfectly until I run a large 'rsync', when the network will freeze after some seemingly-random time (usually under a minute). When it freezes, there is no network activity in or out of the guest. I can still connect to the guest's console via vnc, but it is unable to speak out its network interface. Any attempt to 'ping' from the guest gives a "Destination Host Unreachable" error for 3/4 packets and no reply for every fourth packet. Sometimes (perhaps two thirds of the time), I can bring the interface back to life by doing a "service network restart" from the guest's console. If this works (and if I do it before the rsync times out), the rsync will resume. Usually it will freeze again within a minute or two. If I repeat, the rsync will eventually finish, and I presume the machine goes back to waiting for another period of heavy load. Throughout the whole process, there are no console errors or relevant (that I can see) syslog messages on either guest or host machine. If the "service network restart" doesn't work the first time, trying again (and again and again) never seems to work. The command completes normally, with normal output, but the interface stays frozen. However, a soft reboot of the guest machine (without restarting qemu-kvm) always seems to bring it back. I am aware of the "lowest mac address" assignment problem, where the bridge takes on the mac address of the slave interface with the lowest mac address. This causes temporary network freezes, but is definitely not what's happening for me. My freezes are permanent until manual intervention, and you can see from the 'ip addr show' output above that the mac address being used by br0 is that of the physical ethernet. There are no other virtual machines running on the host. I've verified that each virtual machine on the subnet has its own unique mac address. I have rebuilt the guest machine several times, and I have tried this on three different host machines (identical hardware, built identically). Oddly, I do have one virtual host (the second of this series) which never seemed to have a problem. It never had its network freeze when it was running the same rsync during its build. It's particularly odd because it was the second build. The first, on a different host, did have the freezing problem, but the second did not. I assumed at the time that I had done something wrong with the first build, and that the problem was resolved. Unfortunately, the problem reappeared when I built the third VM. Also unfortunately, I can't do many tests with the working VM, as it's now in production use, and I'm hoping I can find the cause of this issue before that machine starts having problems. It's possible that I just got really lucky while running the rsync on the working machine, and that one time it didn't freeze. Of course it's possible that I somehow changed the build scripts without realizing it and re-broke something, but I can't find any such thing. In any case, I'm hoping someone has some idea what could cause this. Addendum: Preliminary tests suggest that I don't have the problem if I substitute e1000 for virtio in the first -net flag to qemu-kvm. I don't consider this a solution, but it is suitable for a stopgap. Has anyone else had (or better yet, solved) this problem with the virtio network driver?

  Read the article

• Python-daemon doesn't kill its kids

  - by Brian M. Hunt

  When using python-daemon, I'm creating subprocesses likeso: import multiprocessing class Worker(multiprocessing.Process): def __init__(self, queue): self.queue = queue # we wait for things from this in Worker.run() ... q = multiprocessing.Queue() with daemon.DaemonContext(): for i in xrange(3): Worker(q) while True: # let the Workers do their thing q.put(_something_we_wait_for()) When I kill the parent daemonic process (i.e. not a Worker) with a Ctrl-C or SIGTERM, etc., the children don't die. How does one kill the kids? My first thought is to use atexit to kill all the workers, likeso: with daemon.DaemonContext(): workers = list() for i in xrange(3): workers.append(Worker(q)) @atexit.register def kill_the_children(): for w in workers: w.terminate() while True: # let the Workers do their thing q.put(_something_we_wait_for()) However, the children of daemons are tricky things to handle, and I'd be obliged for thoughts and input on how this ought to be done. Thank you.

  Read the article

< Previous Page | 380 381 382 383 384 385 386 387 388 389 390 391  | Next Page >