We have about 30-35 sites per server accross 8-10 servers at a time depending on load, for a total of ~300 sites.  The sites are more or less hosting slightly customized versions of the same .net web app.  One site per app pool.

We find that our web servers get CPU constrained before anything else.  We currently are using CPU throttling on a per app pool basis, set at 80%, which seems a little high to me for a server hosting 30-35 sites per.

Is there an empirical or best practice way to determine the CPU throttling "sweet spot" % for solutions ike our that get CPU constrained mostly, and are ok on all other resources?

Weird hey..

Had some severe AD issues the other day, 2 pairs of shorts later and everything is back and working... except!

1 member server has IIS7 and publishes a file.. and it wasn't working.. I found that giving this file full access to 'everyone' sorts it..

but obviously not ideal eh!.. I know that the file used to have IIS_IUSRS as a member of it's security permissions.. but not anymore!

I try to add IIS_IUSRS in the usual way.. (properties, security, edit, add, IIS, CheckNames......)

just returns with IIS_WPG..  doesn't present IIS_IUSRS as an option..  Object is definately in AD as I can see in it ADUC.. and yes my location search scope includes built-in and all object types!

Thing is.. if I try to add IIS_IUSRS to any file on any of the 3xDC's.. it presents both IIS_WPG and IIS_IUSRS just fine!!!!!!!... 

all PC's or member servers only present IIS_WPG..  (I checked.. my test DC doesn't have IIS installed either...  just to tick that box!..)

DC's comprise 1x2012 and 2x2008 servers...  rest of the member boxes are 2008/2008R2.. PC's win7..  all at latest patches.. oh and all servers are virtual on vmware 5.5.

What could be causing this?  (remember.. it presents fine, if I add the object to a file on any of the 3 DC's.. so my syntax/process is good!)

How may I resolve this?

Run out of ideas at this point.. hopeing someone's seen it before..

help appreciated

AJ

Hello,

I have an FTP site that every now and then doesn't let the client log on. In the log I can see that "sc-status" is 530 and "sc-win32-status" is 64. The client reports "Home directory inaccessible". This can go on for a couple of minutes, and then it starts working again. The home directory of the ftp user is a windows share located on a server in a different country over a WAN link, so I'm thinking that it can sometimes slow to reach the home directory. How can I find out if that is the case? Is there a timeout in IIS for accessing the home directory? Can it be increased? 

URGENT HELP NEEDED

Hello,

I've a new web server windows 2012. I'm running classic ASP pages on it. Most of the pages are extracting datas from a ms access 2007 database.

Everything works fine till the moment when I call one of the above pages and it doesn't return any data while the browser seems to be working. This moment usually seems to happen randomly. What is shown by the browser is a white screen. None of the pages of all web sites running on the server and getting data from any database on the server are working: the screen is white while iexplorer is working.

FTP, Remote desktop, any application in the server (while accessing with the remote desktop) are working fine.

HTML pages work fine.

I need to restart the server (restarting the web site is not solving anything ) to bring the things back to normality.

I've seen the log events but I'm not sure about what to search for.

It seems that the occurances, when asp pages are not producing any response, are slightly increasing.

Please I need help to understand and solve.

many thanks

Alessandro

I'm hoping someone might have some ideas here.

I maintain a C# WinForms application that uses the System.Reflection.Assembly.LoadFrom() method to load UI assemblies from a specified URI, namely a read-access virtual directory that works as a distribution point for UI front end assemblies.  One of our clients has begun to have trouble with this earlier this week, and the application GET verbs from that folder have been responding 403.1.  It was working previously, and nobody seems to know what has changed.

Looking at an error trace, it appears that the MAP_REQUEST_HANDLER is returning the error code.  My understanding of how this is supposed to work is that the StaticFile handler should be handling the request, which shouldn't need execute privledges at all to run and should simply retrieve the file using the x-msdownload mime type.  That doesn't appear to be happening, and I've exhausted myself of things I know to check.

Does anyone have any ideas of what could be going on?  The error trace logs I generated can be found here.

We are getting "Data Loss has occurred" error after Windows Server 2008 r2 upgrade toServer 2012 OR 2012 R2 (yes, tried both OS same issue). Windows is completely updated before the upgrade. The only major application on the IIS is Team Foundation Server 2010.

The actual error is is

One or more problems have been detected during IIS upgrade. As a precaution, one or more IIS services have been disabled. Please review the following error message:

Data loss has occurred - You upgraded your operaing system or swithced from one Windows SKU to another SKu, and data from the previous installation of IIS has been lost or copied incorrectly. As a result, the services W3SVC and MSFTPSVC have been disabled. Before enabling these services, you should review the latest IIS 7.0 compatibility information.

How to fix this issue?

SK

We access another server outside our lan and the firewalls are open I can hit the ip -https://#######  and it displays their IIS

also if I hit an .html on their server it shows that file. If I hit a .aspx I recieve a 404 and the IP of there server resolves to my local host.

Has anyone ever expierenced this kind of problem or found a solution

I am running Server 2012 IIS8

the other server is Windows Server 8 not sure of their IIS version

Thanks

HI, I'm using Windows 2008 R2 with IIS 7.5.

I'm using a default port 21 FTP server just fine.

Now I want to add another FTP server on port 2221.

I tried that and I can't list the folders or transfer files but I can log in.

I'm using the passvive mode of the WS_FTP pro client to connect.

It only fails when I use my public IP, if I use my local IP ex: 192.168.1 ...to connect it works.

The Windows Server 2008 R2 server firewall is on, I added the rules for port 2221 as well

as my router..still no luck.

I also tried to turn off the Windows firewall with no results.

I install FileZilla Server and was up and running in 2 minutes, but I would like everything to run from IIS.

This is the log file from WS_FTP you will see 501 Server cannot accept argument

I think this is a data channel problem?

220 Microsoft FTP Service

Host type (1): Microsoft NT

USER Amazon

331 Password required for Amazon.

PASS (hidden)

230 User logged in.

Host type (I): Microsoft NT

PWD

257 "/" is current directory.

TYPE A

200 Type set to A.

PASV

227 Entering Passive Mode (192,168,1,48,194,152).

connecting data channel to 192.168.1.48:194,152(49816)

Substituting connection address 66.66.66.666.. for private address 192.168.1.48 from PASV

Failed to connect data channel to 66.66.66.666:194,152(49816)

PORT 192,168,1,83,4,31

501 Server cannot accept argument.

Port failed 501 Server cannot accept argument.

QUIT

221 Goodbye.

Connection closed.  Ready for next connection.

Thanks.

I have a MVC 4 application running on IIS 7.5 in Integrated mode.

One of my requests under some circumstances takes too long. It is started let's say at 10:00:00 int the browser. Then on the client side fiddler can see it at 10:00:00. Then it is routed to MCV controller, but in controller's log I can see it 10 seconds later! Than it takes some time to prepare model and render a view but it's not the issue. In the Fiddler's log on browser side i can see that the whole request has taken 15 seconds. In IIS log i can see also 15 seconds. But controller action and view rendering takes only 5 seconds.  Where is my request for the 10 seconds? Server is not busy, this is the only request at the moment. Processor has nothing to do and there is a lot of free memory. I've tried also to: disable pool recycling, disable http compression, disable tpc chimney - but no effect.

I suspect that some module on the way is waiting for some timeout.... I've tried to switch off http modules that I don't need but no effect.

Is there a way to trace how many time it has taken for each http module to process the request until it reaches controller?

What eles could I check?

Piotr

I have a strange scenario where I am setting up a web application and can't assign as user as the primary account for permissions.  The vitual name appears to be set to use a different user than the default for new applications created.  So if my vitual is called x and I make other vituals x1 or x2 with the exact application folders and code x1 and x2 default to the correct user for permissions.  x despite deleting it and starting over will not default to the correct user.  I am not sure where to verify this other than removing the users assigned and watch if it fails.  Also don't know how or why it is unique for x to be assigned to another user be default even when deleted?

Hello,

today I've tried to upgrade one of my Windows 2008 Web Server R2 hosting www servers to Windows Server 2012 R2. As always system upgrade went fine. Also everytime I have to make some changes to applicationhost.config file from IIS7.5 to make it usable in IIS 8.5 (after upgrade IIS is starting with clean applicationhost.config file). I've done that before and every time it worked until now. Issue was with starting service "Windows Process Activation Service". It just didn't wanted to start. I've started checking applicationhost.config files for some errors. If this file has syntax errors then IIS Manager when checking application pools or websites is showing exactly where this issue is (one interesting thing though IIS service has some problems with national characters used in applicationHost.config file - but this is something to check because this can be caused by file character encryption). After fixing some issues with national characters and removing application warmup old entries I was able to see in IIS Manager all of my previous application pools and websites all stopped since IIS Service is not starting because of "Windows Process Activation Service" not startig issue.

So I've started with removing ALL entries from <sites> in applicationHost.config file. When I removed them WAS started correctly so I started to add websites which I've previously removed. I've come to interesting thing. In some point WAS was again stopped to startup correctly. This was very weird because <site> which I've added didn't had anything interesting. Just another website and here we have totally weird situation. I've started changing latest <site> entries in <sites> list and I ended with conclusion that this is not issue with <site> configuration but with just some kind of limit which I could not understand. I've checked my applicationHost.config file where are all of my websites present and he has 77289 lines. Does someone experienced similar issue?

I had to go back to snapshot created befure upgrading to 2012 but I have some error which WAS was returning "%%2148734210". When entering this error in google I'm getting this page:

http://www.experts-exchange.com/OS/Microsoft_Operating_Systems/Server/Windows_Server_2008/Q_27843003.html

but I don't see there any post with this error number or whatever that is.

Regards

EDIT: I've tried to use my production system applicationHost.config file on Windows 2012 R2 Standard test box and I was able to reproduce this issue. When I was trying to start WAS with all of websites in <sites> I'm getting error "Windows could not start the Windows Process Activation Service on Local Computer. For more information, review the System Event Log. If this is a non-Microsoft service, contact the service vendor, and refer to service-specific error code -2147467259". When I removed all of <site> entries in <sites> node from some point which I remember that WAS was starting on my production server to the end of <sites> entries (just before <siteDefaults>) I was able to run WAS service.

Errors from event log:

"The Windows Process Activation Service encountered an error trying to read configuration data with AppHostProvider. The error message is: Bad Data. The data field contains the error number."

"The configuration manager for Windows Process Activation Service (WAS) did not initialize. The data field contains the error number."

"Windows Process Activation Service (WAS) is stopping because it encountered an error. The data field contains the error number."

"The Windows Process Activation Service service terminated with the following service-specific error:  Unspecified error"

applicationHost.config file with removed part of <site> entries which allows WAS to start has now 56503 lines.

EDIT2: Another example that something must be wrong with number of informations stored in applicationHost.config file. For working scenario I've added next <site> entry which started to make WAS service not to start correctly (the same error). After that I've removed first <site> entry which is ID=1 Default Web Site (created for default website configuration with IIS). After removing this first website WAS can start correctly again. <site> node for Default Web site does not contain any specific entries. Interesting part is that WAS allows EXACTLY 100 <site> entries in <sites>. This is veeeery weird and I have no explanation for that. Is there any IIS guru here which can explain this situation? Normally this applicationHost.config file contains entries for almost 2k websites runned currently under IIS 7.5.

Hello

I have a website that's very often does AppDomain restarts, this causes all sessions to be lost. As far as i've manged to figure out, this is the primary suspects in case of appdomain restarts:

1. Modification in Machine.Config, Web.Config or Global.asax files
2. Application's bin directory or its contents modified 
3. Antivirus is running on the  .config /.aspx files
4. The number of re-compilations (aspx, ascx or asax) exceeds the limit specified by the <compilation numRecompilesBeforeAppRestart=/> setting in machine.config or web.config  (by default this is set to 15) 
5. The physical path of the virtual directory is modified 
6. The CAS policy is modified
7. Sub-Directories of Application are deleted or renamed

I've manged to rule out all of them, and is therefor in need for help to figure out what's causing the appdomain restarts.

A small breakdown of the points: 
1) I've searched the whole computer for files that got modified or changed before the appdomain restarts, machine,web or global.asax is not one of them.
2) Same as abow, no changes
3) Tried to disable antivirus all together
4) I thought this was the problem, but since we have to web-front-ends, I adjusted numRecompilesBeforeAppRestart on one server to a very low value, and the other server to a very high value, the problem seems to occur just as often on both of them.
5) I can't imagine that his happens
6) I dumped the CAS policy before and after an appdomain restart, and can't see any changes
7) This does not happens

I turned on  "Application lifetime events" in the web.config under .net framework in hope of that telling me something sensible. But no, it's only telling me "Application is shutting down. Reason: Configuration changed". Is there anything obvious I've left out? I can't see any changes to any config files, neither on application,site or server level.

I do suspect the application of doing this, but I can't prove it. 

Any tips would be highly appreciated:) 

Hi All,

I am facing 2 times a week that the web application down unexpectedly. I have to re-deploy the web application again again in order to let the web application run. Need help from you on IIS issue. 

Here is the log:

Log Name:      Application

Source:        ASP.NET 4.0.30319.0

Date:          4/1/2014 10:07:27 AM

Event ID:      1301

Task Category: Web Event

Level:         Error

Keywords:      Classic

User:          N/A

Computer:      WEB-APPS

Description:

The following exception was thrown by the web event provider 'EventLogProvider' in the application '/WEB_Portal2' (in an application lifetime a maximum of one exception will be logged per provider instance):

System.Web.HttpException (0x80004005): The EventLogWebEventProvider provider failed to log an event with the error code 0x80070057.

   at System.Web.Management.EventLogWebEventProvider.ProcessEvent(WebBaseEvent eventRaised)

   at System.Web.Management.WebBaseEvent.RaiseInternal(WebBaseEvent eventRaised, ArrayList firingRuleInfos, Int32 index0, Int32 index1)

Event Xml:

<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">

  <System>

    <Provider Name="ASP.NET 4.0.30319.0" />

    <EventID Qualifiers="49152">1301</EventID>

    <Level>2</Level>

    <Task>3</Task>

    <Keywords>0x80000000000000</Keywords>

    <TimeCreated SystemTime="2014-04-01T02:07:27.000Z" />

    <EventRecordID>6843</EventRecordID>

    <Channel>Application</Channel>

    <Computer>WEB-APPS</Computer>

    <Security />

  </System>

  <EventData>

    <Data>EventLogProvider</Data>

    <Data>/WEB_Portal2</Data>

    <Data>System.Web.HttpException (0x80004005): The EventLogWebEventProvider provider failed to log an event with the error code 0x80070057.

   at System.Web.Management.EventLogWebEventProvider.ProcessEvent(WebBaseEvent eventRaised)

   at System.Web.Management.WebBaseEvent.RaiseInternal(WebBaseEvent eventRaised, ArrayList firingRuleInfos, Int32 index0, Int32 index1)</Data>

  </EventData>

</Event>

Thanks.

Olá estou configurando um servidor windows 2012 na Amazon e estou encontrando dificuldade com o PHP com o que me parecer ser simples de resolver, porém já vou pra mais de 48 horas de pesquisas e várias tentativas e nada de resolver.

Acredito se simples mas não consigo achar a resposta e preciso com urgência.
Por gentileza dê uma olhada neste link deste site q está hospedado no Amazon: http://ec2-54-207-28-21.sa-east-1.compute.amazonaws.com/clubedapelada/como-funciona.php
Vc pode ver que as imagens estão quase todas quebradas, usei o relative path no código php dessa forma "/imagem/nomeimagem.jpg", qndo mudo o path para "./imagem/nomeimagem.jpg" a imagem é exibida.
O problema é que o site está hospedado em outro alojamento em servidor windows e funciona perfeitamente com o caminho  "/imagem/nomeimagem.jpg", confira no link http://clubedapelada.com.br/como-funciona.php.

Ficaria inviável pra mim alterar todos os caminhos de imagens em todo o site, inclusive na parte administrativa que faz uploads e gera o caminho com a barra normal e funciona perfeitamente no outro servidor.
Por favor alguém pode me dar uma luz ?????
Aguardo respostas com urgência. Grato pela atenção !

We have many ecom customers who use Sagepay for their credit cards.

We moved several sites over to a new windows 2008 web server which was fine then recently we started seeing issues.

The sagepay system works where it posts a response to a given url behind the scenes with the outcome of a credit card transaction, much like the paypal IPN model.

It comes in the form of a form post.

You simply then query the form elements and based on the contents of the fileds sent update your db accordingly.

But more recently this is coming up as an completely empty forms collection.

I have put code in to spit out the entire forms collection to a text file each time so I can see whats being sent, heres an example of the kind of thing I expect to see from past orders that were fine:

VPSProtocol=2.23
TxType=DEFERRED
VendorTxCode=10271
VPSTxId={639E703C-6AB3-3491-65DD-3F1D479DCAA2}
Status=ABORT
StatusDetail=2008 : The Transaction timed-out.
AVSCV2=DATA NOT CHECKED
AddressResult=NOTPROVIDED
PostCodeResult=NOTPROVIDED
CV2Result=NOTPROVIDED
GiftAid=0
3DSecureStatus INCOMPLETE
CardType MC
Last4Digits 999
VPSSignature=342EFE18FCC5F47AD6E845EB6013272C

But now all I get is a completely empty response, well asp.net seems to think so. I KNOW its not sage pay and I have the EXACT same code on our other windows servers and all works so it has to be be an IIS/ASP.NET issue but I have no idea on where or how to start to debug or look into it.

There are no errors in the event log so really stumped

Anyone got any ideas where to start?

I have an app that used the Socket class to do FTP.  When I run the application on Server 2008 r2 (.net 3.5), it works perfectly.  When I run it on Server 2003 (.net 3.5) the Socket.Receive blows up and I get a " An existing connection was forcibly closed by the remote host" error.

The FTP server is running Server 2008 r2.  If I connect to an FTP server running Server 2003 from a server running Server 2003, all is fine or if I connect Server 2008 to Server 2008 all is fine.

This is the line of code that throws the error:

Bob Bevers

I have a W2K R2 SP1 server that refuses to install web deploy 3.5.

I've attempted the suggested actions in this post, but with no change:

http://forums.iis.net/t/1207327.aspx?Unable+to+Install+Web+Deploy+3+5+on+my+windows+Getting+Error+

I've run the extracted MSI with logging and received this error:

MSI (s) (4C:88) [05:49:32:582]: Product: Microsoft Web Deploy 3.5 -- Error 1720. There is a problem with this Windows Installer package. A script required for this install to complete could not be run. Contact your support personnel or package vendor. Custom action InstallEtwManifest script error -2147217406, SWbemServicesEx: Not found  Line 9, Column 13, 

I've seen posts regarding checking the DLL path for vbscript and making sure it's pointed to the correct version (32bit VS 64bit) and the Mcafee script blocking.  I've updated the registry entry "HKEY_LOCAL_MACHINE\SOFTWARE\CLASSES\Wow6432Node\CLSID\{B54F3741-5B07-11cf-A4B0-00AA004A55E8}\InprocServer32"  to "C:\Windows\SysWOW64\vbscript.dll"  No change.

I've also reinstalled IIS numerous times without addressing the issue, using this link as reference:

http://blogs.msdn.com/b/chaun/archive/2010/05/26/iis7-x-a-common-mistake-when-reinstalling-iis-7-x-errors-0x80070490-and-0x80070002.aspx?Redirected=true

I've also run sfc /scannow.  No issues were found.  Server is still running IE 9 (corporate standard) and has all recent Windows OS patches.

Any other suggestions?

We've got a somewhat older ASP.NET Web Forms app, we wrote it back in the early 2000's. It's now at .NET 4, and we edit it/maintain it with VS 2010. It's hosted in an off site VM that's running Windows Server 2008 R2 Standard with Service Pack 1, so it's using IIS 7. It uses SSL. Back when we wrote it we only targetted Internet Explorer. Perhaps that wasn't a wise idea, but that's what we did. It has been working fine, up until IE11 came out. Anyone running IE11 cannot connect to it. Instead they get a "This page can't be displayed" error message. However, if I bring up IE10, I can easily connect to it. Or any older version of IE, I can easily connect to it, no problem at all. I've run Fiddler 2 to see what the HTTP Response is, and it's 302, which makes sense because it's set up to first send the user to the login page, then return them to whatever page they were trying to go to.

The weird thing is that I cannot duplicate this behavior on our test server. It too is running in a VM. It's also Windows Server 2008 R2 Standard with SP1. I've got both web sites configured identically on each machine. But I can easily access the website on the test server with IE11 or lower; no problem at all. The only thing I can think of is the test server is running on a VM that's on-premise, whereas the production server is running in a VM off-site. The production server is hosted by our parent company.

My feeling is that the issue is either one with IIS, or some difference I've not discovered between the two servers. How do I find out what's going on and then fix it?

Hi All,

Im getting that dreaded message: "server application unavailable".  I have recently migrated a website from .net 1.1 to .net 2.  The website is deployed on two load balanced webservers.  On the 2nd web server, the website works fine with no problems, however on the 1st web server Im getting that error when browsing to the site.

Weird thing is if I switch the website on the 1st web server to run under a .net 1.1 app pool then I can browse to the website,(The issue is that my team has added some new functionality to the site which uses the .net2 framework, so we converted the site to .net 2 in visual studio so it needs to run in .net classic mode).

If I put it in a .net2 app pool I get the error again.

I've even tried giving the .net app pool full control on the website directory, but I still get the same error. Whats strange is that I have the same setup/configuration on the 2nd webserver and it works just fine.

I've tried:

recycling the app pool, reinstalling .net (aspnet_regiis -i)

any idea/suggestions?

Thanks in advance