|
First, thanks for the response! Here's what I've found out:
1. I can ping the router, but can't access it with IE like I can on my XP laptop or the 2K box.
2. I can ping the DSL modem from the XP home box.
(didn't think of pinging either of these devices before...thanks)
3. I can access my laptop from the XP home box, even through explorer.
The TCP/IP settings are exactly the same on both XP boxes, so I'm thinking maybe it's just IE that's messed up on his XP home box.
Thanks again, now that I know the card is not messed up, I hope I can figure out what's going on with IE.
Chris Richardson
You can stash and you can seize
In dreams begin, responsibilities U2 - Acrobat[^]
Stop being PC and accounting for everyone and his momma's timeframe. Just enjoy your - Rohit Sinha in the content-challenged thread
|
|
|
|
|
If you can reach the network via ping and perhaps tracert, then the network card and the network doesn't seem to be the problem. Once you have determined you can ping another host on your own network try performing a trace route using tracert to an external host such as codeproject. Keep in mind that depending on the routers that are between your network and the destination network you may get some timeouts. Try several external hosts that you know are not using the same route. Once you can determine that your are at least reaching your ISP's router you know any other network problems are outside your control.
As for application specific issues such as IE, check the proxy settings for the application. IE may have been set to connect to a proxy server that it does not have access to on your network.
You might also try using another application that will utilize an alternate protocol. Try using FTP to go to a public FTP site.
|
|
|
|
|
If the gateway IP is set to the local port of the router, and the XP box is on the same subnet, IE should have no trouble connecting. In Tools/Internet Options/Connections, there should be no connection information. Click on LAN Settings, and make sure that no boxes are checked - the Automatic and Proxy options will interfere with the box working properly, in particular.
I haven't used the Home version - it's so badly crippled it isn't worth considering - but I'd also suggest that you review permissions for the user account. There may be a new one that restricts access to the Internet. Also, many OEMs preconfigure Windows with what they think most users will want. If yours has ICS installed and it's not acting as the gateway for your network, get rid of it, along with that stupid excuse for a firewall that comes with it.
"Please don't put cigarette butts in the urinal. It makes them soggy and hard to light" - Sign in a Bullhead City, AZ Restroom
|
|
|
|
|
How can I connect a client application running behind a proxy to a web server?
The connection cannot be a direct TCP to the server as the proxy is in between. I can have a Http connection , the way a browser works. But how do I achieve it. Any suggestions??
|
|
|
|
|
Your questions is a bit general to give you a specific answer. Let me lay out some examples.
Let's say you have an client application that will be using the HTTP protocol to contact a server. If you want the client app to make all transactions via a proxy you would need to configure the client app to connect to the proxy itself. If you are using the HTTP protocol, then you would tell the proxy what destination server you want your request forwarded to and the proxy would then interpret that request and perform it, and return the results to you.
How you make the connection via a proxy really depends on the type of protocol. Take the SMTP protocol for example. This protocol was not originally designed to be a proxied protocol. However, that doesn't mean you can't proxy it. In this case you would configure your mail client to connect to the proxy when sending mail. The proxy would then be configure to connect to the real server. In this scenario, the client would not know that all of it's requests are being forwarded to the real server.
Hope that helps.
|
|
|
|
|
What I have read about Http CONNECT is that it is used to establish a secure connection to the server.Does that mean that I can connect only to port 443 of the Listner i.e. server?
Can I connect to the port 80 of the server i.e specify the ip address:80 in the CONNECT method
|
|
|
|
|
There is nothing special about the port. It is just a commonly chosen port of 443 for HTTPS and 80 for HTTP. If you want to make a non-secure connection then you make your connection and don't negotiate an SSL session. Of course the server you connect to must be expecting a non secure connection, which most do.
Let's say you want to connect to codeproject from your browser. If you were to connect directly from your application, it would open a TCP connection via port 80 and send something similar to the following.
GET / HTTP/1.1
Accept: */*
Accept-Language: en-us
User-Agent: Your application name here
Host: www.codeproject.com
What will be returned will be the contents of the page requested. You can test this using telnet. Open a telnet session to your own server on port 80 and type the above text. End the text with two carriage returns to get the results.
Now if you wanted to do the same request via a proxy you would do something like this.
GET http://www.codeproject.com/ HTTP/1.1
Accept: */*
Accept-Language: en-us
User-Agent: Your application name here
Host: www.codeproject.com
There is very little difference. In this case you supply the host and protocol portion of the URL instead of just the path. This is because the proxy needs to know the name of the host you want to connect to. Instead of connecting to port 80 on codeproject you would connect to port 80 on your proxy server. Of course your proxy could use any port you wish. Port 80 is the standard, but it could easily be 8080 or some other port of you chosing.
If your proxy understands other protocols such as https, ftp, gopher, or a custom protocol then you could use those by simply modifying the URL passed.
|
|
|
|
|
How can I know when to use a 3 state model or a 5 state model or a 7 state model for processes as they are created, runned and terminated.
This is for a project.
PitoPR
|
|
|
|
|
This is odd. Windows Explorer shows that my C: drive has 5490MB in use, but if I total the used space for all of the folders on the drive it comes to only 3716MB - size on disk, not file size. Which value is 'real'? And why is there such a huge discrepancy?
"Please don't put cigarette butts in the urinal. It makes them soggy and hard to light" - Sign in a Bullhead City, AZ Restroom
|
|
|
|
|
It may be because of cluster size. On a large drive, the cluster size is about 32KB. So space is used up in blocks of 32KB. This means that the minimum amount of space a file can take up is 32KB, and if it was, say, 35KB, it would have to take up 64KB of actual disk space.
"Do unto others as you would have them do unto you." - Jesus
"An eye for an eye only makes the whole world blind." - Mahatma Gandhi
|
|
|
|
|
That's why I used the 'size on disk' figure instead of the file size itself. The former takes into account the cluster waste. I've never needed to defrag a Win2K disk before; perhaps it's time...
"Please don't put cigarette butts in the urinal. It makes them soggy and hard to light" - Sign in a Bullhead City, AZ Restroom
|
|
|
|
|
If it is FAT 32 and is still in the same space problems... then defrag may help it.
If you use that as a standalone PC with less of file permission options... you can try to convert it to a NTFS file system. This gets better.
The above is something I observed with my PC..
I was born intelligent Education ruined me!.
|
|
|
|
|
Nah... this one was born with NTFS. I never use anything else if NTFS is available. I know that NT4 had a hell of a time with fragmentation, and it didn't ship with a defrag routine. I thought that Win2K solved the problem, but I guess there's a good reason that they included a defrag utility with NTFS5.
"Please don't put cigarette butts in the urinal. It makes them soggy and hard to light" - Sign in a Bullhead City, AZ Restroom
|
|
|
|
|
Everything is not stored in files. NTFS keeps
alot of data directly in MFT(Master File Table)
and that is no file. My disk has some 100MB in MFT
skål
jhaga
|
|
|
|
|
That would account for some, but >1500MB? It just doesn't sound right somehow... The idea of defragging the drive makes me nervous, though. There are so many processes running from the C: drive that I'm not sure how defrag will behave. I've had problems with it on lesser systems (Win9x) and haven't heard anything, pro or con, about the robustness of the Win2K Server version.
"Please don't put cigarette butts in the urinal. It makes them soggy and hard to light" - Sign in a Bullhead City, AZ Restroom
|
|
|
|
|
Before you can defrag you must run some kind of
chkdsk /f on the disk. It is possible that you have
1500MB in lost clusters but I don't find it very likely.
A defrag program that uses the underlying operation system
should be totaly safe to use. Of course if you have
a power failure then you will probably loose one file.
jhaga
|
|
|
|
|
Chkdsk finds no errors at all on the drive, so I guess I'll go ahead with the defrag...
"Please don't put cigarette butts in the urinal. It makes them soggy and hard to light" - Sign in a Bullhead City, AZ Restroom
|
|
|
|
|
Here you can find a nice tool for exploring your disks fragmentation
http://www.sysinternals.com/ntw2k/info/defrag.shtml
with source code.
|
|
|
|
|
How can I get the process path by using API "NtQuerySystemInformation".
|
|
|
|
|
Why?
I just installed Sygate Pro 5.0 and it immediately announced that ntoskrnl.exe is trying to connect to the Internet. It's only one of several processes making the attempt, but I can't imagine why the kernel needs access every minute or two to somewhere on the 'net. Does anyone know what it's doing?
"Please don't put cigarette butts in the urinal. It makes them soggy and hard to light" - Sign in a Bullhead City, AZ Restroom
|
|
|
|
|
It may just be a normal network broadcast, the pc is 'announcing itself' to the network, maybe...
Does Sygate report the local port and the remote ip and port ?
Clones are people two.
|
|
|
|
|
It seems to be just the normal NetBIOS announcements. I expected that to be in a separate service, but ntoskrnl is the source reported. It would be so nice to have a reference to the functions that all these various processes in Windows perform, and why they're needed...
"Please don't put cigarette butts in the urinal. It makes them soggy and hard to light" - Sign in a Bullhead City, AZ Restroom
|
|
|
|
|
I regularly use perfmon.exe as one of my debugging tools. Out of nowhere, 2 major counter object categories, Memory and Processor, disappeared from the selection drop-down box. The registry troubleshooting info at MSDN is kind of off, as it relates to Win 2K, not XP. Anybody know how to re-init perfmon or make it sane again?
|
|
|
|
|
Does anyone have information about the user rights required to install browser plugins?
I have some "restricted users" on my Win2K machine that can't install the Flash plugin. It was already installed by an admin user, but the restricted user still got the error message saying that it had not been installed.
Is this expected? Where are plugins saved on disk?
Thanks,
Richard
|
|
|
|
|
Hello,
Application
-----------
I've created a multi-tier application. Its architecture is:
On client machine --> the client application is an ActiveX EXE component (VBasic).
On server machine --> some COM+ server applications and a SQLServer database.
Use Case
--------
CLIENT1
--> asks SQLServer database for an object (= a row in objects table).
The state of this object is stored in a storage file.
--> if the object can be check out, it means nobody else have checked out the object,
the server will set the objects.checkedoutby = ClientID and will copy the stg file
in a shared folder on server machine.
CLIENT2
--> tries to check out the same object
--> the server must know (??? - MY QUESTION IS ABOUT THIS POINT) if the CLIENT1 application
is still running or a crash has occured and CLIENT1 was not able to check in the object.
Remarks
-------
1) CLIENT2 will receive a copy of the storage file if CLIENT1 is still running
2) CLIENT2 will have the possibility to restore the storage file from
an older version if a crash occured in the CLIENT1 application.
3) Many clients can log in using the same account (user name and password)
4) There can be more than one instances of the application on the same client machine
My solution
-----------
My solution is to create an instance of an "watchdog" ActiveX EXE on the client machine
when the client application is started. This "watchdog", from time to time (5 minutes by example),
will modify the value of a date field named NotificationDate in database.
When the client application shuts down, the "watchdog" will stop updating that field.
The stored procedure responsible for check out will use the value of NotificationDate field
to decide whether the CLIENT1 application that use this object is still running or a crash has occured.
Questions
---------
1) Is there a pattern for this problem?
2) What about my solution?
10x,
Ovidiu
|
|
|
|