Strange WideFS problem

[steve_j]

Hi!

I searched the forum, but found no solution for my problem.

I am trying to run a WideFS on a client machine but with no luck:

Wideclient.exe starts up correctly, connects but then doesn't respond. CPU load for Wideclient is also > 80%, so it doesn't respond to mouse clicks anymore. It can only be closed via task manager. Both Wideclient and Wideserver logs show correct connections. When I start e.g. PM Demo it can't find a connection to FSUIPC though Wideclient shows "connected". I've also tried wideclient on a notebook in the network, there everything went fine!

So, has anyone a hint, what is wrong?

OS: WinXP

NC: 3com 3C900B-TPO

Network connected via D-LINK DI-604 Router

thanks,

Steve

[vdkeybus]

Three things come to my mind:

1. Check your network connection by sending a large (> 20 MB or so) file between the computers in both directions using the protocol (TCP or IPX) you've selected for WideFS. Don't tolerate any error.

2. Check the settings of the Win XP firewall.

3. Check the log (WideClient.log) which is in the same directory you start WideClient in.

Regards,

J.

[steve_j]

Hi!

Everything is normal, transmission rate is about 1100kb/sec in both directions. No win XP firewall activated.

regards,

Stephan

[vdkeybus]

transmission rate is about 1100kb/sec in both directions

That looks ok for a 10MBit card. Verify in your network connection property windows (Configuration panel -> Netw. & Internet -> Network, dbl. click the network link) that indeed you're not getting any packet errors.

I exclude your DI-604 because using another PC solves the problem. Neverthelees, there's also a cable tester and diagnostics in the unit which may be useful to trace the problem. I think you have to log in using 'http://192.168.0.1' to use these.

And, finally, try another network card. They're so cheap nowadays that searching for the current problem may be more expensive (time-wise), especially if you could borrow one from a friend.

Regards,

J.

[steve_j]

Hi!

I changed the network card and have the same problems!! Very very strange...

regards,

Stephan

[Pete Dowson]

I changed the network card and have the same problems!! Very very strange...

Where's the Log?

Pete

[steve_j]

Hi!

********* WideServer.DLL Log [version 6.47] *********

Blocksize guide = 4096 (double allowed)

Date (dmy): 02/04/05, Time 19:23:04.250: Server name is FSPC

91578 Initialising TCP/IP server

91594 Initialising IPX/SPX server

91610 ServerNode=0.0.4352.48649.59246

103094 Restarting service due to total lack of use

112766 Incoming connection Accepted ok (skt=1748)

112828 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=1748)

********* WideClient Log [version 6.47] Class=FS98MAIN *********

Date (dmy): 02/04/05, Time 19:25:08.373: Client name is JO-ABVOFK840JCC

150 Attempting to connect now

220 Connection made okay!

I had to kill the wideclient process again via task manager. Closing via the close window button had no effect. I tried now both IPX/SPX and TCP/IP with no luck.

regards,

Stephan

[Pete Dowson]

I had to kill the wideclient process again via task manager. Closing via the close window button had no effect. I tried now both IPX/SPX and TCP/IP with no luck.

There's nothing wrong with either log, and WideClient has nothing to do until you run a program on it, so if the process is hanging it just has to be a bad network driver or card, or some other program is interfering. All WideClient will be doing is idling waiting for a program to be run. It will send an "I'm here" message every couple of seconds to the Server and expect one back. That's it. It's really dead simple basic Networking copied direct out of Microsoft's own examples.

Regards,

Pete

[steve_j]

Here's again a log file, maybe the transmission rates at the can tell more about my problem.

[init skipped]

49593 Incoming connection Accepted ok (skt=592)

49609 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=592)

221422 Incoming connection Accepted ok (skt=656)

221484 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=656)

239593 Incoming connection Accepted ok (skt=3148)

239703 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3148)

257812 Incoming connection Accepted ok (skt=1768)

257875 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=1768)

276000 Incoming connection Accepted ok (skt=3328)

276031 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3328)

294125 Incoming connection Accepted ok (skt=3260)

294156 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3260)

312281 Incoming connection Accepted ok (skt=3240)

312343 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3240)

330422 Incoming connection Accepted ok (skt=3280)

330468 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3280)

348547 Incoming connection Accepted ok (skt=3368)

348593 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3368)

366672 Incoming connection Accepted ok (skt=3396)

366718 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3396)

541359 Incoming connection Accepted ok (skt=3404)

541406 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3404)

588109 Incoming connection Accepted ok (skt=1496)

588156 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=1496)

606468 Incoming connection Accepted ok (skt=3380)

606578 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3380)

624703 Incoming connection Accepted ok (skt=3492)

624734 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3492)

627015 Error 10053: client socket disconnected at Client: removing (skt=592)

627015 Error 10053: client socket disconnected at Client: removing (skt=656)

627015 Error 10053: client socket disconnected at Client: removing (skt=3148)

627015 Error 10053: client socket disconnected at Client: removing (skt=1768)

627015 Error 10053: client socket disconnected at Client: removing (skt=3328)

627015 Error 10053: client socket disconnected at Client: removing (skt=3260)

627015 Error 10053: client socket disconnected at Client: removing (skt=3240)

627015 Error 10053: client socket disconnected at Client: removing (skt=3280)

627015 Error 10053: client socket disconnected at Client: removing (skt=3368)

627015 Error 10054: client socket disconnected at Client: removing (skt=3396)

628031 Closing down now ...

Memory managed: Offset records: 153 alloc, 153 free

Throughput maximum achieved: 1 frames/sec, 825 bytes/sec

Throughput average achieved for complete session: 0 frames/sec, 4 bytes/sec

Average receive rate from "JO-ABVOFK840JCC": 72 frames/sec, 2625 bytes/sec

********* Log file closed *********

[Pete Dowson]

Here's again a log file, maybe the transmission rates at the can tell more about my problem

How is it that the previous logs you showed didn't have any of this?

This extract is from WideServer.log. There MUST be a comparable log from the Client which shows the reconnections, and more importantly probably the reason for them. This is only half of the story.

Interestingly the connection was problem free for the first 3 minutes (almost):

49609 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=592)

221422 Incoming connection Accepted ok (skt=656)

221484 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=656)

After this the Client re-connected every 18 seconds, which is roughly the timeout when it sees nothing arriving from the Server. This continued for 2.5 minutes approx, after which there was a gap for 3 minutes:

366718 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3396)

541359 Incoming connection Accepted ok (skt=3404)

541406 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3404)

Do these odd timings mean anything to you? Were you starting/stopping programs on the client, or doing something in FS which may relate?

In the performance summary the only worrying thing is this:

Average receive rate from "JO-ABVOFK840JCC": 72 frames/sec

What are you running on the Client which is attempting 72 frames per second?

The transmission performance simply shows why the Client is reconnecting -- there appears to be few opportunities for the Server to send things to the Client which is extremely busy sending 72 frames per second to the server!

Check the configuration of the Network at the client. It needs to be full duplex. Half duplex will cause this sort of blockage. Go through all of the settings. Something is wrong.

I cannot tell you anything else because the important half, the Client Log, is not supplied. You really need to think both ends, not one or the other, please.

I also cannot reconcile what you are reporting now with what you reported previously. It does not seem compatible.

If you are running an application which simply sits in a tight loop making constant demands on WideClient, you could try slowing it down somewhat by setting the ApplicationDelay parameter (in WideClient.INI) to something like 30 or 50. Please see the documentation.

Regards,

Pete

[steve_j]

Here's the log file of the client.

********* WideClient Log [version 6.47] Class=FS98MAIN *********

Date (dmy): 03/04/05, Time 09:44:46.143: Client name is JO-ABVOFK840JCC

100 Attempting to connect now

120 Connection made okay!

194750 Error on client post-Connection Select() [Error=10053] Software caused connection abort

194750 Connection closed by server!

194750 The connection was terminated due to a time-out or other failure!

194770 Ready to try connection again

194790 Attempting to connect now

194800 Error on client pre-Connection Select() [Error=10065] No route to host

194800 Ready to try connection again

194880 Attempting to connect now

225804 Connection made okay!

260955 Error on client post-Connection Select() [Error=10053] Software caused connection abort

260955 Connection closed by server!

260955 The connection was terminated due to a time-out or other failure!

260955 Ready to try connection again

261005 Attempting to connect now

261005 Connection made okay!

1116705 Error on client post-Connection Select() [Error=10053] Software caused connection abort

1116705 Connection closed by server!

1116705 The connection was terminated due to a time-out or other failure!

1116705 Ready to try connection again

1116715 Attempting to connect now

1117687 Error on client pre-Connection Select() [Error=10061] Connection refused

1117687 Ready to try connection again

1117727 Attempting to connect now

1141922 ****** End of session performance summary ******

1141922 Total time connected = 1085 seconds

1141922 Reception maximum: 0 frames/sec, 78 bytes/sec

1141922 Reception average whilst connected: 0 frames/sec, 15 bytes/sec

1141922 Transmission maximum: 0 frames/sec, 91 bytes/sec

1141922 Transmission average whilst connected: 0 frames/sec, 20 bytes/sec

1141922 Max receive buffer = 157, Max send depth = 1, Send frames lost = 0

1141922 ********* Log file closed (Buffers: MaxUsed 2, Alloc 1031 Freed 1031 Refused 0) *********

[Pete Dowson]

Here's the log file of the client.

but this doesn't appear to be the Client log matching the Server log above. The server said:

Average receive rate from "JO-ABVOFK840JCC": 72 frames/sec, 2625 bytes/sec

whilst the Client says:

1141922 Transmission maximum: 0 frames/sec, 91 bytes/sec

1141922 Transmission average whilst connected: 0 frames/sec, 20 bytes/sec

As it is, the logs merely show that the Network isn't working correctly, but I'm afraid they don't tell me why.

Pete

[steve_j]

Hi!

Ok, I will make a new try with fresh logs, maybe I have made a mess.

BTW:Thanks for your great support!!

regards,

Stephan

[steve_j]

Hi!

So, here is the next test session. Everything was like ealier. I tried PM ABGC to see if the program can connect via WideFS, but it hung up. I also tried my app I am working on, it also couldn't open a connection.

To sum up what I've done already:

1. Changed network card

2. Tried WideFs with another switch instead of the router

3. Fullduplex enabled on both NCs

4. Transfer of files between server and client is fast

Here the logs:

********* WideServer.DLL Log [version 6.47] *********

Blocksize guide = 4096 (double allowed)

Date (dmy): 05/04/05, Time 16:45:17.140: Server name is FSPC

70609 Initialising TCP/IP server

70625 Initialising IPX/SPX server

70640 ServerNode=0.0.4352.58121.22023

81828 Restarting service due to total lack of use

1705812 Incoming connection Accepted ok (skt=3304)

1705890 Connected to computer "JO-ABVOFK840JCC" running WideClient version 6.470 (skt=3304)

1852406 Closing down now ...

Memory managed: Offset records: 96 alloc, 96 free

Throughput maximum achieved: 12 frames/sec, 432 bytes/sec

Throughput average achieved for complete session: 0 frames/sec, 3 bytes/sec

Average receive rate from "JO-ABVOFK840JCC": 0 frames/sec, 30 bytes/sec

********* Log file closed *********

----------------------------------------------

*********************************

----------------------------------------------

********* WideClient Log [version 6.47] Class=FS98MAIN *********

Date (dmy): 05/04/05, Time 17:13:40.595: Client name is JO-ABVOFK840JCC

160 Attempting to connect now

260 Connection made okay!

34259 New Client Application: "ABGCDemo" (Id=1472)

85142 New Client Application: "AirbusGC" (Id=1524)

147021 Error on client post-Connection Select() [Error=10053] Software caused connection abort

147021 Connection closed by server!

147021 The connection was terminated due to a time-out or other failure!

147021 Ready to try connection again

147071 Attempting to connect now

168322 Error on client pre-Connection Select() [Error=10060] Connection timed out

168332 Ready to try connection again

168332 Attempting to connect now

175142 ****** End of session performance summary ******

175142 Total time connected = 145 seconds

175142 Reception maximum: 17 frames/sec, 460 bytes/sec

175142 Reception average whilst connected: 3 frames/sec, 103 bytes/sec

175142 Transmission maximum: 1 frames/sec, 654 bytes/sec

175142 Transmission average whilst connected: 0 frames/sec, 32 bytes/sec

175142 Max receive buffer = 778, Max send depth = 2, Send frames lost = 0

175142 **************** Individual client application activity ****************

175142 Client 1472 requests: 23 (Ave 0/sec), Data: 45356 bytes (312/sec), Average 1972 bytes/Process

175142 Client 1524 requests: 14 (Ave 0/sec), Data: 756 bytes (5/sec), Average 54 bytes/Process

175142 ********* Log file closed (Buffers: MaxUsed 2, Alloc 556 Freed 556 Refused 0) *********

[Pete Dowson]

So, here is the next test session. Everything was like ealier. I tried PM ABGC to see if the program can connect via WideFS, but it hung up. I also tried my app I am working on, it also couldn't open a connection.

All this really shows is that there's something wrong. I really have no way of finding out what. There is certainly nothing in WideFS you can change for this. It will be something to do with the network, or more likely some program interfering or blocking things.

I'm really not able to help much more. I assume you've been all the way through all the hints and suggestions in the WideFS document? That contains summaries of all I've learned about Networks over the last 7 years of WideFS. If it isn't in there I haven't a clue I'm afraid.

There are only two other suggestions I can make. One is to use the Task Manager (CTRL_ALT_DEL) and delete every unknown or unwanted process in turn. Maybe you'll find the cause that way --- eg. an anti-virus or other blocking type program. Some processes to terminate may crash the system -- just re-boot and remember to leave that one alone next time.

This process has found culprit processes before now, so I think it is your best course of action.

The second suggestion is to see if there's any further advice about Networks you can get from folks who really know them. The expert I usually rely on is Katy Pluta, who you can reach either via PM support or sometimes over in the FS2004 Forum.

Currently there are over 1000 registered WideFS users and none of them have reported anything quite like your problem, not in the 6+ years WideFS has been going. It's a real puzzler I'm afraid.

Regards

Pete