WideFS Connection problems

[CATIII]

Dear Pete,

I have WideFS running using TCP/IP on 8 computers altogether. They all work fine except one. I have recently exchanged it's main board along with processor and memory, updated the OS from W2K to XP but still get the same error?!? On startup of Wideclient.exe it says it's connected via the tooltip message, but a right-click on the taskbar item reveals an empty context menu. In fact, Wideclient doesn't really seem to be connected since my application complains about that: "Maybe running on WideClient, but FS not running on Server, or wrong FSUIPC". It takes many, many seconds that Wideclient accepts a connection from my application "IOS". Here's my Wideclient log:

********* WideClient.DLL Log [version 6.22] Class=FS98MAIN *********

Date (dmy): 25/04/04, Time 17:52:22.062: Client name is IOS

62 Attempting to connect now

62 Connection made okay!

20062 Timed out response: connection assumed lost!

20062 Ready to try connection again

20125 Attempting to connect now

20125 Connection made okay!

45125 Timed out response: connection assumed lost!

45125 Ready to try connection again

45187 Attempting to connect now

45187 Connection made okay!

70187 Timed out response: connection assumed lost!

70187 Ready to try connection again

70250 Attempting to connect now

70250 Connection made okay!

95250 Timed out response: connection assumed lost!

95265 Ready to try connection again

95312 Attempting to connect now

95312 Connection made okay!

231968 New Client Application: "IOS" (Id=320)

3199875 New Client Application: "FSinterrogate" (Id=1516)

3347203 Note: Send() request depth is over 100!

3608687 send() done [25 bytes] after 7 retries, request depth is 4882

3977765 send() done [31 bytes] after 6 retries, request depth is 11834

4025375 send() done [25 bytes] after 6 retries, request depth is 12805

4026562 send() done [25 bytes] after 6 retries, request depth is 12844

4093937 send() done [25 bytes] after 6 retries, request depth is 14135

4099062 send() done [31 bytes] after 6 retries, request depth is 14304

4100312 send() done [25 bytes] after 6 retries, request depth is 14344

4101187 send() done [31 bytes] after 7 retries, request depth is 14373

4102234 send() done [31 bytes] after 7 retries, request depth is 14406

4103453 send() done [31 bytes] after 8 retries, request depth is 14441

4492109 LAN data received:

PREQ ulSize=1116, ulTime=146968, uSum=46782, fUsed=0

CC 02 00 FE 05 00 9A D0 2A 13 94 60 30 00 FE 17

4492109 --> Sumcheck error!

4492109 Size Act=1127 Exp=1127, Sum Act=41131, Exp=46782

4493234 GetRecv() missed block? Sequence 146968 jumped to 147410

4519796 send() done [31 bytes] after 6 retries, request depth is 21978

You can see that it needs 5 attempts to connect properly. The client's hostname is IOS. The other serious error appears from time to time - I have no idea hwo to reproduce it. Even with no actions involved, somehow the transmitting is blocked after a few minutes.

All computers are connected via a 20-port switch (100MBit). I changed network cables and ports, but no change :-(

My only hope is that you can read some information from the logfiles that will enlighten me. These errors are driving me crazy (as well as my application that hangs and has to be killed via the task manager). The reason is that when transmitting is blocked Wideclient.exe takes 99% of processor time and renders the systems unusable.

Let me add that these errors occur with WideFS 6.22 and 6.10. My application has not been changed in the last weeks, thus it can't be the reason for this new phenomenom (<= correctly spelled?).

Anyway, thank you very much for your time and help!

Best regards,

Michael

********* WideServer.DLL Log [version 6.22] *********

Using blocksize guide of 4096 bytes

Date (dmy): 25/04/04, Time 17:49:57.125: Server name is ACFT

16406 Initialising server socket now

16640 Incoming connection Accepted ok (skt=4744)

16640 Incoming connection Accepted ok (skt=4756)

16703 Connected to computer "FO-CDU" (skt=4744)

16781 Connected to computer "EICAS" (skt=4756)

16828 Incoming connection Accepted ok (skt=4768)

16859 Incoming connection Accepted ok (skt=4780)

16937 Incoming connection Accepted ok (skt=4792)

16937 Connected to computer "CPT-CDU" (skt=4768)

16969 Connected to computer "FO-MFD" (skt=4780)

17125 Connected to computer "CPT-MFD" (skt=4792)

102187 Incoming connection Accepted ok (skt=300)

102312 Connected to computer "MAP" (skt=300)

118515 Incoming connection Accepted ok (skt=260)

118562 Connected to computer "IOS" (skt=260)

138594 Incoming connection Accepted ok (skt=364)

145062 Connected to computer "IOS" (skt=364)

163625 Incoming connection Accepted ok (skt=324)

169453 Connected to computer "IOS" (skt=324)

188687 Incoming connection Accepted ok (skt=460)

194828 Connected to computer "IOS" (skt=460)

213750 Incoming connection Accepted ok (skt=4836)

219562 Connected to computer "IOS" (skt=4836)

4148469 Send ok but needed 1219 attempts! (1826 of 1826 sent) (skt=4836)

4314750 Retried 1456 times: sends blocked for over 5 secs! (0 of 123 sent), Error=10035 (skt=4836)

4317062 Send ok but needed 2138 attempts! (133 of 133 sent) (skt=4836)

4378500 Send ok but needed 131 attempts! (1619 of 1619 sent) (skt=4836)

4379390 Send ok but needed 82 attempts! (93 of 93 sent) (skt=4836)

4380422 Send ok but needed 153 attempts! (1152 of 1152 sent) (skt=4836)

4381250 Send ok but needed 39 attempts! (87 of 87 sent) (skt=4836)

4381390 Send ok but needed 64 attempts! (1765 of 1765 sent) (skt=4836)

4382359 Send ok but needed 129 attempts! (649 of 649 sent) (skt=4836)

4383359 Send ok but needed 144 attempts! (660 of 660 sent) (skt=4836)

4384297 Send ok but needed 128 attempts! (1757 of 1757 sent) (skt=4836)

4385297 Send ok but needed 127 attempts! (88 of 88 sent) (skt=4836)

4386265 Send ok but needed 141 attempts! (1746 of 1746 sent) (skt=4836)

4387156 Send ok but needed 59 attempts! (87 of 87 sent) (skt=4836)

4387281 Send ok but needed 43 attempts! (1762 of 1762 sent) (skt=4836)

4388281 Send ok but needed 169 attempts! (601 of 601 sent) (skt=4836)

4389328 Send ok but needed 129 attempts! (1666 of 1666 sent) (skt=4836)

4390109 Send ok but needed 36 attempts! (1675 of 1675 sent) (skt=4836)

4390281 Send ok but needed 110 attempts! (1525 of 1525 sent) (skt=4836)

4391109 Send ok but needed 34 attempts! (1463 of 1463 sent) (skt=4836)

[rest is cut off]

[Pete Dowson]

I have WideFS running using TCP/IP on 8 computers altogether. They all work fine except one. I have recently exchanged it's main board along with processor and memory, updated the OS from W2K to XP but still get the same error?!? On startup of Wideclient.exe it says it's connected via the tooltip message, but a right-click on the taskbar item reveals an empty context menu.

What's a context menu? I don't think I do any of those. I don't use tooltips either -- do you mean the "connected" message in the title bar?

Here's my Wideclient log:

Ugh. There's something wrong there all right, but no specific failure reported from Windows. Very odd.

All I know about Networks, and more (thanks to other's contributions) is embodied in the WideFS DOC. Once you've gone through all those possibilities and there's still something wrong I'm afraid I've had it. I hope some Netwrok expert can help you out here. You could try Katy over in the FS2004 Forum, she's helped me a few times.

Incidentally, the ONLY time I've ever seen blocks actually being corrupted, as this indicates:

4492109 LAN data received:

PREQ ulSize=1116, ulTime=146968, uSum=46782, fUsed=0

CC 02 00 FE 05 00 9A D0 2A 13 94 60 30 00 FE 17

4492109 --> Sumcheck error!

was when I most definitely had a bad Network Card.

Let me add that these errors occur with WideFS 6.22 and 6.10. My application has not been changed in the last weeks, thus it can't be the reason for this new phenomenom (<= correctly spelled?).

No, it can't be the application, and it can't be WideFS either -- as you yourself can attest with 7 other PCs happy. It's going to be hardware or software drivers, more likely hardware. Possibly the Network card, or the PCI socket it's in (if not on-board) or the on-board Network chippery (try a Network card instead -- they are cheap these days in any case), or possibly just the specific port on the Switch (try swapping it with one of the others). I would also have suggested cables, but you say you've tried that already.

Regards,

Pete

[Highvolt]

this is very strange indeed...

the only problem I ever had, and produced similar faults, was a bad networkcable. It was a selfmade one, and the pairs weren't twisted at some spots (due to rolling up and rolling out the cable a lot).

try to ping the pc's or send some large file over to one another.

Kind Regards,

Wim

[CATIII]

Hi Pete,

just want to let you know that I was finally able to solve my problem. I'm pretty sure that the QOS support and an "auto"-setting in the network card's connection properties caused the described behaviour. QOS support was removed and the network card set to 100MBit FullDuplex et voila: everything runs fine :-) It took me 2 days to figure it out, but it was worth it...

Best regards,

Michael

[Pete Dowson]

just want to let you know that I was finally able to solve my problem. I'm pretty sure that the QOS support and an "auto"-setting in the network card's connection properties caused the described behaviour. QOS support was removed and the network card set to 100MBit FullDuplex et voila: everything runs fine :-) It took me 2 days to figure it out, but it was worth it...

Okay, that's great. Thanks for letting me know. I'll add these notes to the list in WideFS.doc!

Regards,

Pete