I don't see any error messages. Could you highlight them somehow? I did a Ctrl+F on "pnErrorDetail=" and only saw "SUCCESS" or CPU credit.

More Prime.log text
 

Here is the Prime.log text between the Thursday evening (reboot) and this afternoon's (reboot after sending my previous message).

[CODE][Sun May 20 18:26:06 2012 - ver 26.2]
Sending result to server: UID: RMAC9.5/DRP_Desktop, M58729757 is not prime. Res64: 52977FCD6A74____. We1: 81AF4982,22961289,00000000, AID: FF9D2C5746DA5FDB676436BC7E09AA2B

URL: http://v5.mersenne.org/v5server/?v=0.95&px=GIMPS&t=ar&g=f482a3d2fefca600c8b67b3c326960e6&k=FF9D2C5746DA5FDB676436BC7E09AA2B&m=UID:+RMAC9.5/DRP_Desktop,+M58729757+is+not+prime.+Res64:+52977FCD6A74____.+We1:+81AF4982,22961289,00000000,+AID:+FF9D2C5746DA5FDB676436BC7E09AA2B%0A&r=100&d=1&n=58729757&sc=22961289&rd=52977FCD6A7440C2&ec=00000000&fftlen=3276800&ss=20221&sh=4E65C56CB2B2A0110F2EEA4D6C07E197

[B][U]Note, the error above occurred on Sunday; no error message; Prime95 processing hangs while waiting for some kind of response; PC rebooted on Thursday[/B][/U]

[Thu May 24 18:15:40 2012 - ver 26.2]
Updating computer information on the server
URL: http://v5.mersenne.org/v5server/?v=0.95&px=GIMPS&t=uc&g=f482a3d2fefca600c8b67b3c326960e6&hg=bc9f42a182f9b81070fe549fc25b3e17&wg=b14168244345408749d06271c30d1ffc&a=Windows,Prime95,v26.2,build+1&c=AMD+Phenom(tm)+II+X4+940+Processor&f=Prefetch,3DNow!,SSE,SSE2&L1=64&L2=512&np=4&hp=1&m=2814&s=3114&h=24&r=1013&L3=6144&ss=32649&sh=F225AAC1FE7264D3C8E4DAEBAB4973A5
== Info: About to connect() to v5.mersenne.org port 80
== Info: Trying 71.6.220.163...
== Info: connected
== Info: Connected to v5.mersenne.org (71.6.220.163) port 80
=> Send header: GET /v5server/?v=0.95&px=GIMPS&t=uc&g=f482a3d2fefca600c8b67b3c326960e6&hg=bc9f42a182f9b81070fe549fc25b3e17&wg=b14168244345408749d06271c30d1ffc&a=Windows,Prime95,v26.2,build+1&c=AMD+Phenom(tm)+II+X4+940+Processor&f=Prefetch,3DNow!,SSE,SSE2&L1=64&L2=512&np=4&hp=1&m=2814&s=3114&h=24&r=1013&L3=6144&ss=32649&sh=F225AAC1FE7264D3C8E4DAEBAB4973A5 HTTP/1.1

Host: v5.mersenne.org

Accept: */*
<= Recv header: HTTP/1.1 200 OK
<= Recv header: Server: Microsoft-IIS/5.0
<= Recv header: Date: Thu, 24 May 2012 23:16:24 GMT
<= Recv header: MicrosoftOfficeWebServer: 5.0_Pub
<= Recv header: X-Powered-By: ASP.NET
<= Recv header: Connection: close
<= Recv header: X-Powered-By: PHP/5.2.5
<= Recv header: Content-type: text/html
<= Recv data: pnErrorResult=0
pnErrorDetail=SUCCESS
g=f482a3d2fefca600c8b67b3c326960e6
u=RMAC9.5
un=RMAC9.5
cn=DRP_Desktop
od=13
==END==

== Info: Closing connection #0
RESPONSE:
pnErrorResult=0
pnErrorDetail=SUCCESS
g=f482a3d2fefca600c8b67b3c326960e6
u=RMAC9.5
un=RMAC9.5
cn=DRP_Desktop
od=13
==END==

Exchanging program options with server
URL: http://v5.mersenne.org/v5server/?v=0.95&px=GIMPS&t=po&g=f482a3d2fefca600c8b67b3c326960e6&ss=32223&sh=7DA3EE7EB12CFCB448C555CA11FC8459
== Info: About to connect() to v5.mersenne.org port 80
== Info: Trying 71.6.220.163...
== Info: connected
== Info: Connected to v5.mersenne.org (71.6.220.163) port 80
=> Send header: GET /v5server/?v=0.95&px=GIMPS&t=po&g=f482a3d2fefca600c8b67b3c326960e6&ss=32223&sh=7DA3EE7EB12CFCB448C555CA11FC8459 HTTP/1.1

Host: v5.mersenne.org

Accept: */*
<= Recv header: HTTP/1.1 200 OK
<= Recv header: Server: Microsoft-IIS/5.0
<= Recv header: Date: Thu, 24 May 2012 23:16:24 GMT
<= Recv header: MicrosoftOfficeWebServer: 5.0_Pub
<= Recv header: X-Powered-By: ASP.NET
<= Recv header: Connection: close
<= Recv header: X-Powered-By: PHP/5.2.5
<= Recv header: Content-type: text/html
<= Recv data: pnErrorResult=0
pnErrorDetail=SUCCESS
od=13
w=0
DayMemory=512
NightMemory=512
Priority=1
DaysOfWork=5
DayStartTime=450
NightStartTime=1410
RunOnBattery=1
==END==

== Info: Closing connection #0
RESPONSE:
pnErrorResult=0
pnErrorDetail=SUCCESS
od=13
w=0
DayMemory=512
NightMemory=512
Priority=1
DaysOfWork=5
DayStartTime=450
NightStartTime=1410
RunOnBattery=1
==END==

Exchanging program options with server
URL: http://v5.mersenne.org/v5server/?v=0.95&px=GIMPS&t=po&g=f482a3d2fefca600c8b67b3c326960e6&c=0&ss=7400&sh=A5C97D5E7CA6ADB9E3B60CA6A8CE8FEA
== Info: About to connect() to v5.mersenne.org port 80
== Info: Trying 71.6.220.163...
== Info: connected
== Info: Connected to v5.mersenne.org (71.6.220.163) port 80
=> Send header: GET /v5server/?v=0.95&px=GIMPS&t=po&g=f482a3d2fefca600c8b67b3c326960e6&c=0&ss=7400&sh=A5C97D5E7CA6ADB9E3B60CA6A8CE8FEA HTTP/1.1

Host: v5.mersenne.org

Accept: */*
<= Recv header: HTTP/1.1 200 OK
<= Recv header: Server: Microsoft-IIS/5.0
<= Recv header: Date: Thu, 24 May 2012 23:16:24 GMT
<= Recv header: MicrosoftOfficeWebServer: 5.0_Pub
<= Recv header: X-Powered-By: ASP.NET
<= Recv header: Connection: close
<= Recv header: X-Powered-By: PHP/5.2.5
<= Recv header: Content-type: text/html
<= Recv data: pnErrorResult=0
pnErrorDetail=SUCCESS
od=13
w=2
DayMemory=512
NightMemory=512
Priority=1
DaysOfWork=5
DayStartTime=450
NightStartTime=1410
RunOnBattery=1
==END==

== Info: Closing connection #0
RESPONSE:
pnErrorResult=0
pnErrorDetail=SUCCESS
od=13
w=2
DayMemory=512
NightMemory=512
Priority=1
DaysOfWork=5
DayStartTime=450
NightStartTime=1410
RunOnBattery=1
==END==
[/CODE]

[I]Multiple additional messages sent to the Prime95 server on Thursday were deleted due to the 10,000 character limit.[/I]

[CODE]Exchanging program options with server
URL: http://v5.mersenne.org/v5server/?v=0.95&px=GIMPS&t=po&g=f482a3d2fefca600c8b67b3c326960e6&w=2&ss=24911&sh=F48FA43092C2DE1F3D1AE9B66E04189E
== Info: About to connect() to v5.mersenne.org port 80
== Info: Trying 71.6.220.163...
== Info: connected
== Info: Connected to v5.mersenne.org (71.6.220.163) port 80
=> Send header: GET /v5server/?v=0.95&px=GIMPS&t=po&g=f482a3d2fefca600c8b67b3c326960e6&w=2&ss=24911&sh=F48FA43092C2DE1F3D1AE9B66E04189E HTTP/1.1

Host: v5.mersenne.org

Accept: */*
<= Recv header: HTTP/1.1 200 OK
<= Recv header: Server: Microsoft-IIS/5.0
<= Recv header: Date: Thu, 24 May 2012 23:16:30 GMT
<= Recv header: MicrosoftOfficeWebServer: 5.0_Pub
<= Recv header: X-Powered-By: ASP.NET
<= Recv header: Connection: close
<= Recv header: X-Powered-By: PHP/5.2.5
<= Recv header: Content-type: text/html
<= Recv data: pnErrorResult=0
pnErrorDetail=SUCCESS
od=14
==END==

== Info: Closing connection #0
RESPONSE:
pnErrorResult=0
pnErrorDetail=SUCCESS
od=14
==END==

[Fri May 25 08:50:22 2012 - ver 26.2]
Sending result to server: UID: RMAC9.5/DRP_Desktop, M56169493 is not prime. Res64: 4A3130B390F4____. We1: DE10989A,37275711,00000000, AID: 579B2FE321C89C93A5D2111C0B2F6928

URL: http://v5.mersenne.org/v5server/?v=0.95&px=GIMPS&t=ar&g=f482a3d2fefca600c8b67b3c326960e6&k=579B2FE321C89C93A5D2111C0B2F6928&m=UID:+RMAC9.5/DRP_Desktop,+M56169493+is+not+prime.+Res64:+4A3130B390F4FBD7.+We1:+DE10989A,37275711,00000000,+AID:+579B2FE321C89C93A5D2111C0B2F6928%0A&r=100&d=1&n=56169493&sc=37275711&rd=4A3130B390F4FBD7&ec=00000000&fftlen=3145728&ss=41&sh=C2EC3AE7CE56AAC9F2077F85A509264A

[B][U]Note, the error above occurred between 05/24 at 23:16 & 05/25 at 08:50; no error message; Prime95 processing hangs while waiting for some kind of response; PC rebooted on Friday[/B][/U]

[Fri May 25 16:02:52 2012 - ver 26.2]
Sending result to server: UID: RMAC9.5/DRP_Desktop, M56169493 is not prime. Res64: 4A3130B390F4FBD7. We1: DE10989A,37275711,00000000, AID: 579B2FE321C89C93A5D2111C0B2F6928

URL: http://v5.mersenne.org/v5server/?v=0.95&px=GIMPS&t=ar&g=f482a3d2fefca600c8b67b3c326960e6&k=579B2FE321C89C93A5D2111C0B2F6928&m=UID:+RMAC9.5/DRP_Desktop,+M56169493+is+not+prime.+Res64:+4A3130B390F4____.+We1:+DE10989A,37275711,00000000,+AID:+579B2FE321C89C93A5D2111C0B2F6928%0A&r=100&d=1&n=56169493&sc=37275711&rd=4A3130B390F4FBD7&ec=00000000&fftlen=3145728&ss=26631&sh=BF613FE7D84525548B8E64966589765D
== Info: About to connect() to v5.mersenne.org port 80
== Info: Trying 71.6.220.163...
== Info: connected
== Info: Connected to v5.mersenne.org (71.6.220.163) port 80
=> Send header: GET /v5server/?v=0.95&px=GIMPS&t=ar&g=f482a3d2fefca600c8b67b3c326960e6&k=579B2FE321C89C93A5D2111C0B2F6928&m=UID:+RMAC9.5/DRP_Desktop,+M56169493+is+not+prime.+Res64:+4A3130B390F4FBD7.+We1:+DE10989A,37275711,00000000,+AID:+579B2FE321C89C93A5D2111C0B2F6928%0A&r=100&d=1&n=56169493&sc=37275711&rd=4A3130B390F4FBD7&ec=00000000&fftlen=3145728&ss=26631&sh=BF613FE7D84525548B8E64966589765D HTTP/1.1

Host: v5.mersenne.org

Accept: */*
<= Recv header: HTTP/1.1 200 OK
<= Recv header: Server: Microsoft-IIS/5.0
<= Recv header: Date: Fri, 25 May 2012 21:03:31 GMT
<= Recv header: MicrosoftOfficeWebServer: 5.0_Pub
<= Recv header: X-Powered-By: ASP.NET
<= Recv header: Connection: close
<= Recv header: X-Powered-By: PHP/5.2.5
<= Recv header: Content-type: text/html
<= Recv data: pnErrorResult=0
pnErrorDetail=CPU credit is 119.0481 GHz-days.
==END==

== Info: Closing connection #0
RESPONSE:
pnErrorResult=0
pnErrorDetail=CPU credit is 119.0481 GHz-days.
==END==
[/CODE]

George, any ideas on changing Prime95 to continue processing while waiting for a server response?

It [i]should[/i] and normally does not depend on actually getting a response. This morning, in fact, I got a timeout on a result but Prime95 continued to work fine. More specifically, I did get an error message. Furthermore, the comm thread is separate from the worker threads, so even if the comm thread runs into an unhandled error, work should still continue...

[code][Comm thread May 25 11:17:14] Sending result to server: UID: Dubslow/Gravemind, M54197029 is not prime. Res64: 472273CC1E771B__. We8: 2616FE68,50510346,00000000, AID: A88F2DBB01299A1E1E2B6E2083629790
[Comm thread May 25 11:17:14]
[Worker #1 May 25 11:17:14] Starting primality test of M42430763 using AVX FFT length 2240K, Pass1=448, Pass2=5K
[Worker #4 May 25 11:17:58] Iteration: 8780000 / 25723987 [34.1315%]. Per iteration time: 11.938 ms.
[Worker #2 May 25 11:17:58] Iteration: 23210000 / 25318487 [91.6721%]. Per iteration time: 11.937 ms.
[Worker #3 May 25 11:18:17] Iteration: 12110000 / 25726303 [47.0724%]. Per iteration time: 12.284 ms.
[Worker #4 May 25 11:19:58] Iteration: 8790000 / 25723987 [34.1704%]. Per iteration time: 11.977 ms.
[Worker #2 May 25 11:19:58] Iteration: 23220000 / 25318487 [91.7116%]. Per iteration time: 11.978 ms.
[Comm thread May 25 11:20:14] CURL library error: Operation timed out after 180006 milliseconds with 0 bytes received
[Comm thread May 25 11:20:14] CURL library error: Operation timed out after 180006 milliseconds with 0 bytes received
[Comm thread May 25 11:20:14] Visit http://mersenneforum.org for help.
[Comm thread May 25 11:20:14] Will try contacting server again in 300 minutes.[/code]

Perhaps it's a bug in an older CURL version that has since been fixed?

[strike]Does this happen with 26.6?[/strike] Yes, apparently. Hmm.

More Clues which May or May not be Applicable
 

On restart this PC gets message about .NET-BroadcastEventWindow.1.0.5 ... needing to be manually stopped. During the most recent restart, I clicked Cancel on the End Program Windows popup window for this item and canceled the PC restart. Canceling the PC restart caused Prime96 to "freeze up" again. Stopping the Prime95.exe using Task Manager and then starting it again by double clicking the Prime95.exe restarted it in its "frozen" state (i.e. Task Manager shows it using 0 CPU and 76K of memory). The Event Viewer doesn't display any thing of interest for this "freeze up" time.

Logging off to restart this PC for real.

The log file isn't informative. Basically it shows that prime95 called the CURL library which called Windows and the OS decided to hang prime95. I'm afraid networking internals are outside my area of expertise.

The prime95 workers are in separate threads, so in theory the workers shouldn't stall. I don't think 27.7 will behave any differently as it is linked with the same cURL library.

Unfortunately, you seem to be alone in having this problem. There must be something unusual about your Windows install.

I'm afraid the only things I can think of to try are 1) building your own prime95 with the latest CURL library. Maybe the cURL people fixed your problem. And 2) you can try putting UseCURL=0 in the [primenet] section of prime.txt. This is described in undoc.txt and should use some really old sockets based comm code. It hasn't been tested in years.