Error reported by Windows BurnIn, not Linux

[Tim (PassMark)]

We're working on the next release of BurnInTest at the moment so i'll send you a build with some extra debugging relating to the serial test to check that the values relating to the CTS check are returning what's expected.

It could simply be that this situation isn't occuring in Linux like in Windows due to differences in device drivers and hardware interaction in Windows.

[wcnackers]

Hi Tim,

It is one machine. I have 2 drives built up, one Windows XP and the other Red Hat. I originally have a USB thumb drive with SLAX. Linux OSs don't flag the error, Windows does.

Errors are constant and right off the bat. I have copied a log file below. Ran for less than a minute but it should give you an idea of the error. The issue of course is why Linux BurnIn isn't flagging the error.

The SP2 is bad by the way. Its not an erroneous error report from the Windows BurnIn.

Thanks!

Bill

SYSTEM INFORMATION
******************
Computer Name: TEST-D3E896E2C8
Date: 03/01/10 15:38:56
Operating system: Windows XP Professional Service Pack 2 build 2600
Number of CPUs: 1 (4 Core(s)/CPU, 1 Logical(s)/Core)
CPU manufacturer: GenuineIntel
CPU type: Intel(R) Core(TM)2 Quad CPU Q9400 @ 2.66GHz
CPU features: MMX SSE SSE2 SSE3 DEP PAE
CPU1 speed: 2659.8 MHz
CPU L2 Cache: 3 MB
RAM: 1014 MB
Video card: Intel(R) Q35 Express Chipset Family (Resolution: 800x600x32)
Disk drive: Model ST380815AS (Size: 74.5GB)
Disk drive: Model USB 2.0 USB Flash Drive USB Device (Size: 3.8GB)

**************
RESULT SUMMARY
**************
Test Start time: Mon Mar 01 15:37:37 2010
Test Stop time: Mon Mar 01 15:38:31 2010
Test Duration: 000h 00m 54s
Test Name Cycles Operations Result Errors Last Error
Serial Port 1 1 72000 PASS 0 No errors
Serial Port 2 4 3000 FAIL 5 COM port Clear To Send (CTS) line stuck high
TEST RUN FAILED
******************
DETAILED EVENT LOG
******************
SERIOUS : 2010-03-01 15:37:39, Serial, COM port Clear To Send (CTS) line stuck high
LOG NOTE: 2010-03-01 15:37:39, Serial, \\.\COM2 Speed: 9600baud ErrorCode: 0
SERIOUS : 2010-03-01 15:37:51, Serial, COM port Clear To Send (CTS) line stuck high
LOG NOTE: 2010-03-01 15:37:51, Serial, \\.\COM2 Speed: 9600baud ErrorCode: 0
SERIOUS : 2010-03-01 15:38:03, Serial, COM port Clear To Send (CTS) line stuck high
LOG NOTE: 2010-03-01 15:38:03, Serial, \\.\COM2 Speed: 9600baud ErrorCode: 0
SERIOUS : 2010-03-01 15:38:15, Serial, COM port Clear To Send (CTS) line stuck high
LOG NOTE: 2010-03-01 15:38:15, Serial, \\.\COM2 Speed: 9600baud ErrorCode: 0
SERIOUS : 2010-03-01 15:38:27, Serial, COM port Clear To Send (CTS) line stuck high
LOG NOTE: 2010-03-01 15:38:27, Serial, \\.\COM2 Speed: 9600baud ErrorCode: 0
LOG NOTE: 2010-03-01 15:38:33, Status, Test run stopped

[Tim (PassMark)]

There may be configuration differences in Windows that could be triggering the error and the same situation isn't reached in linux (assuming this is the same machine and you've been switching between OS's). How often is the error occurring in Windows?

[wcnackers]

Tim, I tried installing Red Hat RHEL 5.4 and got the same results. Linux BurnIn does not flag the error, Windows BurnIn does.

I also tried downloading the latest Linux BurnIn.

Any ideas would be appreciated.

Thanks!

Bill

[wcnackers]

Tim,

Oops. Didn't realize til I read back over my original post that I didn't explain that Serial Port 1 is working fine, its Serial Port 2 where Windows is reporting the error and Slax is not.

Sorry.

[wcnackers]

Hi Tim,

Thanks for the reply.

Disable RTS/CTS is NOT checked. And yes, the test in Linux is chugging along as if nothing is wrong. It is logging the loopback results right along with Serial Port 1.

I assume that whatever the hardware reports back to BurnIn still has to go through the OS. Is it possible that somehow Slax is ignoring, blocking or missing what the hardware is reporting? Does it make any sense to try another Linux distro?

Again, thanks!

[Tim (PassMark)]

As long as the "Disable RTS/CTS" option isn't checked it should be logging this error if detected by Linux, however BurnInTest is logging what the hardware reports which in this case seems to be different between Windows and Linux.

Is the test is still running in BurnInTest Linux (rather than stalling and not logging a particular error)?

[wcnackers]

Hi All,

I have come across an anomaly where Windows BurnIn 5.3 Serial Port loopback test is reporting: "COM port Clear To Send (CTS) line stuck high".

However, both the GUI and the Text mode version of Linux BurnIn 2.0 is not?

Any idea of why Windows BurnIn would report this error but not Linux BurnIn?

Thanks!