X58A-UD3R: BSODs with Intel RST console

[Gary123]

Twice today, I got a BSOD within about ten seconds after opening the Intel RST console: one Cache Manager error, one Bad Pool Handler. No crashes at any other times.

I then installed the Intel RST 10.0.0.1046 WHQL drivers (from station-drivers; previously, I was using 9.​6.​0.​1014 from the Gigabyte product download page). But afterward, I got a third BSOD a few seconds after opening the RST console (this time a System Service Exception).

I don't really need to use the RST console, but the crashes are disconcerting. The system is stable at all other times. The RST console usually works fine, but sometimes the BSODs happen. Possibly, the crashes are occurring when some of my hard drives have spun down before I open the RST console (though I'm not sure of that).

(AHCI enabled, no RAID, SMART disk disabled. Specs in signature.)

[Dark Mantis]

Hi Gary
What exactly was the error code on the BSODs?
Why are you using the RST anyway?

[Peteruk]

They run fine here but this is with RAID but if 9.​6.​0.​1014 was working for you go back to them.

Their was a post on OCZ I saw about the RST 10 not working well with their SSD's or something....
http://www.ocztechnologyforum.com/forum/showthread.php?79426-Intel-RST-Version-10.0.0.1046-WHQL

[Gary123]

What exactly was the error code on the BSODs?

CACHE_MANAGER = 0x34
BAD_POOL_HEADER = 0x19 (I incorrectly wrote 'handler' instead of 'header' earlier)
SYSTEM_SERVICE_EXCEPTION = 0x3B

Why are you using the RST anyway?

It installed with the drivers, so I opened it just out of curiosity. As I said, I don't actually need it. But it shouldn't crash the PC, and I'm concerned that that might indicate a deeper problem.

Gigabyte tech support wants me to try running with one memory stick at a time, even though the problem arises only when I open the RST console.

[Gary123]

if 9.​6.​0.​1014 was working for you go back to them.

No, as I mentioned, the first two crashes happened with 9.6.0.1014. The third crash happened with the newer WHQL driver.

Their was a post on OCZ I saw about the RST 10 not working well with their SSD's or something....

Thanks, saw that, but it's referring to different SSDs than mine, with a particular firmware version that doesn't apply to mine.

[Dark Mantis]

It installed with the drivers, so I opened it just out of curiosity. As I said, I don't actually need it. But it shouldn't crash the PC, and I'm concerned that that might indicate a deeper problem.

Gigabyte tech support wants me to try running with one memory stick at a time, even though the problem arises only when I open the RST console.

It might also pay to check your memory even though it seems fine normally it is still possible that it could be faulty. Try running Memtest86+ on it one stick at a time for at least ten loops each.

http://www.memtest.org/download/beta/415b2/mt415.rar

[Gary123]

It might also pay to check your memory even though it seems fine normally it is still possible that it could be faulty. Try running Memtest86+ on it one stick at a time for at least ten loops each.

Ok, but may I ask the rationale for testing one module at a time, even if the memory passes Memtest86+/Prine95 with all three modules installed? I'm aware that even if each stick passes individually, the memory can still fail the test when all three are installed. But does the converse hold too? That is, are there kinds of errors that will only show up with a single module, but not with all three installed?

[Dark Mantis]

That is what I understand but I must admit I have never had it explained why.

[Gary123]

That is what I understand but I must admit I have never had it explained why.

Thanks, I appreciate both your advice and your honesty.

But the recommended test is quite time-consuming; and whenever components are shuffled around, there's a slight chance of damaging something. I wonder, then, if this test procedure should really be recommended, unless someone can find evidence that it's ever helpful. (The memtest documentation itself does not suggest testing modules individually unless they've shown an error when tested together.)

[Dark Mantis]

Yes I will have to check into it when I have a bit more time to spare. Anyway thanks for your thoughts on it. More often than not we are looking for a faulty memory module anyway so it makes sense to test individually.