W7 sometimes freezes a few minutes after startup

[Roger Mills]

Paul wrote:

It's a free program supplied - by one of my banks - which is supposed to
make on-line banking safer by checking that when you enter user details
and password you are talking to the bank's genuine site and not to a
spoof phishing site.

"Recently, updates made to Rapport have caused user machines to fail at
boot-up with a Blue Screen of Death; the problems are resolved by
re-naming the file RapportEI.sys.[10]"
(from the above site)

Ah, now then. Given that Win7 uses the BSOD as seldom as possible, that
could be it.
Give it a try. Remove it from loading at boot and see what happens.

It *could* be that - but that article was written in mid-2010, and I've
only be using Rapport since mid-2011 - and hopefully have a later
version with the known bugs removed.

How do I go about finding what causes it to load at boot time?
--
Cheers,
Roger
____________
Please reply to Newsgroup. Whilst email address is valid, it is seldom
checked.

 

[Paul]

Paul wrote:

It's a free program supplied - by one of my banks - which is supposed to
make on-line banking safer by checking that when you enter user details
and password you are talking to the bank's genuine site and not to a
spoof phishing site.

"Recently, updates made to Rapport have caused user machines to fail at
boot-up with a Blue Screen of Death; the problems are resolved by
re-naming the file RapportEI.sys.[10]"
(from the above site)

Ah, now then. Given that Win7 uses the BSOD as seldom as possible, that
could be it.
Give it a try. Remove it from loading at boot and see what happens.

It *could* be that - but that article was written in mid-2010, and I've
only be using Rapport since mid-2011 - and hopefully have a later
version with the known bugs removed.

How do I go about finding what causes it to load at boot time?

Autoruns is one way. It will show you various things like
registry keys that control launch.

http://technet.microsoft.com/en-us/sysinternals/bb963902

Paul

 

[Ed Cryer]

Paul wrote:

It's a free program supplied - by one of my banks - which is supposed to
make on-line banking safer by checking that when you enter user details
and password you are talking to the bank's genuine site and not to a
spoof phishing site.

"Recently, updates made to Rapport have caused user machines to fail at
boot-up with a Blue Screen of Death; the problems are resolved by
re-naming the file RapportEI.sys.[10]"
(from the above site)

Ah, now then. Given that Win7 uses the BSOD as seldom as possible, that
could be it.
Give it a try. Remove it from loading at boot and see what happens.

It *could* be that - but that article was written in mid-2010, and I've
only be using Rapport since mid-2011 - and hopefully have a later
version with the known bugs removed.

How do I go about finding what causes it to load at boot time?

It's usually in the program's settings. If not there, then download
Autoruns;
http://technet.microsoft.com/en-us/sysinternals/bb963902

Ed

 

[Ed Cryer]

Paul wrote:

I noticed in your log, AVG as well as "Trusteer RapportCerberus" ? I
don't know what the latter one does, but that might not be something
the rest of us are using. Just a thought, in terms of things we can
see from here, that are out of the ordinary.

http://en.wikipedia.org/wiki/Trusteer

It's a free program supplied - by one of my banks - which is supposed
to make on-line banking safer by checking that when you enter user
details and password you are talking to the bank's genuine site and
not to a spoof phishing site.

"Recently, updates made to Rapport have caused user machines to fail at
boot-up with a Blue Screen of Death; the problems are resolved by
re-naming the file RapportEI.sys.[10]"
(from the above site)

Ah, now then. Given that Win7 uses the BSOD as seldom as possible, that
could be it.
Give it a try. Remove it from loading at boot and see what happens.

It *could* be that - but that article was written in mid-2010, and
I've only be using Rapport since mid-2011 - and hopefully have a later
version with the known bugs removed.

How do I go about finding what causes it to load at boot time?

Autoruns is one way. It will show you various things like
registry keys that control launch.

http://technet.microsoft.com/en-us/sysinternals/bb963902

Paul

Snap, Paul.
I get identical time-stamps for our two messages.

Ed

 

[ray]

What happens if you shut it down for, say, 5 or 10 minutes after it has
warmed up?

Frankly, I've never seen a hardware problem that was solved by the
machine warming up - generally, heat causes the problem.

 

[Paul]

On 09/02/2012 17:49, Ed Cryer wrote:
Paul wrote:


I noticed in your log, AVG as well as "Trusteer RapportCerberus" ? I
don't know what the latter one does, but that might not be something
the rest of us are using. Just a thought, in terms of things we can
see from here, that are out of the ordinary.

http://en.wikipedia.org/wiki/Trusteer


It's a free program supplied - by one of my banks - which is supposed
to make on-line banking safer by checking that when you enter user
details and password you are talking to the bank's genuine site and
not to a spoof phishing site.


"Recently, updates made to Rapport have caused user machines to fail at
boot-up with a Blue Screen of Death; the problems are resolved by
re-naming the file RapportEI.sys.[10]"
(from the above site)

Ah, now then. Given that Win7 uses the BSOD as seldom as possible, that
could be it.
Give it a try. Remove it from loading at boot and see what happens.

It *could* be that - but that article was written in mid-2010, and
I've only be using Rapport since mid-2011 - and hopefully have a later
version with the known bugs removed.

How do I go about finding what causes it to load at boot time?

Autoruns is one way. It will show you various things like
registry keys that control launch.

http://technet.microsoft.com/en-us/sysinternals/bb963902

Paul

Snap, Paul.
I get identical time-stamps for our two messages.

Ed

Great minds... live in different time zones

Paul

 

[Roger Mills]

What happens if you shut it down for, say, 5 or 10 minutes after it has
warmed up?

Frankly, I've never seen a hardware problem that was solved by the
machine warming up - generally, heat causes the problem.

Dunno - but I doubt whether that would prove anything, bearing in mind
that it doesn't *always* crash just after first boot-up. For example,
today, it's still going after 12 hours.
--
Cheers,
Roger
____________
Please reply to Newsgroup. Whilst email address is valid, it is seldom
checked.

 

[Char Jackson]

It's a free program supplied - by one of my banks - which is supposed to
make on-line banking safer by checking that when you enter user details
and password you are talking to the bank's genuine site and not to a
spoof phishing site.

You shouldn't need that. Always use a saved link rather than typing
the address each time in order to avoid typos, and look for the locked
icon to indicate an HTTPS connection. That should accomplish the same
thing.

 

[ray]

Dunno - but I doubt whether that would prove anything, bearing in mind
that it doesn't *always* crash just after first boot-up. For example,
today, it's still going after 12 hours.

If it would crash after warm up as well as on a cold reboot, that would
tend to indicate that it is not a hardware problem since it is
temperature independent. If it ain't a hardware problem, what's left?

 

[Gene E. Bloch]

If it would crash after warm up as well as on a cold reboot, that would
tend to indicate that it is not a hardware problem since it is
temperature independent. If it ain't a hardware problem, what's left?

Not all hardware problems depend on temperature...

For one example, it could be a bad memory cell that isn't always
addressed, depending on the sequence of events at a particular moment.

 

[ray]

Not all hardware problems depend on temperature...

For one example, it could be a bad memory cell that isn't always
addressed, depending on the sequence of events at a particular moment.

Could be. Fortunately, memory is very easy to test. Nearly all Linux Live
CDs have a memory tester as a boot option.

 

[Gene E. Bloch]

Could be. Fortunately, memory is very easy to test. Nearly all Linux Live
CDs have a memory tester as a boot option.

Memory is very easy to test, but I had a bad experience (OK, I'm
exaggerating).

It was maybe 8 years ago, so we can hope that by now both memory and
the test programs have been improved.

Anyway - my computer back then was crashing often, but it passed
MemTest86 with flying colors more than once. It also passed another
memory test program (my own memory needs testing - I forgot the name).

But I continued to feel from the way it crashed that it was memory, so
to test my idea, I ran the computer on one stick at a time. On stick A
it was consistently fine, and on stick B it consistently crashed...

My only explanation (OK, let's call it a guess) is that the particulars
of the failure didn't match the particulars of what the tests did.
Exact sequence of addressed memory, exact data, whatever...

 

[Paul]

Memory is very easy to test, but I had a bad experience (OK, I'm
exaggerating).

It was maybe 8 years ago, so we can hope that by now both memory and the
test programs have been improved.

Anyway - my computer back then was crashing often, but it passed
MemTest86 with flying colors more than once. It also passed another
memory test program (my own memory needs testing - I forgot the name).

But I continued to feel from the way it crashed that it was memory, so
to test my idea, I ran the computer on one stick at a time. On stick A
it was consistently fine, and on stick B it consistently crashed...

My only explanation (OK, let's call it a guess) is that the particulars
of the failure didn't match the particulars of what the tests did. Exact
sequence of addressed memory, exact data, whatever...

This is why I recommend for memory testing:

1) Use two sticks of RAM at a time.
2) Arrange them in single channel mode.
3) Run the test. One pass is fine for a first test.
4) Swap the two sticks in the two slots. This puts the
high memory stick in the low memory slot, and the low
memory stick in the high memory slot. Run the test again.
Single channel mode is used for these two tests, so the
sticks are given the roles of "high memory" and "low memory"
stick. If you test in dual channel mode, the BIOS reserved
ares is not tested on either stick, and swapping the two
sticks in dual channel mode would achieve nothing. But single
channel mode is different.

By doing it that way, you cover the BIOS reserved area which
memtest86+ (or any other memory tester for that matter) cannot
test.

On motherboards with one slot per bus (exists on some microATX
and also on some of the new X79 boards), well, you're screwed...
In a case like that, the BIOS reserved area cannot be tested.
It's only a megabyte, but a problem there can still cause
issues.

If a test program ignores the BIOS reserved area and tests it
anyway, the BIOS will piss all over the test results. Which is
why programs tend to respect that dividing line.

Paul