Knowledgeable RAID diagnosis requested

  • Thread starter Thread starter ColBlip
  • Start date Start date
C

ColBlip

I recently installed two SATA 80 gb drives as RAID0. All was going well for
about a week. Checks on the speed of the drive came out in the 70 mb/s
range. Then I began to have problems with what can best be described as
"hesitation". Checking the array showed it was going down to about 1 mb/sec
and then up to 60 mb/sec range and then back down again. The peak to peak
cycle was maybe 15-20 seconds.

I then started chasing the problem to see if it was a HD or MB problem. Here
is what I found. I deleted the RAID0 and created two RAID 0+1 so I could
look at them independently. I call the m/b connectors T and B and the drives
d1 and d2. I checked each of the drives/connections separately (i.e. first
connecting T to d1 and testing and T to d2, etc). Here are my results where
OK is normal consistently high performance and not ok is cycle in the drive:

Tests with both drives connected:
T to d1 OK
B to d2 Not OK

T to d2 OK
B to d1 Not OK

B to d1 Ok
T to d2 Not OK
(note - may have written down results on last test in reverse)


Tests with only one drive connected:
T to d1 OK

B to d1 OK

T to d2 OK

B to d2 OK

My conclusion from all of this is that the m/b has a RAID problem that only
shows up when both drives are connected. Could someone who knows this stuff
better than me (which is probably most people here) comment on this and
whether the conclusion is likely valid or did I break some rule of
connecting these things that makes the testing invalid.

If this approach is not valid, suggestions on some other way to get to the
bottom of the problem?

Thanks.

ColBlip.
E-mail: (e-mail address removed)
 
Before you get to tech, you've ran the hd manu checking utility on the hd's?

ColBlip said:
I recently installed two SATA 80 gb drives as RAID0. All was going well for
about a week. Checks on the speed of the drive came out in the 70 mb/s
range. Then I began to have problems with what can best be described as
"hesitation". Checking the array showed it was going down to about 1 mb/sec
and then up to 60 mb/sec range and then back down again. The peak to peak
cycle was maybe 15-20 seconds.

I then started chasing the problem to see if it was a HD or MB problem. Here
is what I found. I deleted the RAID0 and created two RAID 0+1 so I could
look at them independently. I call the m/b connectors T and B and the drives
d1 and d2. I checked each of the drives/connections separately (i.e. first
connecting T to d1 and testing and T to d2, etc). Here are my results where
OK is normal consistently high performance and not ok is cycle in the drive:

Tests with both drives connected:
T to d1 OK
B to d2 Not OK

T to d2 OK
B to d1 Not OK

B to d1 Ok
T to d2 Not OK
(note - may have written down results on last test in reverse)


Tests with only one drive connected:
T to d1 OK

B to d1 OK

T to d2 OK

B to d2 OK

My conclusion from all of this is that the m/b has a RAID problem that only
shows up when both drives are connected. Could someone who knows this stuff
better than me (which is probably most people here) comment on this and
whether the conclusion is likely valid or did I break some rule of
connecting these things that makes the testing invalid.

If this approach is not valid, suggestions on some other way to get to the
bottom of the problem?

Thanks.

ColBlip.
E-mail: (e-mail address removed)
=----
 
Hello, MrGrumpy!
You wrote on Fri, 15 Apr 2005 22:08:51 GMT:

Sorry, what is the "hd manu checking utility on the hd's"?

Thanks.

ColBlip.
E-mail: (e-mail address removed)

M> Before you get to tech, you've ran the hd manu checking utility on the
M> hd's?

M> ??>> I recently installed two SATA 80 gb drives as RAID0. All was going
??>> well
M> for
??>> about a week. Checks on the speed of the drive came out in the 70 mb/s
??>> range. Then I began to have problems with what can best be described
??>> as "hesitation". Checking the array showed it was going down to about
??>> 1
M> mb/sec
??>> and then up to 60 mb/sec range and then back down again. The peak to
??>> peak cycle was maybe 15-20 seconds.
??>>
??>> I then started chasing the problem to see if it was a HD or MB
??>> problem.
M> Here
??>> is what I found. I deleted the RAID0 and created two RAID 0+1 so I
??>> could look at them independently. I call the m/b connectors T and B
??>> and the
M> drives
??>> d1 and d2. I checked each of the drives/connections separately (i.e.
??>> first connecting T to d1 and testing and T to d2, etc). Here are my
??>> results
M> where
??>> OK is normal consistently high performance and not ok is cycle in the
M> drive:
??>>
??>> Tests with both drives connected:
??>> T to d1 OK
??>> B to d2 Not OK
??>>
??>> T to d2 OK
??>> B to d1 Not OK
??>>
??>> B to d1 Ok
??>> T to d2 Not OK
??>> (note - may have written down results on last test in reverse)
??>>
??>> Tests with only one drive connected:
??>> T to d1 OK
??>>
??>> B to d1 OK
??>>
??>> T to d2 OK
??>>
??>> B to d2 OK
??>>
??>> My conclusion from all of this is that the m/b has a RAID problem that
M> only
??>> shows up when both drives are connected. Could someone who knows this
M> stuff
??>> better than me (which is probably most people here) comment on this
??>> and whether the conclusion is likely valid or did I break some rule of
??>> connecting these things that makes the testing invalid.
??>>
??>> If this approach is not valid, suggestions on some other way to get to
??>> the bottom of the problem?
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
M> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
M> =----
 
Hello, Marc!
You wrote on Fri, 15 Apr 2005 18:20:40 -0400:

I am using an Asus A7V8X with onboard RAID.

Thanks.

ColBlip.
E-mail: (e-mail address removed)

MS> Onboard RAID or add-on RAID card?
MS> What model motherboard?

MS> MC

MS> ??>> I recently installed two SATA 80 gb drives as RAID0. All was going
??>> well for about a week. Checks on the speed of the drive came out in
??>> the 70 mb/s range. Then I began to have problems with what can best be
??>> described as "hesitation". Checking the array showed it was going down
??>> to about 1 mb/sec and then up to 60 mb/sec range and then back down
??>> again. The peak to peak cycle was maybe 15-20 seconds.
??>>
??>> I then started chasing the problem to see if it was a HD or MB
??>> problem. Here is what I found. I deleted the RAID0 and created two
??>> RAID 0+1 so I could look at them independently. I call the m/b
??>> connectors T and B and the drives d1 and d2. I checked each of the
??>> drives/connections separately (i.e. first connecting T to d1 and
??>> testing and T to d2, etc). Here are my results where OK is normal
??>> consistently high performance and not ok is cycle in the drive:
??>>
??>> Tests with both drives connected:
??>> T to d1 OK
??>> B to d2 Not OK
??>>
??>> T to d2 OK
??>> B to d1 Not OK
??>>
??>> B to d1 Ok
??>> T to d2 Not OK
??>> (note - may have written down results on last test in reverse)
??>>
??>> Tests with only one drive connected:
??>> T to d1 OK
??>>
??>> B to d1 OK
??>>
??>> T to d2 OK
??>>
??>> B to d2 OK
??>>
??>> My conclusion from all of this is that the m/b has a RAID problem that
??>> only shows up when both drives are connected. Could someone who knows
??>> this stuff better than me (which is probably most people here) comment
??>> on this and whether the conclusion is likely valid or did I break some
??>> rule of connecting these things that makes the testing invalid.
??>>
??>> If this approach is not valid, suggestions on some other way to get to
??>> the bottom of the problem?
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
??>> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
??>> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
??>> =----
 
How did you "check the array"?
??>> T to d2 OK
??>> B to d1 Not OK
??>>
??>> B to d1 Ok
??>> T to d2 Not OK

Isn't above a bit contradicting. Can you explain?
Hello, Marc!
You wrote on Fri, 15 Apr 2005 18:20:40 -0400:

I am using an Asus A7V8X with onboard RAID.

Thanks.

ColBlip.
E-mail: (e-mail address removed)

MS> Onboard RAID or add-on RAID card?
MS> What model motherboard?

MS> MC

MS> ??>> I recently installed two SATA 80 gb drives as RAID0. All was going
??>> well for about a week. Checks on the speed of the drive came out in
??>> the 70 mb/s range. Then I began to have problems with what can best be
??>> described as "hesitation". Checking the array showed it was going down
??>> to about 1 mb/sec and then up to 60 mb/sec range and then back down
??>> again. The peak to peak cycle was maybe 15-20 seconds.
??>>
??>> I then started chasing the problem to see if it was a HD or MB
??>> problem. Here is what I found. I deleted the RAID0 and created two
??>> RAID 0+1 so I could look at them independently. I call the m/b
??>> connectors T and B and the drives d1 and d2. I checked each of the
??>> drives/connections separately (i.e. first connecting T to d1 and
??>> testing and T to d2, etc). Here are my results where OK is normal
??>> consistently high performance and not ok is cycle in the drive:
??>>
??>> Tests with both drives connected:
??>> T to d1 OK
??>> B to d2 Not OK
??>>
??>> T to d2 OK
??>> B to d1 Not OK
??>>
??>> B to d1 Ok
??>> T to d2 Not OK
??>> (note - may have written down results on last test in reverse)
??>>
??>> Tests with only one drive connected:
??>> T to d1 OK
??>>
??>> B to d1 OK
??>>
??>> T to d2 OK
??>>
??>> B to d2 OK
??>>
??>> My conclusion from all of this is that the m/b has a RAID problem that
??>> only shows up when both drives are connected. Could someone who knows
??>> this stuff better than me (which is probably most people here) comment
??>> on this and whether the conclusion is likely valid or did I break some
??>> rule of connecting these things that makes the testing invalid.
??>>
??>> If this approach is not valid, suggestions on some other way to get to
??>> the bottom of the problem?
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
??>> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
??>> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
??>> =----
=----
 
The hard disk manufacturers disk diagnostic checking utility. They are
freely available from the hd manu. web site. Very usefull for diagnosing hd
problems, and can in some cases reserect an apparently failed hd to enable
data to be recovered. Which might be usefull, in your case, since you have
chosen to impliment a raid level which has no fault tollerance.
http://www.acnc.com/raid.html

ColBlip said:
Hello, MrGrumpy!
You wrote on Fri, 15 Apr 2005 22:08:51 GMT:

Sorry, what is the "hd manu checking utility on the hd's"?

Thanks.

ColBlip.
E-mail: (e-mail address removed)

M> Before you get to tech, you've ran the hd manu checking utility on the
M> hd's?

M> ??>> I recently installed two SATA 80 gb drives as RAID0. All was going
??>> well
M> for
??>> about a week. Checks on the speed of the drive came out in the 70 mb/s
??>> range. Then I began to have problems with what can best be described
??>> as "hesitation". Checking the array showed it was going down to about
??>> 1
M> mb/sec
??>> and then up to 60 mb/sec range and then back down again. The peak to
??>> peak cycle was maybe 15-20 seconds.
??>>
??>> I then started chasing the problem to see if it was a HD or MB
??>> problem.
M> Here
??>> is what I found. I deleted the RAID0 and created two RAID 0+1 so I
??>> could look at them independently. I call the m/b connectors T and B
??>> and the
M> drives
??>> d1 and d2. I checked each of the drives/connections separately (i.e.
??>> first connecting T to d1 and testing and T to d2, etc). Here are my
??>> results
M> where
??>> OK is normal consistently high performance and not ok is cycle in the
M> drive:
??>>
??>> Tests with both drives connected:
??>> T to d1 OK
??>> B to d2 Not OK
??>>
??>> T to d2 OK
??>> B to d1 Not OK
??>>
??>> B to d1 Ok
??>> T to d2 Not OK
??>> (note - may have written down results on last test in reverse)
??>>
??>> Tests with only one drive connected:
??>> T to d1 OK
??>>
??>> B to d1 OK
??>>
??>> T to d2 OK
??>>
??>> B to d2 OK
??>>
??>> My conclusion from all of this is that the m/b has a RAID problem that
M> only
??>> shows up when both drives are connected. Could someone who knows this
M> stuff
??>> better than me (which is probably most people here) comment on this
??>> and whether the conclusion is likely valid or did I break some rule of
??>> connecting these things that makes the testing invalid.
??>>
??>> If this approach is not valid, suggestions on some other way to get to
??>> the bottom of the problem?
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
M> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
M> =----
=----
 
Hello, Peter!
You wrote on Fri, 15 Apr 2005 21:19:21 -0400:

P> How did you "check the array"?

??>>>> T to d2 OK
??>>>> B to d1 Not OK
??>>>>
??>>>> B to d1 Ok
??>>>> T to d2 Not OK

P> Isn't above a bit contradicting. Can you explain?

The only way I could do it was to delete the RAID0 array and recreate two
RAID0+1 disks 'arrays', one each on each of the hd's. Then I could see each
of the disks independently. I had no way to check the disks in the array
independently when set as RAID0 since all one sees is a single volume.


[Sorry, skipped]


Thanks.

ColBlip.
E-mail: (e-mail address removed)
 
Hello, MrGrumpy!
You wrote on Sat, 16 Apr 2005 08:21:22 GMT:

I had done this before even starting the thread. These are Hitachi
deathstars but the diagnostic they provide doesn't find a SATA array on my
m/b, only the other 4 IDE devices (PATA) I have running off of the two m/b
connectors.

Thanks.

ColBlip.
E-mail: (e-mail address removed)

M> The hard disk manufacturers disk diagnostic checking utility. They are
M> freely available from the hd manu. web site. Very usefull for diagnosing
M> hd problems, and can in some cases reserect an apparently failed hd to
M> enable data to be recovered. Which might be usefull, in your case, since
M> you have chosen to impliment a raid level which has no fault tollerance.
M> http://www.acnc.com/raid.html

M> ??>> Hello, MrGrumpy!
??>> You wrote on Fri, 15 Apr 2005 22:08:51 GMT:
??>>
??>> Sorry, what is the "hd manu checking utility on the hd's"?
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
M>>> Before you get to tech, you've ran the hd manu checking utility on the
M>>> hd's?
??>>
M>>> ??>>>> I recently installed two SATA 80 gb drives as RAID0. All was going
??>>>> well
M>>> for
??>>>> about a week. Checks on the speed of the drive came out in the 70
M> mb/s
??>>>> range. Then I began to have problems with what can best be described
??>>>> as "hesitation". Checking the array showed it was going down to
M> about
??>>>> 1
M>>> mb/sec
??>>>> and then up to 60 mb/sec range and then back down again. The peak to
??>>>> peak cycle was maybe 15-20 seconds.
??>>>>
??>>>> I then started chasing the problem to see if it was a HD or MB
??>>>> problem.
M>>> Here
??>>>> is what I found. I deleted the RAID0 and created two RAID 0+1 so I
??>>>> could look at them independently. I call the m/b connectors T and B
??>>>> and the
M>>> drives
??>>>> d1 and d2. I checked each of the drives/connections separately (i.e.
??>>>> first connecting T to d1 and testing and T to d2, etc). Here are my
??>>>> results
M>>> where
??>>>> OK is normal consistently high performance and not ok is cycle in
M> the
M>>> drive:
??>>>>
??>>>> Tests with both drives connected:
??>>>> T to d1 OK
??>>>> B to d2 Not OK
??>>>>
??>>>> T to d2 OK
??>>>> B to d1 Not OK
??>>>>
??>>>> B to d1 Ok
??>>>> T to d2 Not OK
??>>>> (note - may have written down results on last test in reverse)
??>>>>
??>>>> Tests with only one drive connected:
??>>>> T to d1 OK
??>>>>
??>>>> B to d1 OK
??>>>>
??>>>> T to d2 OK
??>>>>
??>>>> B to d2 OK
??>>>>
??>>>> My conclusion from all of this is that the m/b has a RAID problem
M> that
M>>> only
??>>>> shows up when both drives are connected. Could someone who knows
M> this
M>>> stuff
??>>>> better than me (which is probably most people here) comment on this
??>>>> and whether the conclusion is likely valid or did I break some rule
M> of
??>>>> connecting these things that makes the testing invalid.
??>>>>
??>>>> If this approach is not valid, suggestions on some other way to get
M> to
??>>>> the bottom of the problem?
??>>>>
??>>>> Thanks.
??>>>>
??>>>> ColBlip.
??>>>> E-mail: (e-mail address removed)
??>>>>
??>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M>>> News==----
??>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>> 120,000+
M>>> Newsgroups
??>>>> ----= East and West-Coast Server Farms - Total Privacy via
M> Encryption
M>>> =----
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
M> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
M> =----
 
According to Hitachi web site their diagnostics support Serial ATA, if the
tool is not finding your drives something is not quite right.
I certainly had no trouble with a Samsung diagnostics testing my Serial
drives, and finding one failing.
Asus have acknowledged problems with Sil controller on earlier bios - I know
this is not relevant, just shows that Asus have had problems with raid
controllers.

ColBlip said:
Hello, MrGrumpy!
You wrote on Sat, 16 Apr 2005 08:21:22 GMT:

I had done this before even starting the thread. These are Hitachi
deathstars but the diagnostic they provide doesn't find a SATA array on my
m/b, only the other 4 IDE devices (PATA) I have running off of the two m/b
connectors.

Thanks.

ColBlip.
E-mail: (e-mail address removed)

M> The hard disk manufacturers disk diagnostic checking utility. They are
M> freely available from the hd manu. web site. Very usefull for diagnosing
M> hd problems, and can in some cases reserect an apparently failed hd to
M> enable data to be recovered. Which might be usefull, in your case, since
M> you have chosen to impliment a raid level which has no fault tollerance.
M> http://www.acnc.com/raid.html

M> ??>> Hello, MrGrumpy!
??>> You wrote on Fri, 15 Apr 2005 22:08:51 GMT:
??>>
??>> Sorry, what is the "hd manu checking utility on the hd's"?
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
M>>> Before you get to tech, you've ran the hd manu checking utility on the
M>>> hd's?
??>>
M>>> ??>>>> I recently installed two SATA 80 gb drives as RAID0. All was going
??>>>> well
M>>> for
??>>>> about a week. Checks on the speed of the drive came out in the 70
M> mb/s
??>>>> range. Then I began to have problems with what can best be described
??>>>> as "hesitation". Checking the array showed it was going down to
M> about
??>>>> 1
M>>> mb/sec
??>>>> and then up to 60 mb/sec range and then back down again. The peak to
??>>>> peak cycle was maybe 15-20 seconds.
??>>>>
??>>>> I then started chasing the problem to see if it was a HD or MB
??>>>> problem.
M>>> Here
??>>>> is what I found. I deleted the RAID0 and created two RAID 0+1 so I
??>>>> could look at them independently. I call the m/b connectors T and B
??>>>> and the
M>>> drives
??>>>> d1 and d2. I checked each of the drives/connections separately (i.e.
??>>>> first connecting T to d1 and testing and T to d2, etc). Here are my
??>>>> results
M>>> where
??>>>> OK is normal consistently high performance and not ok is cycle in
M> the
M>>> drive:
??>>>>
??>>>> Tests with both drives connected:
??>>>> T to d1 OK
??>>>> B to d2 Not OK
??>>>>
??>>>> T to d2 OK
??>>>> B to d1 Not OK
??>>>>
??>>>> B to d1 Ok
??>>>> T to d2 Not OK
??>>>> (note - may have written down results on last test in reverse)
??>>>>
??>>>> Tests with only one drive connected:
??>>>> T to d1 OK
??>>>>
??>>>> B to d1 OK
??>>>>
??>>>> T to d2 OK
??>>>>
??>>>> B to d2 OK
??>>>>
??>>>> My conclusion from all of this is that the m/b has a RAID problem
M> that
M>>> only
??>>>> shows up when both drives are connected. Could someone who knows
M> this
M>>> stuff
??>>>> better than me (which is probably most people here) comment on this
??>>>> and whether the conclusion is likely valid or did I break some rule
M> of
??>>>> connecting these things that makes the testing invalid.
??>>>>
??>>>> If this approach is not valid, suggestions on some other way to get
M> to
??>>>> the bottom of the problem?
??>>>>
??>>>> Thanks.
??>>>>
??>>>> ColBlip.
??>>>> E-mail: (e-mail address removed)
??>>>>
??>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M>>> News==----
??>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>> 120,000+
M>>> Newsgroups
??>>>> ----= East and West-Coast Server Farms - Total Privacy via
M> Encryption
M>>> =----
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
M> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
M> =----
=----
 
Hello, MrGrumpy!
You wrote on Sat, 16 Apr 2005 13:11:42 GMT:

Thanks. Missed the point about SATA when I got the DFT tool. Seems to add
fuel to my fire that the m/b is the source of the problem and not the
drives.


ColBlip.
E-mail: (e-mail address removed)

M> According to Hitachi web site their diagnostics support Serial ATA, if
M> the tool is not finding your drives something is not quite right.
M> I certainly had no trouble with a Samsung diagnostics testing my Serial
M> drives, and finding one failing.
M> Asus have acknowledged problems with Sil controller on earlier bios - I
M> know this is not relevant, just shows that Asus have had problems with
M> raid controllers.

M> ??>> Hello, MrGrumpy!
??>> You wrote on Sat, 16 Apr 2005 08:21:22 GMT:
??>>
??>> I had done this before even starting the thread. These are Hitachi
??>> deathstars but the diagnostic they provide doesn't find a SATA array
??>> on my m/b, only the other 4 IDE devices (PATA) I have running off of
??>> the two m/b connectors.
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
M>>> The hard disk manufacturers disk diagnostic checking utility. They are
M>>> freely available from the hd manu. web site. Very usefull for
M> diagnosing
M>>> hd problems, and can in some cases reserect an apparently failed hd to
M>>> enable data to be recovered. Which might be usefull, in your case,
M> since
M>>> you have chosen to impliment a raid level which has no fault
M> tollerance.
M>>> http://www.acnc.com/raid.html
??>>
M>>> ??>>>> Hello, MrGrumpy!
??>>>> You wrote on Fri, 15 Apr 2005 22:08:51 GMT:
??>>>>
??>>>> Sorry, what is the "hd manu checking utility on the hd's"?
??>>>>
??>>>> Thanks.
??>>>>
??>>>> ColBlip.
??>>>> E-mail: (e-mail address removed)
??>>>>
M>>>>> Before you get to tech, you've ran the hd manu checking utility on
M> the
M>>>>> hd's?
??>>>>
M>>>>> ??>>>>>> I recently installed two SATA 80 gb drives as RAID0. All was going
??>>>>>> well
M>>>>> for
??>>>>>> about a week. Checks on the speed of the drive came out in the 70
M>>> mb/s
??>>>>>> range. Then I began to have problems with what can best be
M> described
??>>>>>> as "hesitation". Checking the array showed it was going down to
M>>> about
??>>>>>> 1
M>>>>> mb/sec
??>>>>>> and then up to 60 mb/sec range and then back down again. The peak
M> to
??>>>>>> peak cycle was maybe 15-20 seconds.
??>>>>>>
??>>>>>> I then started chasing the problem to see if it was a HD or MB
??>>>>>> problem.
M>>>>> Here
??>>>>>> is what I found. I deleted the RAID0 and created two RAID 0+1 so I
??>>>>>> could look at them independently. I call the m/b connectors T and
M> B
??>>>>>> and the
M>>>>> drives
??>>>>>> d1 and d2. I checked each of the drives/connections separately
M> (i.e.
??>>>>>> first connecting T to d1 and testing and T to d2, etc). Here are
M> my
??>>>>>> results
M>>>>> where
??>>>>>> OK is normal consistently high performance and not ok is cycle in
M>>> the
M>>>>> drive:
??>>>>>>
??>>>>>> Tests with both drives connected:
??>>>>>> T to d1 OK
??>>>>>> B to d2 Not OK
??>>>>>>
??>>>>>> T to d2 OK
??>>>>>> B to d1 Not OK
??>>>>>>
??>>>>>> B to d1 Ok
??>>>>>> T to d2 Not OK
??>>>>>> (note - may have written down results on last test in reverse)
??>>>>>>
??>>>>>> Tests with only one drive connected:
??>>>>>> T to d1 OK
??>>>>>>
??>>>>>> B to d1 OK
??>>>>>>
??>>>>>> T to d2 OK
??>>>>>>
??>>>>>> B to d2 OK
??>>>>>>
??>>>>>> My conclusion from all of this is that the m/b has a RAID problem
M>>> that
M>>>>> only
??>>>>>> shows up when both drives are connected. Could someone who knows
M>>> this
M>>>>> stuff
??>>>>>> better than me (which is probably most people here) comment on
M> this
??>>>>>> and whether the conclusion is likely valid or did I break some
M> rule
M>>> of
??>>>>>> connecting these things that makes the testing invalid.
??>>>>>>
??>>>>>> If this approach is not valid, suggestions on some other way to
M> get
M>>> to
??>>>>>> the bottom of the problem?
??>>>>>>
??>>>>>> Thanks.
??>>>>>>
??>>>>>> ColBlip.
??>>>>>> E-mail: (e-mail address removed)
??>>>>>>
??>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M> Usenet
M>>>>> News==----
??>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>> 120,000+
M>>>>> Newsgroups
??>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>> Encryption
M>>>>> =----
??>>>>
??>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M>>> News==----
??>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>> 120,000+
M>>> Newsgroups
??>>>> ----= East and West-Coast Server Farms - Total Privacy via
M> Encryption
M>>> =----
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
M> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
M> =----
 
Hello, MrGrumpy!
You wrote on Sat, 16 Apr 2005 13:11:42 GMT:

Your note prompted me to try this again since deleting the raid0 array and
recreating a couple of raid0+1 drives. Still doesn't see the drives.

Thanks.

ColBlip.
E-mail: (e-mail address removed)

M> According to Hitachi web site their diagnostics support Serial ATA, if
M> the tool is not finding your drives something is not quite right.
M> I certainly had no trouble with a Samsung diagnostics testing my Serial
M> drives, and finding one failing.
M> Asus have acknowledged problems with Sil controller on earlier bios - I
M> know this is not relevant, just shows that Asus have had problems with
M> raid controllers.

M> ??>> Hello, MrGrumpy!
??>> You wrote on Sat, 16 Apr 2005 08:21:22 GMT:
??>>
??>> I had done this before even starting the thread. These are Hitachi
??>> deathstars but the diagnostic they provide doesn't find a SATA array
??>> on my m/b, only the other 4 IDE devices (PATA) I have running off of
??>> the two m/b connectors.
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
M>>> The hard disk manufacturers disk diagnostic checking utility. They are
M>>> freely available from the hd manu. web site. Very usefull for
M> diagnosing
M>>> hd problems, and can in some cases reserect an apparently failed hd to
M>>> enable data to be recovered. Which might be usefull, in your case,
M> since
M>>> you have chosen to impliment a raid level which has no fault
M> tollerance.
M>>> http://www.acnc.com/raid.html
??>>
M>>> ??>>>> Hello, MrGrumpy!
??>>>> You wrote on Fri, 15 Apr 2005 22:08:51 GMT:
??>>>>
??>>>> Sorry, what is the "hd manu checking utility on the hd's"?
??>>>>
??>>>> Thanks.
??>>>>
??>>>> ColBlip.
??>>>> E-mail: (e-mail address removed)
??>>>>
M>>>>> Before you get to tech, you've ran the hd manu checking utility on
M> the
M>>>>> hd's?
??>>>>
M>>>>> ??>>>>>> I recently installed two SATA 80 gb drives as RAID0. All was going
??>>>>>> well
M>>>>> for
??>>>>>> about a week. Checks on the speed of the drive came out in the 70
M>>> mb/s
??>>>>>> range. Then I began to have problems with what can best be
M> described
??>>>>>> as "hesitation". Checking the array showed it was going down to
M>>> about
??>>>>>> 1
M>>>>> mb/sec
??>>>>>> and then up to 60 mb/sec range and then back down again. The peak
M> to
??>>>>>> peak cycle was maybe 15-20 seconds.
??>>>>>>
??>>>>>> I then started chasing the problem to see if it was a HD or MB
??>>>>>> problem.
M>>>>> Here
??>>>>>> is what I found. I deleted the RAID0 and created two RAID 0+1 so I
??>>>>>> could look at them independently. I call the m/b connectors T and
M> B
??>>>>>> and the
M>>>>> drives
??>>>>>> d1 and d2. I checked each of the drives/connections separately
M> (i.e.
??>>>>>> first connecting T to d1 and testing and T to d2, etc). Here are
M> my
??>>>>>> results
M>>>>> where
??>>>>>> OK is normal consistently high performance and not ok is cycle in
M>>> the
M>>>>> drive:
??>>>>>>
??>>>>>> Tests with both drives connected:
??>>>>>> T to d1 OK
??>>>>>> B to d2 Not OK
??>>>>>>
??>>>>>> T to d2 OK
??>>>>>> B to d1 Not OK
??>>>>>>
??>>>>>> B to d1 Ok
??>>>>>> T to d2 Not OK
??>>>>>> (note - may have written down results on last test in reverse)
??>>>>>>
??>>>>>> Tests with only one drive connected:
??>>>>>> T to d1 OK
??>>>>>>
??>>>>>> B to d1 OK
??>>>>>>
??>>>>>> T to d2 OK
??>>>>>>
??>>>>>> B to d2 OK
??>>>>>>
??>>>>>> My conclusion from all of this is that the m/b has a RAID problem
M>>> that
M>>>>> only
??>>>>>> shows up when both drives are connected. Could someone who knows
M>>> this
M>>>>> stuff
??>>>>>> better than me (which is probably most people here) comment on
M> this
??>>>>>> and whether the conclusion is likely valid or did I break some
M> rule
M>>> of
??>>>>>> connecting these things that makes the testing invalid.
??>>>>>>
??>>>>>> If this approach is not valid, suggestions on some other way to
M> get
M>>> to
??>>>>>> the bottom of the problem?
??>>>>>>
??>>>>>> Thanks.
??>>>>>>
??>>>>>> ColBlip.
??>>>>>> E-mail: (e-mail address removed)
??>>>>>>
??>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M> Usenet
M>>>>> News==----
??>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>> 120,000+
M>>>>> Newsgroups
??>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>> Encryption
M>>>>> =----
??>>>>
??>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M>>> News==----
??>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>> 120,000+
M>>> Newsgroups
??>>>> ----= East and West-Coast Server Farms - Total Privacy via
M> Encryption
M>>> =----
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
M> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
M> =----
 
Being one for testing/experimenting, I would beg, borrow or steal an add-on
Raid card (Adaptec, Promise etc.)
Disable the on-board controll in the BIOS and then retest.

MS
 
Excepting if the raid has been built a different raid controller may not see
the raid, and if it does it may corrupt the raid. - Info from Adaptec,
further they state if changing raid controller, sys must be rebuilt as
otherwise they cannot guarantee the raid -
 
Hello, MrGrumpy!
You wrote on Sat, 16 Apr 2005 18:08:19 GMT:

Fortunately, or unfortunately, I'm not concerned about the old array. Let me
rephrase that - based on my limited knowledge I don't THINK I concerned.
I've got the whole Array backed up to an image file so I can restore it. I
think I've already destroyed the raid0 array since I created the 2 raid0+1
disks. Having said that it may be that if I were to create the 2 disk array
via a different controller my backup restore would be worthless?

Thanks.

ColBlip.
E-mail: (e-mail address removed)

M> Excepting if the raid has been built a different raid controller may not
M> see the raid, and if it does it may corrupt the raid. - Info from
M> Adaptec, further they state if changing raid controller, sys must be
M> rebuilt as otherwise they cannot guarantee the raid -

M> ??>> Being one for testing/experimenting, I would beg, borrow or steal an
M> add-on
??>> Raid card (Adaptec, Promise etc.)
??>> Disable the on-board controll in the BIOS and then retest.
??>>
??>> MS
??>> ??>>> Hello, MrGrumpy!
??>>> You wrote on Sat, 16 Apr 2005 13:11:42 GMT:
??>>>
??>>> Your note prompted me to try this again since deleting the raid0
??>>> array
M> and
??>>> recreating a couple of raid0+1 drives. Still doesn't see the drives.
??>>>
??>>> Thanks.
??>>>
??>>> ColBlip.
??>>> E-mail: (e-mail address removed)
??>>>
M>>>> According to Hitachi web site their diagnostics support Serial ATA,
M> if
M>>>> the tool is not finding your drives something is not quite right.
M>>>> I certainly had no trouble with a Samsung diagnostics testing my
M> Serial
M>>>> drives, and finding one failing.
M>>>> Asus have acknowledged problems with Sil controller on earlier bios -
M> I
M>>>> know this is not relevant, just shows that Asus have had problems
M> with
M>>>> raid controllers.
??>>>
M>>>> ??>>>>> Hello, MrGrumpy!
??>>>>> You wrote on Sat, 16 Apr 2005 08:21:22 GMT:
??>>>>>
??>>>>> I had done this before even starting the thread. These are Hitachi
??>>>>> deathstars but the diagnostic they provide doesn't find a SATA
M> array
??>>>>> on my m/b, only the other 4 IDE devices (PATA) I have running off
M> of
??>>>>> the two m/b connectors.
??>>>>>
??>>>>> Thanks.
??>>>>>
??>>>>> ColBlip.
??>>>>> E-mail: (e-mail address removed)
??>>>>>
M>>>>>> The hard disk manufacturers disk diagnostic checking utility. They
??>>> are
M>>>>>> freely available from the hd manu. web site. Very usefull for
M>>>> diagnosing
M>>>>>> hd problems, and can in some cases reserect an apparently failed hd
??>>> to
M>>>>>> enable data to be recovered. Which might be usefull, in your case,
M>>>> since
M>>>>>> you have chosen to impliment a raid level which has no fault
M>>>> tollerance.
M>>>>>> http://www.acnc.com/raid.html
??>>>>>
M>>>>>> ??>>>>>>> Hello, MrGrumpy!
??>>>>>>> You wrote on Fri, 15 Apr 2005 22:08:51 GMT:
??>>>>>>>
??>>>>>>> Sorry, what is the "hd manu checking utility on the hd's"?
??>>>>>>>
??>>>>>>> Thanks.
??>>>>>>>
??>>>>>>> ColBlip.
??>>>>>>> E-mail: (e-mail address removed)
??>>>>>>>
M>>>>>>>> Before you get to tech, you've ran the hd manu checking utility
M> on
M>>>> the
M>>>>>>>> hd's?
??>>>>>>>
M>>>>>>>> ??>>>>>>>>> I recently installed two SATA 80 gb drives as RAID0. All was
??>>> going
??>>>>>>>>> well
M>>>>>>>> for
??>>>>>>>>> about a week. Checks on the speed of the drive came out in the
M> 70
M>>>>>> mb/s
??>>>>>>>>> range. Then I began to have problems with what can best be
M>>>> described
??>>>>>>>>> as "hesitation". Checking the array showed it was going down to
M>>>>>> about
??>>>>>>>>> 1
M>>>>>>>> mb/sec
??>>>>>>>>> and then up to 60 mb/sec range and then back down again. The
M> peak
M>>>> to
??>>>>>>>>> peak cycle was maybe 15-20 seconds.
??>>>>>>>>>
??>>>>>>>>> I then started chasing the problem to see if it was a HD or MB
??>>>>>>>>> problem.
M>>>>>>>> Here
??>>>>>>>>> is what I found. I deleted the RAID0 and created two RAID 0+1
M> so
??>>> I
??>>>>>>>>> could look at them independently. I call the m/b connectors T
M> and
M>>>> B
??>>>>>>>>> and the
M>>>>>>>> drives
??>>>>>>>>> d1 and d2. I checked each of the drives/connections separately
M>>>> (i.e.
??>>>>>>>>> first connecting T to d1 and testing and T to d2, etc). Here
M> are
M>>>> my
??>>>>>>>>> results
M>>>>>>>> where
??>>>>>>>>> OK is normal consistently high performance and not ok is cycle
M> in
M>>>>>> the
M>>>>>>>> drive:
??>>>>>>>>>
??>>>>>>>>> Tests with both drives connected:
??>>>>>>>>> T to d1 OK
??>>>>>>>>> B to d2 Not OK
??>>>>>>>>>
??>>>>>>>>> T to d2 OK
??>>>>>>>>> B to d1 Not OK
??>>>>>>>>>
??>>>>>>>>> B to d1 Ok
??>>>>>>>>> T to d2 Not OK
??>>>>>>>>> (note - may have written down results on last test in reverse)
??>>>>>>>>>
??>>>>>>>>> Tests with only one drive connected:
??>>>>>>>>> T to d1 OK
??>>>>>>>>>
??>>>>>>>>> B to d1 OK
??>>>>>>>>>
??>>>>>>>>> T to d2 OK
??>>>>>>>>>
??>>>>>>>>> B to d2 OK
??>>>>>>>>>
??>>>>>>>>> My conclusion from all of this is that the m/b has a RAID
M> problem
M>>>>>> that
M>>>>>>>> only
??>>>>>>>>> shows up when both drives are connected. Could someone who
M> knows
M>>>>>> this
M>>>>>>>> stuff
??>>>>>>>>> better than me (which is probably most people here) comment on
M>>>> this
??>>>>>>>>> and whether the conclusion is likely valid or did I break some
M>>>> rule
M>>>>>> of
??>>>>>>>>> connecting these things that makes the testing invalid.
??>>>>>>>>>
??>>>>>>>>> If this approach is not valid, suggestions on some other way to
M>>>> get
M>>>>>> to
??>>>>>>>>> the bottom of the problem?
??>>>>>>>>>
??>>>>>>>>> Thanks.
??>>>>>>>>>
??>>>>>>>>> ColBlip.
??>>>>>>>>> E-mail: (e-mail address removed)
??>>>>>>>>>
??>>>>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M>>>> Usenet
M>>>>>>>> News==----
??>>>>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>>>>> 120,000+
M>>>>>>>> Newsgroups
??>>>>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>>>>> Encryption
M>>>>>>>> =----
??>>>>>>>
??>>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
??>>> Usenet
M>>>>>> News==----
??>>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>>> 120,000+
M>>>>>> Newsgroups
??>>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>>> Encryption
M>>>>>> =----
??>>>>>
??>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M> Usenet
M>>>> News==----
??>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>> 120,000+
M>>>> Newsgroups
??>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M> Encryption
M>>>> =----
??>>>
??>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
??>>> News==----
??>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
??>>> Newsgroups
??>>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
??>>> =----
??>>
 
Hello, Marc!
You wrote on Sat, 16 Apr 2005 13:20:08 -0400:

I've got a promise controller that I think can do raid but not sata. I'm out
of luck on laying my hands on a raid sata card. I'm the expert among my
friends so that should scare you. <g>

Thanks.

ColBlip.
E-mail: (e-mail address removed)

MS> Being one for testing/experimenting, I would beg, borrow or steal an
MS> add-on Raid card (Adaptec, Promise etc.)
MS> Disable the on-board controll in the BIOS and then retest.

MS> MS
MS> ??>> Hello, MrGrumpy!
??>> You wrote on Sat, 16 Apr 2005 13:11:42 GMT:
??>>
??>> Your note prompted me to try this again since deleting the raid0 array
??>> and recreating a couple of raid0+1 drives. Still doesn't see the
??>> drives.
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
M>>> According to Hitachi web site their diagnostics support Serial ATA, if
M>>> the tool is not finding your drives something is not quite right.
M>>> I certainly had no trouble with a Samsung diagnostics testing my
M>>> Serial drives, and finding one failing. Asus have acknowledged
M>>> problems with Sil controller on earlier bios - I know this is not
M>>> relevant, just shows that Asus have had problems with raid
M>>> controllers.
??>>
M>>> ??>>>> Hello, MrGrumpy!
??>>>> You wrote on Sat, 16 Apr 2005 08:21:22 GMT:
??>>>>
??>>>> I had done this before even starting the thread. These are Hitachi
??>>>> deathstars but the diagnostic they provide doesn't find a SATA array
??>>>> on my m/b, only the other 4 IDE devices (PATA) I have running off of
??>>>> the two m/b connectors.
??>>>>
??>>>> Thanks.
??>>>>
??>>>> ColBlip.
??>>>> E-mail: (e-mail address removed)
??>>>>
M>>>>> The hard disk manufacturers disk diagnostic checking utility. They
??>> are
M>>>>> freely available from the hd manu. web site. Very usefull for
M>>> diagnosing
M>>>>> hd problems, and can in some cases reserect an apparently failed hd
??>> to
M>>>>> enable data to be recovered. Which might be usefull, in your case,
M>>> since
M>>>>> you have chosen to impliment a raid level which has no fault
M>>> tollerance.
M>>>>> http://www.acnc.com/raid.html
??>>>>
M>>>>> ??>>>>>> Hello, MrGrumpy!
??>>>>>> You wrote on Fri, 15 Apr 2005 22:08:51 GMT:
??>>>>>>
??>>>>>> Sorry, what is the "hd manu checking utility on the hd's"?
??>>>>>>
??>>>>>> Thanks.
??>>>>>>
??>>>>>> ColBlip.
??>>>>>> E-mail: (e-mail address removed)
??>>>>>>
M>>>>>>> Before you get to tech, you've ran the hd manu checking utility on
M>>> the
M>>>>>>> hd's?
??>>>>>>
M>>>>>>> ??>>>>>>>> I recently installed two SATA 80 gb drives as RAID0. All was
??>> going
??>>>>>>>> well
M>>>>>>> for
??>>>>>>>> about a week. Checks on the speed of the drive came out in the
??>>>>>>>> 70
M>>>>> mb/s
??>>>>>>>> range. Then I began to have problems with what can best be
M>>> described
??>>>>>>>> as "hesitation". Checking the array showed it was going down to
M>>>>> about
??>>>>>>>> 1
M>>>>>>> mb/sec
??>>>>>>>> and then up to 60 mb/sec range and then back down again. The
??>>>>>>>> peak
M>>> to
??>>>>>>>> peak cycle was maybe 15-20 seconds.
??>>>>>>>>
??>>>>>>>> I then started chasing the problem to see if it was a HD or MB
??>>>>>>>> problem.
M>>>>>>> Here
??>>>>>>>> is what I found. I deleted the RAID0 and created two RAID 0+1 so

??>> I
??>>>>>>>> could look at them independently. I call the m/b connectors T
??>>>>>>>> and
M>>> B
??>>>>>>>> and the
M>>>>>>> drives
??>>>>>>>> d1 and d2. I checked each of the drives/connections separately
M>>> (i.e.
??>>>>>>>> first connecting T to d1 and testing and T to d2, etc). Here are
M>>> my
??>>>>>>>> results
M>>>>>>> where
??>>>>>>>> OK is normal consistently high performance and not ok is cycle
??>>>>>>>> in
M>>>>> the
M>>>>>>> drive:
??>>>>>>>>
??>>>>>>>> Tests with both drives connected:
??>>>>>>>> T to d1 OK
??>>>>>>>> B to d2 Not OK
??>>>>>>>>
??>>>>>>>> T to d2 OK
??>>>>>>>> B to d1 Not OK
??>>>>>>>>
??>>>>>>>> B to d1 Ok
??>>>>>>>> T to d2 Not OK
??>>>>>>>> (note - may have written down results on last test in reverse)
??>>>>>>>>
??>>>>>>>> Tests with only one drive connected:
??>>>>>>>> T to d1 OK
??>>>>>>>>
??>>>>>>>> B to d1 OK
??>>>>>>>>
??>>>>>>>> T to d2 OK
??>>>>>>>>
??>>>>>>>> B to d2 OK
??>>>>>>>>
??>>>>>>>> My conclusion from all of this is that the m/b has a RAID
??>>>>>>>> problem
M>>>>> that
M>>>>>>> only
??>>>>>>>> shows up when both drives are connected. Could someone who knows
M>>>>> this
M>>>>>>> stuff
??>>>>>>>> better than me (which is probably most people here) comment on
M>>> this
??>>>>>>>> and whether the conclusion is likely valid or did I break some
M>>> rule
M>>>>> of
??>>>>>>>> connecting these things that makes the testing invalid.
??>>>>>>>>
??>>>>>>>> If this approach is not valid, suggestions on some other way to
M>>> get
M>>>>> to
??>>>>>>>> the bottom of the problem?
??>>>>>>>>
??>>>>>>>> Thanks.
??>>>>>>>>
??>>>>>>>> ColBlip.
??>>>>>>>> E-mail: (e-mail address removed)
??>>>>>>>>
??>>>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M>>> Usenet
M>>>>>>> News==----
??>>>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>>>> 120,000+
M>>>>>>> Newsgroups
??>>>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>>>> Encryption
M>>>>>>> =----
??>>>>>>
??>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
??>> Usenet
M>>>>> News==----
??>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>> 120,000+
M>>>>> Newsgroups
??>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>> Encryption
M>>>>> =----
??>>>>
??>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M>>> News==----
??>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>> 120,000+
M>>> Newsgroups
??>>>> ----= East and West-Coast Server Farms - Total Privacy via
??>>>> Encryption
M>>> =----
??>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
??>> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
??>> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
??>> =----
 
The raid controler writes hidden info to the disk, which is inturn used by
the controler to 'access/manipulate the raid' I would assume this info would
be returned if you did an image restore. If however you simply restored data
I would think there would'nt be a problem
I will, in the next couple of days be replacing my Sil raid controller with
an Adaptec. I had mirror raid. I propose, in the first instance to break the
origonal mirror, shutdown, disconect one sata and boot to the other.
assuming everything is ok, I'll install adaptec reboot update the
firmware/drivers then shutdown, connect the single sata to the Adaptec
reboot and test. If all ok connect 2nd sata, reboot and build the mirror.
I'm simply trying to avoid reinstalling everything, as I have full onsite
and offsite backups I'm not to bothered about data loss. - I'm a bit
paranoid re data, as to date I've lost three hd's

ColBlip said:
Hello, MrGrumpy!
You wrote on Sat, 16 Apr 2005 18:08:19 GMT:

Fortunately, or unfortunately, I'm not concerned about the old array. Let me
rephrase that - based on my limited knowledge I don't THINK I concerned.
I've got the whole Array backed up to an image file so I can restore it. I
think I've already destroyed the raid0 array since I created the 2 raid0+1
disks. Having said that it may be that if I were to create the 2 disk array
via a different controller my backup restore would be worthless?

Thanks.

ColBlip.
E-mail: (e-mail address removed)

M> Excepting if the raid has been built a different raid controller may not
M> see the raid, and if it does it may corrupt the raid. - Info from
M> Adaptec, further they state if changing raid controller, sys must be
M> rebuilt as otherwise they cannot guarantee the raid -

M> ??>> Being one for testing/experimenting, I would beg, borrow or steal an
M> add-on
??>> Raid card (Adaptec, Promise etc.)
??>> Disable the on-board controll in the BIOS and then retest.
??>>
??>> MS
??>> ??>>> Hello, MrGrumpy!
??>>> You wrote on Sat, 16 Apr 2005 13:11:42 GMT:
??>>>
??>>> Your note prompted me to try this again since deleting the raid0
??>>> array
M> and
??>>> recreating a couple of raid0+1 drives. Still doesn't see the drives.
??>>>
??>>> Thanks.
??>>>
??>>> ColBlip.
??>>> E-mail: (e-mail address removed)
??>>>
M>>>> According to Hitachi web site their diagnostics support Serial ATA,
M> if
M>>>> the tool is not finding your drives something is not quite right.
M>>>> I certainly had no trouble with a Samsung diagnostics testing my
M> Serial
M>>>> drives, and finding one failing.
M>>>> Asus have acknowledged problems with Sil controller on earlier bios -
M> I
M>>>> know this is not relevant, just shows that Asus have had problems
M> with
M>>>> raid controllers.
??>>>
M>>>> ??>>>>> Hello, MrGrumpy!
??>>>>> You wrote on Sat, 16 Apr 2005 08:21:22 GMT:
??>>>>>
??>>>>> I had done this before even starting the thread. These are Hitachi
??>>>>> deathstars but the diagnostic they provide doesn't find a SATA
M> array
??>>>>> on my m/b, only the other 4 IDE devices (PATA) I have running off
M> of
??>>>>> the two m/b connectors.
??>>>>>
??>>>>> Thanks.
??>>>>>
??>>>>> ColBlip.
??>>>>> E-mail: (e-mail address removed)
??>>>>>
M>>>>>> The hard disk manufacturers disk diagnostic checking utility. They
??>>> are
M>>>>>> freely available from the hd manu. web site. Very usefull for
M>>>> diagnosing
M>>>>>> hd problems, and can in some cases reserect an apparently failed hd
??>>> to
M>>>>>> enable data to be recovered. Which might be usefull, in your case,
M>>>> since
M>>>>>> you have chosen to impliment a raid level which has no fault
M>>>> tollerance.
M>>>>>> http://www.acnc.com/raid.html
??>>>>>
M>>>>>> ??>>>>>>> Hello, MrGrumpy!
??>>>>>>> You wrote on Fri, 15 Apr 2005 22:08:51 GMT:
??>>>>>>>
??>>>>>>> Sorry, what is the "hd manu checking utility on the hd's"?
??>>>>>>>
??>>>>>>> Thanks.
??>>>>>>>
??>>>>>>> ColBlip.
??>>>>>>> E-mail: (e-mail address removed)
??>>>>>>>
M>>>>>>>> Before you get to tech, you've ran the hd manu checking utility
M> on
M>>>> the
M>>>>>>>> hd's?
??>>>>>>>
M>>>>>>>> ??>>>>>>>>> I recently installed two SATA 80 gb drives as RAID0. All was
??>>> going
??>>>>>>>>> well
M>>>>>>>> for
??>>>>>>>>> about a week. Checks on the speed of the drive came out in the
M> 70
M>>>>>> mb/s
??>>>>>>>>> range. Then I began to have problems with what can best be
M>>>> described
??>>>>>>>>> as "hesitation". Checking the array showed it was going down to
M>>>>>> about
??>>>>>>>>> 1
M>>>>>>>> mb/sec
??>>>>>>>>> and then up to 60 mb/sec range and then back down again. The
M> peak
M>>>> to
??>>>>>>>>> peak cycle was maybe 15-20 seconds.
??>>>>>>>>>
??>>>>>>>>> I then started chasing the problem to see if it was a HD or MB
??>>>>>>>>> problem.
M>>>>>>>> Here
??>>>>>>>>> is what I found. I deleted the RAID0 and created two RAID 0+1
M> so
??>>> I
??>>>>>>>>> could look at them independently. I call the m/b connectors T
M> and
M>>>> B
??>>>>>>>>> and the
M>>>>>>>> drives
??>>>>>>>>> d1 and d2. I checked each of the drives/connections separately
M>>>> (i.e.
??>>>>>>>>> first connecting T to d1 and testing and T to d2, etc). Here
M> are
M>>>> my
??>>>>>>>>> results
M>>>>>>>> where
??>>>>>>>>> OK is normal consistently high performance and not ok is cycle
M> in
M>>>>>> the
M>>>>>>>> drive:
??>>>>>>>>>
??>>>>>>>>> Tests with both drives connected:
??>>>>>>>>> T to d1 OK
??>>>>>>>>> B to d2 Not OK
??>>>>>>>>>
??>>>>>>>>> T to d2 OK
??>>>>>>>>> B to d1 Not OK
??>>>>>>>>>
??>>>>>>>>> B to d1 Ok
??>>>>>>>>> T to d2 Not OK
??>>>>>>>>> (note - may have written down results on last test in reverse)
??>>>>>>>>>
??>>>>>>>>> Tests with only one drive connected:
??>>>>>>>>> T to d1 OK
??>>>>>>>>>
??>>>>>>>>> B to d1 OK
??>>>>>>>>>
??>>>>>>>>> T to d2 OK
??>>>>>>>>>
??>>>>>>>>> B to d2 OK
??>>>>>>>>>
??>>>>>>>>> My conclusion from all of this is that the m/b has a RAID
M> problem
M>>>>>> that
M>>>>>>>> only
??>>>>>>>>> shows up when both drives are connected. Could someone who
M> knows
M>>>>>> this
M>>>>>>>> stuff
??>>>>>>>>> better than me (which is probably most people here) comment on
M>>>> this
??>>>>>>>>> and whether the conclusion is likely valid or did I break some
M>>>> rule
M>>>>>> of
??>>>>>>>>> connecting these things that makes the testing invalid.
??>>>>>>>>>
??>>>>>>>>> If this approach is not valid, suggestions on some other way to
M>>>> get
M>>>>>> to
??>>>>>>>>> the bottom of the problem?
??>>>>>>>>>
??>>>>>>>>> Thanks.
??>>>>>>>>>
??>>>>>>>>> ColBlip.
??>>>>>>>>> E-mail: (e-mail address removed)
??>>>>>>>>>
??>>>>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M>>>> Usenet
M>>>>>>>> News==----
??>>>>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>>>>> 120,000+
M>>>>>>>> Newsgroups
??>>>>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>>>>> Encryption
M>>>>>>>> =----
??>>>>>>>
??>>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
??>>> Usenet
M>>>>>> News==----
??>>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>>> 120,000+
M>>>>>> Newsgroups
??>>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>>> Encryption
M>>>>>> =----
??>>>>>
??>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M> Usenet
M>>>> News==----
??>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>> 120,000+
M>>>> Newsgroups
??>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M> Encryption
M>>>> =----
??>>>
??>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
??>>> News==----
??>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
??>>> Newsgroups
??>>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
??>>> =----
??>>
=----
 
Hello, MrGrumpy!
You wrote on Sat, 16 Apr 2005 19:59:24 GMT:

It looks like if I want RAID0 I now have an excuse to upgrade my m/b and
chip. <g>

Been thinking about it but had decided to postpone for a while but believe I
will likely get an A8V Deluxe from Asus. I think it will handle all of my
PATA and ATAPI (or whatever a CD is called) devices as well as the SATA
raid0. If I can't restore the array I'll be aggravated but it is only OS and
programs and with changing a m/b and going to the AMD 64 stuff I would
probably reinstall the the OS anyway.

Thanks.

ColBlip.
E-mail: (e-mail address removed)

M> The raid controler writes hidden info to the disk, which is inturn used
M> by the controler to 'access/manipulate the raid' I would assume this
M> info would be returned if you did an image restore. If however you
M> simply restored data I would think there would'nt be a problem
M> I will, in the next couple of days be replacing my Sil raid controller
M> with an Adaptec. I had mirror raid. I propose, in the first instance to
M> break the origonal mirror, shutdown, disconect one sata and boot to the
M> other. assuming everything is ok, I'll install adaptec reboot update the
M> firmware/drivers then shutdown, connect the single sata to the Adaptec
M> reboot and test. If all ok connect 2nd sata, reboot and build the
M> mirror. I'm simply trying to avoid reinstalling everything, as I have
M> full onsite and offsite backups I'm not to bothered about data loss. -
M> I'm a bit paranoid re data, as to date I've lost three hd's

M> ??>> Hello, MrGrumpy!
??>> You wrote on Sat, 16 Apr 2005 18:08:19 GMT:
??>>
??>> Fortunately, or unfortunately, I'm not concerned about the old array.
??>> Let
M> me
??>> rephrase that - based on my limited knowledge I don't THINK I
??>> concerned. I've got the whole Array backed up to an image file so I
??>> can restore it. I think I've already destroyed the raid0 array since I
??>> created the 2 raid0+1 disks. Having said that it may be that if I were
??>> to create the 2 disk
M> array
??>> via a different controller my backup restore would be worthless?
??>>
??>> Thanks.
??>>
??>> ColBlip.
??>> E-mail: (e-mail address removed)
??>>
M>>> Excepting if the raid has been built a different raid controller may
M> not
M>>> see the raid, and if it does it may corrupt the raid. - Info from
M>>> Adaptec, further they state if changing raid controller, sys must be
M>>> rebuilt as otherwise they cannot guarantee the raid -
??>>
M>>> ??>>>> Being one for testing/experimenting, I would beg, borrow or steal
M> an
M>>> add-on
??>>>> Raid card (Adaptec, Promise etc.)
??>>>> Disable the on-board controll in the BIOS and then retest.
??>>>>
??>>>> MS
??>>>> ??>>>>> Hello, MrGrumpy!
??>>>>> You wrote on Sat, 16 Apr 2005 13:11:42 GMT:
??>>>>>
??>>>>> Your note prompted me to try this again since deleting the raid0
??>>>>> array
M>>> and
??>>>>> recreating a couple of raid0+1 drives. Still doesn't see the
M> drives.
??>>>>>
??>>>>> Thanks.
??>>>>>
??>>>>> ColBlip.
??>>>>> E-mail: (e-mail address removed)
??>>>>>
M>>>>>> According to Hitachi web site their diagnostics support Serial ATA,
M>>> if
M>>>>>> the tool is not finding your drives something is not quite right.
M>>>>>> I certainly had no trouble with a Samsung diagnostics testing my
M>>> Serial
M>>>>>> drives, and finding one failing.
M>>>>>> Asus have acknowledged problems with Sil controller on earlier
M> bios -
M>>> I
M>>>>>> know this is not relevant, just shows that Asus have had problems
M>>> with
M>>>>>> raid controllers.
??>>>>>
M>>>>>> ??>>>>>>> Hello, MrGrumpy!
??>>>>>>> You wrote on Sat, 16 Apr 2005 08:21:22 GMT:
??>>>>>>>
??>>>>>>> I had done this before even starting the thread. These are
M> Hitachi
??>>>>>>> deathstars but the diagnostic they provide doesn't find a SATA
M>>> array
??>>>>>>> on my m/b, only the other 4 IDE devices (PATA) I have running off
M>>> of
??>>>>>>> the two m/b connectors.
??>>>>>>>
??>>>>>>> Thanks.
??>>>>>>>
??>>>>>>> ColBlip.
??>>>>>>> E-mail: (e-mail address removed)
??>>>>>>>
M>>>>>>>> The hard disk manufacturers disk diagnostic checking utility.
M> They
??>>>>> are
M>>>>>>>> freely available from the hd manu. web site. Very usefull for
M>>>>>> diagnosing
M>>>>>>>> hd problems, and can in some cases reserect an apparently failed
M> hd
??>>>>> to
M>>>>>>>> enable data to be recovered. Which might be usefull, in your
M> case,
M>>>>>> since
M>>>>>>>> you have chosen to impliment a raid level which has no fault
M>>>>>> tollerance.
M>>>>>>>> http://www.acnc.com/raid.html
??>>>>>>>
M>>>>>>>> ??>>>>>>>>> Hello, MrGrumpy!
??>>>>>>>>> You wrote on Fri, 15 Apr 2005 22:08:51 GMT:
??>>>>>>>>>
??>>>>>>>>> Sorry, what is the "hd manu checking utility on the hd's"?
??>>>>>>>>>
??>>>>>>>>> Thanks.
??>>>>>>>>>
??>>>>>>>>> ColBlip.
??>>>>>>>>> E-mail: (e-mail address removed)
??>>>>>>>>>
M>>>>>>>>>> Before you get to tech, you've ran the hd manu checking utility
M>>> on
M>>>>>> the
M>>>>>>>>>> hd's?
??>>>>>>>>>
M> message
M>>>>>>>>>> ??>>>>>>>>>>> I recently installed two SATA 80 gb drives as RAID0. All was
??>>>>> going
??>>>>>>>>>>> well
M>>>>>>>>>> for
??>>>>>>>>>>> about a week. Checks on the speed of the drive came out in
M> the
M>>> 70
M>>>>>>>> mb/s
??>>>>>>>>>>> range. Then I began to have problems with what can best be
M>>>>>> described
??>>>>>>>>>>> as "hesitation". Checking the array showed it was going down
M> to
M>>>>>>>> about
??>>>>>>>>>>> 1
M>>>>>>>>>> mb/sec
??>>>>>>>>>>> and then up to 60 mb/sec range and then back down again. The
M>>> peak
M>>>>>> to
??>>>>>>>>>>> peak cycle was maybe 15-20 seconds.
??>>>>>>>>>>>
??>>>>>>>>>>> I then started chasing the problem to see if it was a HD or
M> MB
??>>>>>>>>>>> problem.
M>>>>>>>>>> Here
??>>>>>>>>>>> is what I found. I deleted the RAID0 and created two RAID 0+1
M>>> so
??>>>>> I
??>>>>>>>>>>> could look at them independently. I call the m/b connectors T
M>>> and
M>>>>>> B
??>>>>>>>>>>> and the
M>>>>>>>>>> drives
??>>>>>>>>>>> d1 and d2. I checked each of the drives/connections
M> separately
M>>>>>> (i.e.
??>>>>>>>>>>> first connecting T to d1 and testing and T to d2, etc). Here
M>>> are
M>>>>>> my
??>>>>>>>>>>> results
M>>>>>>>>>> where
??>>>>>>>>>>> OK is normal consistently high performance and not ok is
M> cycle
M>>> in
M>>>>>>>> the
M>>>>>>>>>> drive:
??>>>>>>>>>>>
??>>>>>>>>>>> Tests with both drives connected:
??>>>>>>>>>>> T to d1 OK
??>>>>>>>>>>> B to d2 Not OK
??>>>>>>>>>>>
??>>>>>>>>>>> T to d2 OK
??>>>>>>>>>>> B to d1 Not OK
??>>>>>>>>>>>
??>>>>>>>>>>> B to d1 Ok
??>>>>>>>>>>> T to d2 Not OK
??>>>>>>>>>>> (note - may have written down results on last test in
M> reverse)
??>>>>>>>>>>>
??>>>>>>>>>>> Tests with only one drive connected:
??>>>>>>>>>>> T to d1 OK
??>>>>>>>>>>>
??>>>>>>>>>>> B to d1 OK
??>>>>>>>>>>>
??>>>>>>>>>>> T to d2 OK
??>>>>>>>>>>>
??>>>>>>>>>>> B to d2 OK
??>>>>>>>>>>>
??>>>>>>>>>>> My conclusion from all of this is that the m/b has a RAID
M>>> problem
M>>>>>>>> that
M>>>>>>>>>> only
??>>>>>>>>>>> shows up when both drives are connected. Could someone who
M>>> knows
M>>>>>>>> this
M>>>>>>>>>> stuff
??>>>>>>>>>>> better than me (which is probably most people here) comment
M> on
M>>>>>> this
??>>>>>>>>>>> and whether the conclusion is likely valid or did I break
M> some
M>>>>>> rule
M>>>>>>>> of
??>>>>>>>>>>> connecting these things that makes the testing invalid.
??>>>>>>>>>>>
??>>>>>>>>>>> If this approach is not valid, suggestions on some other way
M> to
M>>>>>> get
M>>>>>>>> to
??>>>>>>>>>>> the bottom of the problem?
??>>>>>>>>>>>
??>>>>>>>>>>> Thanks.
??>>>>>>>>>>>
??>>>>>>>>>>> ColBlip.
??>>>>>>>>>>> E-mail: (e-mail address removed)
??>>>>>>>>>>>
??>>>>>>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M>>>>>> Usenet
M>>>>>>>>>> News==----
??>>>>>>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the
M> World!
??>>>>>>>>> 120,000+
M>>>>>>>>>> Newsgroups
??>>>>>>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>>>>>>> Encryption
M>>>>>>>>>> =----
??>>>>>>>>>
??>>>>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
??>>>>> Usenet
M>>>>>>>> News==----
??>>>>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>>>>> 120,000+
M>>>>>>>> Newsgroups
??>>>>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>>>>> Encryption
M>>>>>>>> =----
??>>>>>>>
??>>>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M>>> Usenet
M>>>>>> News==----
??>>>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>>>>> 120,000+
M>>>>>> Newsgroups
??>>>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M>>> Encryption
M>>>>>> =----
??>>>>>
??>>>>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure
M> Usenet
??>>>>> News==----
??>>>>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
??>> 120,000+
??>>>>> Newsgroups
??>>>>> ----= East and West-Coast Server Farms - Total Privacy via
M> Encryption
??>>>>> =----
??>>>>
??>> ----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet
M> News==----
??>> http://www.newsfeeds.com The #1 Newsgroup Service in the World!
120,000+
M> Newsgroups
??>> ----= East and West-Coast Server Farms - Total Privacy via Encryption
M> =----
 
MMmmm...too bad
Adaptec makes the Serial ATA RAID 1210SA - which is approx. $50US. It
will probably out-perform the one on your mobo.
SIIG, Startech and Promise also makes them for slightly less.

I'd buy the Adaptec - then you can use it with your drives in any (older)
system.

MS
 
The raid controler writes hidden info to the disk, which is inturn used by
the controler to 'access/manipulate the raid' I would assume this info
would
be returned if you did an image restore. If however you simply restored
data
I would think there would'nt be a problem

This is recommended in some situatons - as I say in another post if there is
an issue which affects both drives and one fails first, the one that is left
may also have damaged meta data. They seem to include meta data at the disc
sector level as well so if an image restore is performed a RAID 1 rebuild
may fail at exactly the same point.

I think the OP needs to go back to basics and backup data, reformat both
drives, recreate raid and restore.
 
Back
Top