New SQL Server, RAID Planning

We are installing a new SQL server. Right now, I am planning to get 10
drives:
2 - RAID 1 [Mirror] : OS
4 - RAID 10 [Mirrored Stripes] : SQL DB
4 - RAID 10 [Mirrored Stripes] : SQL Logs
There would be a dedicated controller for the OS and one dedicated two
channel controller for the SQL sets.
From everything I have been reading, this is an ideal setup for a new SQL
server.
Is there any reason to consider creating one large eight drive RAID 10 or
RAID 5 for both SQL DB and the logs? Or possibly run six drives in a RAID
10 for SQL DB and then two drives RAID 1 for the log files?
Any other considerations?
Thank You,
Kevin
you want the logs to be on RAID 1+0
Data best on RAID 1+0 also.
Your setup looks good to me.
Greg Jackson
PDX, Oregon
|||Hi Kevin
You generally don't get much from striping the log unless you have an app
that tends to read the log heavily for rollbacks or transactional
replication. So your second suggestion to have 6 drives in RAID10 config and
RAID 1 for log files may have more merit as you'll get the benefit of having
more physical drives to perform the regular data file read / write activity.
You'd still get the redundancy you're after on the log with RAID1 but you'd
be giving the extra spindle to the main data array where it's probably
needed more.
Regards,
Greg Linwood
SQL Server MVP
"Kevin Hammond" <kghammond@.nrscorp.com> wrote in message
news:c6jnf2$gfo$1@.grandcanyon.binc.net...
> We are installing a new SQL server. Right now, I am planning to get 10
> drives:
> 2 - RAID 1 [Mirror] : OS
> 4 - RAID 10 [Mirrored Stripes] : SQL DB
> 4 - RAID 10 [Mirrored Stripes] : SQL Logs
> There would be a dedicated controller for the OS and one dedicated two
> channel controller for the SQL sets.
> From everything I have been reading, this is an ideal setup for a new SQL
> server.
> Is there any reason to consider creating one large eight drive RAID 10 or
> RAID 5 for both SQL DB and the logs? Or possibly run six drives in a RAID
> 10 for SQL DB and then two drives RAID 1 for the log files?
> Any other considerations?
> Thank You,
> Kevin
>

New SQL Server, RAID Planning

We are installing a new SQL server. Right now, I am planning to get 10
drives:
2 - RAID 1 [Mirror] : OS
4 - RAID 10 [Mirrored Stripes] : SQL DB
4 - RAID 10 [Mirrored Stripes] : SQL Logs
There would be a dedicated controller for the OS and one dedicated two
channel controller for the SQL sets.
From everything I have been reading, this is an ideal setup for a new SQL
server.
Is there any reason to consider creating one large eight drive RAID 10 or
RAID 5 for both SQL DB and the logs? Or possibly run six drives in a RAID
10 for SQL DB and then two drives RAID 1 for the log files?
Any other considerations?
Thank You,
Kevinyou want the logs to be on RAID 1+0
Data best on RAID 1+0 also.
Your setup looks good to me.
Greg Jackson
PDX, Oregon|||Hi Kevin
You generally don't get much from striping the log unless you have an app
that tends to read the log heavily for rollbacks or transactional
replication. So your second suggestion to have 6 drives in RAID10 config and
RAID 1 for log files may have more merit as you'll get the benefit of having
more physical drives to perform the regular data file read / write activity.
You'd still get the redundancy you're after on the log with RAID1 but you'd
be giving the extra spindle to the main data array where it's probably
needed more.
Regards,
Greg Linwood
SQL Server MVP
"Kevin Hammond" <kghammond@.nrscorp.com> wrote in message
news:c6jnf2$gfo$1@.grandcanyon.binc.net...
> We are installing a new SQL server. Right now, I am planning to get 10
> drives:
> 2 - RAID 1 [Mirror] : OS
> 4 - RAID 10 [Mirrored Stripes] : SQL DB
> 4 - RAID 10 [Mirrored Stripes] : SQL Logs
> There would be a dedicated controller for the OS and one dedicated two
> channel controller for the SQL sets.
> From everything I have been reading, this is an ideal setup for a new SQL
> server.
> Is there any reason to consider creating one large eight drive RAID 10 or
> RAID 5 for both SQL DB and the logs? Or possibly run six drives in a RAID
> 10 for SQL DB and then two drives RAID 1 for the log files?
> Any other considerations?
> Thank You,
> Kevin
>

New SQL Server, RAID Planning

We are installing a new SQL server. Right now, I am planning to get 10
drives:
2 - RAID 1 [Mirror] : OS
4 - RAID 10 [Mirrored Stripes] : SQL DB
4 - RAID 10 [Mirrored Stripes] : SQL Logs
There would be a dedicated controller for the OS and one dedicated two
channel controller for the SQL sets.
From everything I have been reading, this is an ideal setup for a new SQL
server.
Is there any reason to consider creating one large eight drive RAID 10 or
RAID 5 for both SQL DB and the logs? Or possibly run six drives in a RAID
10 for SQL DB and then two drives RAID 1 for the log files?
Any other considerations?
Thank You,
Kevinyou want the logs to be on RAID 1+0
Data best on RAID 1+0 also.
Your setup looks good to me.
Greg Jackson
PDX, Oregon|||Hi Kevin
You generally don't get much from striping the log unless you have an app
that tends to read the log heavily for rollbacks or transactional
replication. So your second suggestion to have 6 drives in RAID10 config and
RAID 1 for log files may have more merit as you'll get the benefit of having
more physical drives to perform the regular data file read / write activity.
You'd still get the redundancy you're after on the log with RAID1 but you'd
be giving the extra spindle to the main data array where it's probably
needed more.
Regards,
Greg Linwood
SQL Server MVP
"Kevin Hammond" <kghammond@.nrscorp.com> wrote in message
news:c6jnf2$gfo$1@.grandcanyon.binc.net...
> We are installing a new SQL server. Right now, I am planning to get 10
> drives:
> 2 - RAID 1 [Mirror] : OS
> 4 - RAID 10 [Mirrored Stripes] : SQL DB
> 4 - RAID 10 [Mirrored Stripes] : SQL Logs
> There would be a dedicated controller for the OS and one dedicated two
> channel controller for the SQL sets.
> From everything I have been reading, this is an ideal setup for a new SQL
> server.
> Is there any reason to consider creating one large eight drive RAID 10 or
> RAID 5 for both SQL DB and the logs? Or possibly run six drives in a RAID
> 10 for SQL DB and then two drives RAID 1 for the log files?
> Any other considerations?
> Thank You,
> Kevin
>sql

New SQL Server Install Advice

I have a new Dell PowerEdge 2800 running windows 2003. I have 2 RAID
containers as the machine has a PERC 4 controller. The machine has x2 73GB
SCSI and x4 146GB SCSI HDD.
Container 1
--
The OS C drive (partition of 12GB) is in RAID 1 and so is E drive which has
56GB free.
Container 2
--
A D drive for Data I created and also L for Log which I assigned as dynamic
drives.
Is this setup Ok for running SQL2000. The Data files will reside in D and
the Log files will reside in L. Is this Ok for the read/write access that
occurs when SQL is accessed?
Please help.
skcI assume there is at least one processor and some memory on the server as
well.
It will work. It depends on what performance, capacity, and availability
you are needing.
"Skc" <Skc@.discussions.microsoft.com> wrote in message
news:46CDFADB-7AC6-4719-B0A8-8154D6977397@.microsoft.com...
>I have a new Dell PowerEdge 2800 running windows 2003. I have 2 RAID
> containers as the machine has a PERC 4 controller. The machine has x2
> 73GB
> SCSI and x4 146GB SCSI HDD.
> Container 1
> --
> The OS C drive (partition of 12GB) is in RAID 1 and so is E drive which
> has
> 56GB free.
> Container 2
> --
> A D drive for Data I created and also L for Log which I assigned as
> dynamic
> drives.
> Is this setup Ok for running SQL2000. The Data files will reside in D and
> the Log files will reside in L. Is this Ok for the read/write access that
> occurs when SQL is accessed?
> Please help.
> skc|||Xeon 2.8 x2 and 2GB RAM.
We want max. performance. Is this OK?
Furthermore, I installed SQL2000 and it came with 4 CDs:
1> SQL 2000 Standard (Personal)
2> SQL 2000 Standard
3> SP4
4> Reporting Tools
I installed Disk #1, asked me no product key, I selected 5 CALS from the
dropdown (which was how many I bought) and that was it. I have the SQL
Service running and the SQL Tools (enterprise manager). What is the
difference between the 2 Cds mentioned above?
In Enterprise Manager, when I right-click the (local) Server, it says I have
Microsoft SQL Server Std Edition installed. This is very odd, I hope I have
done this right as it did not ask for a serial product key or anything.
Please verify the above it OK.
"Danny" wrote:
> I assume there is at least one processor and some memory on the server as
> well.
> It will work. It depends on what performance, capacity, and availability
> you are needing.
> "Skc" <Skc@.discussions.microsoft.com> wrote in message
> news:46CDFADB-7AC6-4719-B0A8-8154D6977397@.microsoft.com...
> >I have a new Dell PowerEdge 2800 running windows 2003. I have 2 RAID
> > containers as the machine has a PERC 4 controller. The machine has x2
> > 73GB
> > SCSI and x4 146GB SCSI HDD.
> >
> > Container 1
> > --
> > The OS C drive (partition of 12GB) is in RAID 1 and so is E drive which
> > has
> > 56GB free.
> >
> > Container 2
> > --
> > A D drive for Data I created and also L for Log which I assigned as
> > dynamic
> > drives.
> >
> > Is this setup Ok for running SQL2000. The Data files will reside in D and
> > the Log files will reside in L. Is this Ok for the read/write access that
> > occurs when SQL is accessed?
> >
> > Please help.
> >
> > skc
>
>

New SQL Server Install

We just went through a catastrophic failure of an Oracle database. This
database was running on a server with a RAID 5 installed. One of the disks
was having read/write errors. When we pulled the disk from the array to
have the hot spare kick in, the Oracle server failed and it took a lot of
effort to get it back. (I hate oracle).
We are moving to SQL Server 2005 and want to avoid this problem. I have
been asked to get the specs for a cluster. I have been reading up on the
clustering, but came across Transaction Log Shipping. I know that the
Transaction Log shipping does not have an automatic fail over, but it will
move the transaction log to another server and provide another server we can
use for reporting and in the event of a failure on the main server, we could
use the other server until we can bring it up on line.
What in your opinion would be a better solution, cluster or transaction log
shipping? All of our programs will be using this data, and it is important
to recover from a disaster like this as quickly as possible. The cost of
clustering may be preventative, so I would like your input for those who
have used one or both methods for pros and cons. Thanks.
JohnSince you are using SQL Server 2005, consider database mirroring. You don't
need the fancy hardware that clustering does and failover is very quick -
and can be automatic. Also, you can do a database snapshot on the mirror
and use it for reporting.
--
Tom
----
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
https://mvp.support.microsoft.com/profile/Tom.Moreau
"John Wright" <riley_wrightx@.hotmail.com> wrote in message
news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
We just went through a catastrophic failure of an Oracle database. This
database was running on a server with a RAID 5 installed. One of the disks
was having read/write errors. When we pulled the disk from the array to
have the hot spare kick in, the Oracle server failed and it took a lot of
effort to get it back. (I hate oracle).
We are moving to SQL Server 2005 and want to avoid this problem. I have
been asked to get the specs for a cluster. I have been reading up on the
clustering, but came across Transaction Log Shipping. I know that the
Transaction Log shipping does not have an automatic fail over, but it will
move the transaction log to another server and provide another server we can
use for reporting and in the event of a failure on the main server, we could
use the other server until we can bring it up on line.
What in your opinion would be a better solution, cluster or transaction log
shipping? All of our programs will be using this data, and it is important
to recover from a disaster like this as quickly as possible. The cost of
clustering may be preventative, so I would like your input for those who
have used one or both methods for pros and cons. Thanks.
John|||Isn't SQL Server Enterprise edition the only edition that allows for
snapshots? We plan on using Standard Edition which will support two
clusters.
"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:ee6ZxueXIHA.5208@.TK2MSFTNGP04.phx.gbl...
> Since you are using SQL Server 2005, consider database mirroring. You
> don't
> need the fancy hardware that clustering does and failover is very quick -
> and can be automatic. Also, you can do a database snapshot on the mirror
> and use it for reporting.
> --
> Tom
> ----
> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
> SQL Server MVP
> Toronto, ON Canada
> https://mvp.support.microsoft.com/profile/Tom.Moreau
>
> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
> news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
> We just went through a catastrophic failure of an Oracle database. This
> database was running on a server with a RAID 5 installed. One of the disks
> was having read/write errors. When we pulled the disk from the array to
> have the hot spare kick in, the Oracle server failed and it took a lot of
> effort to get it back. (I hate oracle).
> We are moving to SQL Server 2005 and want to avoid this problem. I have
> been asked to get the specs for a cluster. I have been reading up on the
> clustering, but came across Transaction Log Shipping. I know that the
> Transaction Log shipping does not have an automatic fail over, but it will
> move the transaction log to another server and provide another server we
> can
> use for reporting and in the event of a failure on the main server, we
> could
> use the other server until we can bring it up on line.
> What in your opinion would be a better solution, cluster or transaction
> log
> shipping? All of our programs will be using this data, and it is
> important
> to recover from a disaster like this as quickly as possible. The cost of
> clustering may be preventative, so I would like your input for those who
> have used one or both methods for pros and cons. Thanks.
> John
>
>|||Yes, EE is required for database snapshots. Mirroring is available for SE but only on sync mode and
only one REDO thread.
--
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://sqlblog.com/blogs/tibor_karaszi
"John Wright" <riley_wrightx@.hotmail.com> wrote in message
news:%23In0b5eXIHA.4440@.TK2MSFTNGP06.phx.gbl...
> Isn't SQL Server Enterprise edition the only edition that allows for snapshots? We plan on using
> Standard Edition which will support two clusters.
>
> "Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
> news:ee6ZxueXIHA.5208@.TK2MSFTNGP04.phx.gbl...
>> Since you are using SQL Server 2005, consider database mirroring. You don't
>> need the fancy hardware that clustering does and failover is very quick -
>> and can be automatic. Also, you can do a database snapshot on the mirror
>> and use it for reporting.
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
>> SQL Server MVP
>> Toronto, ON Canada
>> https://mvp.support.microsoft.com/profile/Tom.Moreau
>>
>> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
>> news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
>> We just went through a catastrophic failure of an Oracle database. This
>> database was running on a server with a RAID 5 installed. One of the disks
>> was having read/write errors. When we pulled the disk from the array to
>> have the hot spare kick in, the Oracle server failed and it took a lot of
>> effort to get it back. (I hate oracle).
>> We are moving to SQL Server 2005 and want to avoid this problem. I have
>> been asked to get the specs for a cluster. I have been reading up on the
>> clustering, but came across Transaction Log Shipping. I know that the
>> Transaction Log shipping does not have an automatic fail over, but it will
>> move the transaction log to another server and provide another server we can
>> use for reporting and in the event of a failure on the main server, we could
>> use the other server until we can bring it up on line.
>> What in your opinion would be a better solution, cluster or transaction log
>> shipping? All of our programs will be using this data, and it is important
>> to recover from a disaster like this as quickly as possible. The cost of
>> clustering may be preventative, so I would like your input for those who
>> have used one or both methods for pros and cons. Thanks.
>> John
>>
>|||Yes, snapshots are supported only in Ent Ed. Also, Std Edition supports
2-node clusters - not "two clusters". It depends on where you want to put
your money and what your needs are. When it comes to High Availability
(HA), then you often have to go to Ent Ed. If you want clustering and have
only 2 nodes, then you can get away with Std Ed, but clustering hardware is
usually a bit more expensive and a little trickier to set up. Mirroring
will work on regular hardware and can go between dissimilar platforms.
Automatic failover, though is available in the Ent Ed.
When it comes to reporting, snapshots are your friend. Log shipping is fine
for HA, but you can't report off of the destination DB.
--
Tom
----
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
https://mvp.support.microsoft.com/profile/Tom.Moreau
"John Wright" <riley_wrightx@.hotmail.com> wrote in message
news:%23In0b5eXIHA.4440@.TK2MSFTNGP06.phx.gbl...
Isn't SQL Server Enterprise edition the only edition that allows for
snapshots? We plan on using Standard Edition which will support two
clusters.
"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:ee6ZxueXIHA.5208@.TK2MSFTNGP04.phx.gbl...
> Since you are using SQL Server 2005, consider database mirroring. You
> don't
> need the fancy hardware that clustering does and failover is very quick -
> and can be automatic. Also, you can do a database snapshot on the mirror
> and use it for reporting.
> --
> Tom
> ----
> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
> SQL Server MVP
> Toronto, ON Canada
> https://mvp.support.microsoft.com/profile/Tom.Moreau
>
> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
> news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
> We just went through a catastrophic failure of an Oracle database. This
> database was running on a server with a RAID 5 installed. One of the disks
> was having read/write errors. When we pulled the disk from the array to
> have the hot spare kick in, the Oracle server failed and it took a lot of
> effort to get it back. (I hate oracle).
> We are moving to SQL Server 2005 and want to avoid this problem. I have
> been asked to get the specs for a cluster. I have been reading up on the
> clustering, but came across Transaction Log Shipping. I know that the
> Transaction Log shipping does not have an automatic fail over, but it will
> move the transaction log to another server and provide another server we
> can
> use for reporting and in the event of a failure on the main server, we
> could
> use the other server until we can bring it up on line.
> What in your opinion would be a better solution, cluster or transaction
> log
> shipping? All of our programs will be using this data, and it is
> important
> to recover from a disaster like this as quickly as possible. The cost of
> clustering may be preventative, so I would like your input for those who
> have used one or both methods for pros and cons. Thanks.
> John
>
>|||> Log shipping is fine
> for HA, but you can't report off of the destination DB.
Well, you can if you do the restores using STANDBY. But you'd have to kick out the users each time
you want to do the next restore. Which means that you really ... can't use a log shipped database
for reporting ;-).
(Just wanted to mention this before anyone else say "You can". )
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://sqlblog.com/blogs/tibor_karaszi
"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:uy0aAMfXIHA.1188@.TK2MSFTNGP04.phx.gbl...
> Yes, snapshots are supported only in Ent Ed. Also, Std Edition supports
> 2-node clusters - not "two clusters". It depends on where you want to put
> your money and what your needs are. When it comes to High Availability
> (HA), then you often have to go to Ent Ed. If you want clustering and have
> only 2 nodes, then you can get away with Std Ed, but clustering hardware is
> usually a bit more expensive and a little trickier to set up. Mirroring
> will work on regular hardware and can go between dissimilar platforms.
> Automatic failover, though is available in the Ent Ed.
> When it comes to reporting, snapshots are your friend. Log shipping is fine
> for HA, but you can't report off of the destination DB.
> --
> Tom
> ----
> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
> SQL Server MVP
> Toronto, ON Canada
> https://mvp.support.microsoft.com/profile/Tom.Moreau
>
> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
> news:%23In0b5eXIHA.4440@.TK2MSFTNGP06.phx.gbl...
> Isn't SQL Server Enterprise edition the only edition that allows for
> snapshots? We plan on using Standard Edition which will support two
> clusters.
>
> "Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
> news:ee6ZxueXIHA.5208@.TK2MSFTNGP04.phx.gbl...
>> Since you are using SQL Server 2005, consider database mirroring. You
>> don't
>> need the fancy hardware that clustering does and failover is very quick -
>> and can be automatic. Also, you can do a database snapshot on the mirror
>> and use it for reporting.
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
>> SQL Server MVP
>> Toronto, ON Canada
>> https://mvp.support.microsoft.com/profile/Tom.Moreau
>>
>> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
>> news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
>> We just went through a catastrophic failure of an Oracle database. This
>> database was running on a server with a RAID 5 installed. One of the disks
>> was having read/write errors. When we pulled the disk from the array to
>> have the hot spare kick in, the Oracle server failed and it took a lot of
>> effort to get it back. (I hate oracle).
>> We are moving to SQL Server 2005 and want to avoid this problem. I have
>> been asked to get the specs for a cluster. I have been reading up on the
>> clustering, but came across Transaction Log Shipping. I know that the
>> Transaction Log shipping does not have an automatic fail over, but it will
>> move the transaction log to another server and provide another server we
>> can
>> use for reporting and in the event of a failure on the main server, we
>> could
>> use the other server until we can bring it up on line.
>> What in your opinion would be a better solution, cluster or transaction
>> log
>> shipping? All of our programs will be using this data, and it is
>> important
>> to recover from a disaster like this as quickly as possible. The cost of
>> clustering may be preventative, so I would like your input for those who
>> have used one or both methods for pros and cons. Thanks.
>> John
>>
>
>|||Low-end clusters have definitely gotten cheaper over the past few years.
Look at some of the dual-socket HP stuff and you can put together a good,
stable, basic cluster for under six figures.
--
Geoff N. Hiten
Senior SQL Infrastructure Consultant
Microsoft SQL Server MVP
"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:uy0aAMfXIHA.1188@.TK2MSFTNGP04.phx.gbl...
> Yes, snapshots are supported only in Ent Ed. Also, Std Edition supports
> 2-node clusters - not "two clusters". It depends on where you want to put
> your money and what your needs are. When it comes to High Availability
> (HA), then you often have to go to Ent Ed. If you want clustering and
> have
> only 2 nodes, then you can get away with Std Ed, but clustering hardware
> is
> usually a bit more expensive and a little trickier to set up. Mirroring
> will work on regular hardware and can go between dissimilar platforms.
> Automatic failover, though is available in the Ent Ed.
> When it comes to reporting, snapshots are your friend. Log shipping is
> fine
> for HA, but you can't report off of the destination DB.
> --
> Tom
> ----
> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
> SQL Server MVP
> Toronto, ON Canada
> https://mvp.support.microsoft.com/profile/Tom.Moreau
>
> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
> news:%23In0b5eXIHA.4440@.TK2MSFTNGP06.phx.gbl...
> Isn't SQL Server Enterprise edition the only edition that allows for
> snapshots? We plan on using Standard Edition which will support two
> clusters.
>
> "Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
> news:ee6ZxueXIHA.5208@.TK2MSFTNGP04.phx.gbl...
>> Since you are using SQL Server 2005, consider database mirroring. You
>> don't
>> need the fancy hardware that clustering does and failover is very quick -
>> and can be automatic. Also, you can do a database snapshot on the mirror
>> and use it for reporting.
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
>> SQL Server MVP
>> Toronto, ON Canada
>> https://mvp.support.microsoft.com/profile/Tom.Moreau
>>
>> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
>> news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
>> We just went through a catastrophic failure of an Oracle database. This
>> database was running on a server with a RAID 5 installed. One of the
>> disks
>> was having read/write errors. When we pulled the disk from the array to
>> have the hot spare kick in, the Oracle server failed and it took a lot of
>> effort to get it back. (I hate oracle).
>> We are moving to SQL Server 2005 and want to avoid this problem. I have
>> been asked to get the specs for a cluster. I have been reading up on the
>> clustering, but came across Transaction Log Shipping. I know that the
>> Transaction Log shipping does not have an automatic fail over, but it
>> will
>> move the transaction log to another server and provide another server we
>> can
>> use for reporting and in the event of a failure on the main server, we
>> could
>> use the other server until we can bring it up on line.
>> What in your opinion would be a better solution, cluster or transaction
>> log
>> shipping? All of our programs will be using this data, and it is
>> important
>> to recover from a disaster like this as quickly as possible. The cost of
>> clustering may be preventative, so I would like your input for those who
>> have used one or both methods for pros and cons. Thanks.
>> John
>>
>
>|||Waaaaaay back in 2000, we created our own jobs to logship and only had the
restore command run every 8 hours. All the files were there, and we could
catch the whole thing up in about 20 minutes if the primary went down, but
it was a reporting server all day :)
Same thing could be done here if Log Shipping was desired, as well as
reporting. May not be the most glamorous solution, but it works...
--
Kevin3NF
SQL Server dude
You want fries with that?
http://kevin3nf.blogspot.com/
I only check the newsgroups during work hours, M-F.
Hit my blog and the contact links if necessary...I may be available.
"Tibor Karaszi" <tibor_please.no.email_karaszi@.hotmail.nomail.com> wrote in
message news:0875841B-779E-4639-9C7F-31D19FCF060A@.microsoft.com...
>> Log shipping is fine
>> for HA, but you can't report off of the destination DB.
> Well, you can if you do the restores using STANDBY. But you'd have to kick
> out the users each time you want to do the next restore. Which means that
> you really ... can't use a log shipped database for reporting ;-).
> (Just wanted to mention this before anyone else say "You can". )
>
> --
> Tibor Karaszi, SQL Server MVP
> http://www.karaszi.com/sqlserver/default.asp
> http://sqlblog.com/blogs/tibor_karaszi
>
> "Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
> news:uy0aAMfXIHA.1188@.TK2MSFTNGP04.phx.gbl...
>> Yes, snapshots are supported only in Ent Ed. Also, Std Edition supports
>> 2-node clusters - not "two clusters". It depends on where you want to
>> put
>> your money and what your needs are. When it comes to High Availability
>> (HA), then you often have to go to Ent Ed. If you want clustering and
>> have
>> only 2 nodes, then you can get away with Std Ed, but clustering hardware
>> is
>> usually a bit more expensive and a little trickier to set up. Mirroring
>> will work on regular hardware and can go between dissimilar platforms.
>> Automatic failover, though is available in the Ent Ed.
>> When it comes to reporting, snapshots are your friend. Log shipping is
>> fine
>> for HA, but you can't report off of the destination DB.
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
>> SQL Server MVP
>> Toronto, ON Canada
>> https://mvp.support.microsoft.com/profile/Tom.Moreau
>>
>> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
>> news:%23In0b5eXIHA.4440@.TK2MSFTNGP06.phx.gbl...
>> Isn't SQL Server Enterprise edition the only edition that allows for
>> snapshots? We plan on using Standard Edition which will support two
>> clusters.
>>
>> "Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
>> news:ee6ZxueXIHA.5208@.TK2MSFTNGP04.phx.gbl...
>> Since you are using SQL Server 2005, consider database mirroring. You
>> don't
>> need the fancy hardware that clustering does and failover is very
>> quick -
>> and can be automatic. Also, you can do a database snapshot on the
>> mirror
>> and use it for reporting.
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
>> SQL Server MVP
>> Toronto, ON Canada
>> https://mvp.support.microsoft.com/profile/Tom.Moreau
>>
>> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
>> news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
>> We just went through a catastrophic failure of an Oracle database. This
>> database was running on a server with a RAID 5 installed. One of the
>> disks
>> was having read/write errors. When we pulled the disk from the array to
>> have the hot spare kick in, the Oracle server failed and it took a lot
>> of
>> effort to get it back. (I hate oracle).
>> We are moving to SQL Server 2005 and want to avoid this problem. I have
>> been asked to get the specs for a cluster. I have been reading up on
>> the
>> clustering, but came across Transaction Log Shipping. I know that the
>> Transaction Log shipping does not have an automatic fail over, but it
>> will
>> move the transaction log to another server and provide another server we
>> can
>> use for reporting and in the event of a failure on the main server, we
>> could
>> use the other server until we can bring it up on line.
>> What in your opinion would be a better solution, cluster or transaction
>> log
>> shipping? All of our programs will be using this data, and it is
>> important
>> to recover from a disaster like this as quickly as possible. The cost
>> of
>> clustering may be preventative, so I would like your input for those who
>> have used one or both methods for pros and cons. Thanks.
>> John
>>
>>
>>
>|||Yeah, I figured I'd give him the "short answer". Of course, if you want,
you can have the log shipping sync on a daily schedule and use STANDBY.
That could be a poor man's way of doing mirroring with a daily snapshot.
--
Tom
----
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
SQL Server MVP
Toronto, ON Canada
https://mvp.support.microsoft.com/profile/Tom.Moreau
"Tibor Karaszi" <tibor_please.no.email_karaszi@.hotmail.nomail.com> wrote in
message news:0875841B-779E-4639-9C7F-31D19FCF060A@.microsoft.com...
> Log shipping is fine
> for HA, but you can't report off of the destination DB.
Well, you can if you do the restores using STANDBY. But you'd have to kick
out the users each time
you want to do the next restore. Which means that you really ... can't use a
log shipped database
for reporting ;-).
(Just wanted to mention this before anyone else say "You can". )
Tibor Karaszi, SQL Server MVP
http://www.karaszi.com/sqlserver/default.asp
http://sqlblog.com/blogs/tibor_karaszi
"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:uy0aAMfXIHA.1188@.TK2MSFTNGP04.phx.gbl...
> Yes, snapshots are supported only in Ent Ed. Also, Std Edition supports
> 2-node clusters - not "two clusters". It depends on where you want to put
> your money and what your needs are. When it comes to High Availability
> (HA), then you often have to go to Ent Ed. If you want clustering and
have
> only 2 nodes, then you can get away with Std Ed, but clustering hardware
is
> usually a bit more expensive and a little trickier to set up. Mirroring
> will work on regular hardware and can go between dissimilar platforms.
> Automatic failover, though is available in the Ent Ed.
> When it comes to reporting, snapshots are your friend. Log shipping is
fine
> for HA, but you can't report off of the destination DB.
> --
> Tom
> ----
> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
> SQL Server MVP
> Toronto, ON Canada
> https://mvp.support.microsoft.com/profile/Tom.Moreau
>
> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
> news:%23In0b5eXIHA.4440@.TK2MSFTNGP06.phx.gbl...
> Isn't SQL Server Enterprise edition the only edition that allows for
> snapshots? We plan on using Standard Edition which will support two
> clusters.
>
> "Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
> news:ee6ZxueXIHA.5208@.TK2MSFTNGP04.phx.gbl...
>> Since you are using SQL Server 2005, consider database mirroring. You
>> don't
>> need the fancy hardware that clustering does and failover is very quick -
>> and can be automatic. Also, you can do a database snapshot on the mirror
>> and use it for reporting.
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
>> SQL Server MVP
>> Toronto, ON Canada
>> https://mvp.support.microsoft.com/profile/Tom.Moreau
>>
>> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
>> news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
>> We just went through a catastrophic failure of an Oracle database. This
>> database was running on a server with a RAID 5 installed. One of the
disks
>> was having read/write errors. When we pulled the disk from the array to
>> have the hot spare kick in, the Oracle server failed and it took a lot of
>> effort to get it back. (I hate oracle).
>> We are moving to SQL Server 2005 and want to avoid this problem. I have
>> been asked to get the specs for a cluster. I have been reading up on the
>> clustering, but came across Transaction Log Shipping. I know that the
>> Transaction Log shipping does not have an automatic fail over, but it
will
>> move the transaction log to another server and provide another server we
>> can
>> use for reporting and in the event of a failure on the main server, we
>> could
>> use the other server until we can bring it up on line.
>> What in your opinion would be a better solution, cluster or transaction
>> log
>> shipping? All of our programs will be using this data, and it is
>> important
>> to recover from a disaster like this as quickly as possible. The cost of
>> clustering may be preventative, so I would like your input for those who
>> have used one or both methods for pros and cons. Thanks.
>> John
>>
>
>|||Way back 1.5 years ago I did the same thing for a client with thousands of
databases. Worked like a champ.
--
Kevin G. Boles
Indicium Resources, Inc.
SQL Server MVP
kgboles a earthlink dt net
"Kevin3NF" <kevin@.SPAMTRAP.3nf-inc.com> wrote in message
news:%23v1x6WfXIHA.4696@.TK2MSFTNGP05.phx.gbl...
> Waaaaaay back in 2000, we created our own jobs to logship and only had the
> restore command run every 8 hours. All the files were there, and we could
> catch the whole thing up in about 20 minutes if the primary went down, but
> it was a reporting server all day :)
> Same thing could be done here if Log Shipping was desired, as well as
> reporting. May not be the most glamorous solution, but it works...
> --
> Kevin3NF
> SQL Server dude
> You want fries with that?
> http://kevin3nf.blogspot.com/
> I only check the newsgroups during work hours, M-F.
> Hit my blog and the contact links if necessary...I may be available.
>
> "Tibor Karaszi" <tibor_please.no.email_karaszi@.hotmail.nomail.com> wrote
> in message news:0875841B-779E-4639-9C7F-31D19FCF060A@.microsoft.com...
>> Log shipping is fine
>> for HA, but you can't report off of the destination DB.
>> Well, you can if you do the restores using STANDBY. But you'd have to
>> kick out the users each time you want to do the next restore. Which means
>> that you really ... can't use a log shipped database for reporting ;-).
>> (Just wanted to mention this before anyone else say "You can". )
>>
>> --
>> Tibor Karaszi, SQL Server MVP
>> http://www.karaszi.com/sqlserver/default.asp
>> http://sqlblog.com/blogs/tibor_karaszi
>>
>> "Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
>> news:uy0aAMfXIHA.1188@.TK2MSFTNGP04.phx.gbl...
>> Yes, snapshots are supported only in Ent Ed. Also, Std Edition supports
>> 2-node clusters - not "two clusters". It depends on where you want to
>> put
>> your money and what your needs are. When it comes to High Availability
>> (HA), then you often have to go to Ent Ed. If you want clustering and
>> have
>> only 2 nodes, then you can get away with Std Ed, but clustering hardware
>> is
>> usually a bit more expensive and a little trickier to set up. Mirroring
>> will work on regular hardware and can go between dissimilar platforms.
>> Automatic failover, though is available in the Ent Ed.
>> When it comes to reporting, snapshots are your friend. Log shipping is
>> fine
>> for HA, but you can't report off of the destination DB.
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
>> SQL Server MVP
>> Toronto, ON Canada
>> https://mvp.support.microsoft.com/profile/Tom.Moreau
>>
>> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
>> news:%23In0b5eXIHA.4440@.TK2MSFTNGP06.phx.gbl...
>> Isn't SQL Server Enterprise edition the only edition that allows for
>> snapshots? We plan on using Standard Edition which will support two
>> clusters.
>>
>> "Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
>> news:ee6ZxueXIHA.5208@.TK2MSFTNGP04.phx.gbl...
>> Since you are using SQL Server 2005, consider database mirroring. You
>> don't
>> need the fancy hardware that clustering does and failover is very
>> quick -
>> and can be automatic. Also, you can do a database snapshot on the
>> mirror
>> and use it for reporting.
>> --
>> Tom
>> ----
>> Thomas A. Moreau, BSc, PhD, MCSE, MCDBA, MCITP, MCTS
>> SQL Server MVP
>> Toronto, ON Canada
>> https://mvp.support.microsoft.com/profile/Tom.Moreau
>>
>> "John Wright" <riley_wrightx@.hotmail.com> wrote in message
>> news:u6BTxreXIHA.3400@.TK2MSFTNGP03.phx.gbl...
>> We just went through a catastrophic failure of an Oracle database.
>> This
>> database was running on a server with a RAID 5 installed. One of the
>> disks
>> was having read/write errors. When we pulled the disk from the array
>> to
>> have the hot spare kick in, the Oracle server failed and it took a lot
>> of
>> effort to get it back. (I hate oracle).
>> We are moving to SQL Server 2005 and want to avoid this problem. I
>> have
>> been asked to get the specs for a cluster. I have been reading up on
>> the
>> clustering, but came across Transaction Log Shipping. I know that the
>> Transaction Log shipping does not have an automatic fail over, but it
>> will
>> move the transaction log to another server and provide another server
>> we
>> can
>> use for reporting and in the event of a failure on the main server, we
>> could
>> use the other server until we can bring it up on line.
>> What in your opinion would be a better solution, cluster or transaction
>> log
>> shipping? All of our programs will be using this data, and it is
>> important
>> to recover from a disaster like this as quickly as possible. The cost
>> of
>> clustering may be preventative, so I would like your input for those
>> who
>> have used one or both methods for pros and cons. Thanks.
>> John
>>
>>
>>
>|||Just a word of caution to help manage the expectation. If your storage goes
bad, you can for sure leverage certain DBMS features to alleviate the pain.
But there is no magic. SQL Server offers you many good features, but the same
potential storage problems won't magically go away. Due diligence is required
to plan for storage disasters.
Linchi
"John Wright" wrote:
> We just went through a catastrophic failure of an Oracle database. This
> database was running on a server with a RAID 5 installed. One of the disks
> was having read/write errors. When we pulled the disk from the array to
> have the hot spare kick in, the Oracle server failed and it took a lot of
> effort to get it back. (I hate oracle).
> We are moving to SQL Server 2005 and want to avoid this problem. I have
> been asked to get the specs for a cluster. I have been reading up on the
> clustering, but came across Transaction Log Shipping. I know that the
> Transaction Log shipping does not have an automatic fail over, but it will
> move the transaction log to another server and provide another server we can
> use for reporting and in the event of a failure on the main server, we could
> use the other server until we can bring it up on line.
> What in your opinion would be a better solution, cluster or transaction log
> shipping? All of our programs will be using this data, and it is important
> to recover from a disaster like this as quickly as possible. The cost of
> clustering may be preventative, so I would like your input for those who
> have used one or both methods for pros and cons. Thanks.
> John
>
>|||After much reading and study, I think we are going to implement the Database
Mirroring option with a witness to facilitate automatic fail-over.
Operating in High-Availability with full safety and a witness server should
do it for us. The servers are connected via a gigabit connection so latency
should not be a problem. Thanks one and all for the advice.
John
"Linchi Shea" <LinchiShea@.discussions.microsoft.com> wrote in message
news:C8570CAA-676A-4FE3-963D-21F4ECAC794B@.microsoft.com...
> Just a word of caution to help manage the expectation. If your storage
> goes
> bad, you can for sure leverage certain DBMS features to alleviate the
> pain.
> But there is no magic. SQL Server offers you many good features, but the
> same
> potential storage problems won't magically go away. Due diligence is
> required
> to plan for storage disasters.
> Linchi
> "John Wright" wrote:
>> We just went through a catastrophic failure of an Oracle database. This
>> database was running on a server with a RAID 5 installed. One of the
>> disks
>> was having read/write errors. When we pulled the disk from the array to
>> have the hot spare kick in, the Oracle server failed and it took a lot of
>> effort to get it back. (I hate oracle).
>> We are moving to SQL Server 2005 and want to avoid this problem. I have
>> been asked to get the specs for a cluster. I have been reading up on the
>> clustering, but came across Transaction Log Shipping. I know that the
>> Transaction Log shipping does not have an automatic fail over, but it
>> will
>> move the transaction log to another server and provide another server we
>> can
>> use for reporting and in the event of a failure on the main server, we
>> could
>> use the other server until we can bring it up on line.
>> What in your opinion would be a better solution, cluster or transaction
>> log
>> shipping? All of our programs will be using this data, and it is
>> important
>> to recover from a disaster like this as quickly as possible. The cost of
>> clustering may be preventative, so I would like your input for those who
>> have used one or both methods for pros and cons. Thanks.
>> John
>>

New server setup and RAID recommendation

I have a database that is around 2 to 4 GB.
If I were to estimate some numbers like 4x growth or
even 10x, the database size could reach 40GB.
The new server will be running SQL Server 2005.

I am not sure which configuration option to take.
I've gathered some information from different places:

Configuration #1:
OS - Raid1 2x36GB
Logs - Raid1 2x36GB
Data - Raid 5 4x73GB

Configuration #2:
OS - Raid1 2x36GB
Logs - Raid5 (not sure how many drives)
Data - Raid5 (not sure how many drives)

Now if I am using a separate RAID array disks
for the database's transaction log, should I also put the
TempDB in this RAID also?

Here's the configuration I am thinking of right now.
Please give me your comments:

OS - Raid1 2x36GB
Logs & TempDB - Raid5 3x36GB = 2x36GB usable space
Data - Raid5 3x73GB = 2x73GB usable space

If you have other configurations you recommend please let
me know.

Thank youFor the install and forget about I want protection and speed at a low
cost use RAID 5 for everything.

Now if you want the best performance and protect for the type of work
being done use the follow.

For the system files and executables, reading binaries doesn't really
put much overhead on the servers, so it is not a major concern. Any
RAID level could be used.

For data files, RAID 1 is preferable because it gives the best tradeoff
between performance, protection and cost. Because it is 100 percent
redundant, it both protects the data and the availability of your
system. Availability is the reason that you for using RAID with a
database, and even the loss of a small internal database can bring down
the database server. In this case, you will not lose any data but the
users will have to live without your system. If performance and
security are your major factors then RAID 10 should be used. No matter
what RAID level is used each RAID array should have as many physical
disks in the array as the controller will support. This allows reads
and writes to be performed simultaneously on each physical drive in the
array, significantly boosting disk I/O.

Log files, by their nature, are mostly written to, which means that
often RAID 1 is your best choice for performance. As with the
database files if performance and security are your major issues RAID
10 should be used.

New Installation Error

Hello all.
After suffering a complete RAID failure, I am now tasked with the
responsibility of setting up reporting services again, on a new and
different machine. I had been successful getting it running last time,
so I figured this wouldn't be too much problem. The installation goes
fine, until the end, when it reports that the service wont start for
the first time.
As for information. Server2003-sp1. For the ReportServer Windows
service I'm using account (NT AUTHORITY\NETWORK SERVICE). I've left
virtual directories the same, except that I've disabled SSL
connections. The Database is running locally, and for the runtime
connection credentials, I've chosen to use a Domain User Account (set
up especially for this purpose) "RSExec".
As I said, the install goes fine, but fails. The only thing I can
think of is that maybe my new RSExec user needs more permission than a
standard user. But I've read the books (Lachev, and Blackburn &
Vaughn), as well as the BOL, and I can't find any indication of what I
might be missing.
Opening the //machine/ReportServer page, I get the following error
message:
* The report server cannot open a connection to the report server
database. The logon failed. (rsReportServerDatabaseLogonFailed) Get
Online Help
o Logon failure: the user has not been granted the requested logon
type at this computer.
The error message from the log shows:
w3wp!ui!dc0!9/27/2005-08:25:37:: e ERROR:
System.Web.Services.Protocols.SoapException: The report server cannot
open a connection to the report server database. The logon failed. -->
Microsoft.ReportingServices.Diagnostics.Utilities.ReportServerDatabaseLogonFailedException:
The report server cannot open a connection to the report server
database. The logon failed. -->
System.Runtime.InteropServices.COMException (0x80070569): Logon
failure: the user has not been granted the requested logon type at this
computer.
at System.Runtime.InteropServices.Marshal.ThrowExceptionForHR(Int32
errorCode, IntPtr errorInfo)
at RSRemoteRpcClient.RemoteLogon.GetRemoteImpToken(String pUserName,
String pDomain, String pPassword, Boolean bTryRemote, IntPtr&
pImpToken)
at
Microsoft.ReportingServices.Library.ConnectionManager.GetImpersonationToken(String
userName, String domain, String userPwd)
-- End of inner exception stack trace --
Thanks for all your help. It is sincerely appreciated.
Brian AckermannDid you grant RSExec permissions to logon to SQL server? If so, can you
check to see if the user has permission to the RS database?
--
-Daniel
This posting is provided "AS IS" with no warranties, and confers no rights.
"Brian Ackermann" <brian.ackermann@.gmail.com> wrote in message
news:1127830923.220944.153010@.g44g2000cwa.googlegroups.com...
> Hello all.
> After suffering a complete RAID failure, I am now tasked with the
> responsibility of setting up reporting services again, on a new and
> different machine. I had been successful getting it running last time,
> so I figured this wouldn't be too much problem. The installation goes
> fine, until the end, when it reports that the service wont start for
> the first time.
> As for information. Server2003-sp1. For the ReportServer Windows
> service I'm using account (NT AUTHORITY\NETWORK SERVICE). I've left
> virtual directories the same, except that I've disabled SSL
> connections. The Database is running locally, and for the runtime
> connection credentials, I've chosen to use a Domain User Account (set
> up especially for this purpose) "RSExec".
> As I said, the install goes fine, but fails. The only thing I can
> think of is that maybe my new RSExec user needs more permission than a
> standard user. But I've read the books (Lachev, and Blackburn &
> Vaughn), as well as the BOL, and I can't find any indication of what I
> might be missing.
> Opening the //machine/ReportServer page, I get the following error
> message:
> * The report server cannot open a connection to the report server
> database. The logon failed. (rsReportServerDatabaseLogonFailed) Get
> Online Help
> o Logon failure: the user has not been granted the requested logon
> type at this computer.
> The error message from the log shows:
> w3wp!ui!dc0!9/27/2005-08:25:37:: e ERROR:
> System.Web.Services.Protocols.SoapException: The report server cannot
> open a connection to the report server database. The logon failed. -->
> Microsoft.ReportingServices.Diagnostics.Utilities.ReportServerDatabaseLogonFailedException:
> The report server cannot open a connection to the report server
> database. The logon failed. -->
> System.Runtime.InteropServices.COMException (0x80070569): Logon
> failure: the user has not been granted the requested logon type at this
> computer.
> at System.Runtime.InteropServices.Marshal.ThrowExceptionForHR(Int32
> errorCode, IntPtr errorInfo)
> at RSRemoteRpcClient.RemoteLogon.GetRemoteImpToken(String pUserName,
> String pDomain, String pPassword, Boolean bTryRemote, IntPtr&
> pImpToken)
> at
> Microsoft.ReportingServices.Library.ConnectionManager.GetImpersonationToken(String
> userName, String domain, String userPwd)
> -- End of inner exception stack trace --
> Thanks for all your help. It is sincerely appreciated.
> Brian Ackermann
>|||Daniel,
Thanks for the reply.
As far as I can tell, the installer gives the RSExec user permissions
on the database. I must be missing it if there was something else.
There was a user called RSExec intalled on the database, with what
appeared to be appropriate values.
Can you be more explicit as to what settings exactly need to be set.
I'm sure its quite simple, and that I'm just not seeing it.
Thanks
Brian Ackermann|||From the look of the call stack, it appears that you are using a domain
account to talk to the RS database server. It doesn't appear that user has
permission to logon to the RS box. You should verify if that is the case.
--
-Daniel
This posting is provided "AS IS" with no warranties, and confers no rights.
"Brian Ackermann" <brian.ackermann@.gmail.com> wrote in message
news:1127939219.565435.135110@.o13g2000cwo.googlegroups.com...
> Daniel,
> Thanks for the reply.
> As far as I can tell, the installer gives the RSExec user permissions
> on the database. I must be missing it if there was something else.
> There was a user called RSExec intalled on the database, with what
> appeared to be appropriate values.
> Can you be more explicit as to what settings exactly need to be set.
> I'm sure its quite simple, and that I'm just not seeing it.
> Thanks
> Brian Ackermann
>