All backups failing

Having worked fine for 4 years, all our SQL backups are failing:
[Microsoft SQL-DMO (ODBC SQLState: 42000)] Error 3201: [Microsoft][ODBC SQL
Server Driver][SQL Server]Cannot open backup device '\\.\Tape0'. Device
error or device off-line. See the SQL Server error log for more details.
BackupTapeFile::OpenMedia: Backup device '\\.\Tape0' failed to open.
Operating system error = 32(The process cannot access the file because it
is being used by another process.).
The first db in the maintenance plan claims to back up successfully in 1
second (it used to take 10s), but its verify and all subsequent backups
fail.
I've rebooted and tried with a new tape to no avail. Using filemon from
www.sysinternals.com shows no other process using the tape drive. Backing up
to disk fails for about half of the dbs: the OS error is 2(The system cannot
find the file specified.). Backing up individual dbs 'by hand' from
Enterprise Manager gives the same OS error 32. If I use isql, I can seem to
backup OK.
The day before backups started failing, there was an (apparently
unsuccessful) attack on our server via IIS and MDAC, so I installed the
latest MDAC and JET drivers. After that I reinstalled NT4 SP6a, the SRP and
subsequent security patches. I also changed some permissions, as part of
adding a new user; I don't think I touched anything to do with SQLserver,
and I've later increased the permissions of the SQLServer agent to no
effect.
The system is SBS 4.5, with latest service packs for all components. We have
four dbs, two <5MB and two <100MB, all used very lightly (mostly 0-2 users,
say 5 transactions an hour, almost always no use during backup).
I'd really need to find out what's going on here: could MDAC and/or JET
break the backup (since the maintenance plan apparently uses ODBC drivers)?
What other cause might there be for this apparently erroneous OS error = 32?
What services need to be running for backup to work (other than MSSQLServer
and SQLServerAgent; I currently have MSDTC and MSSearch too, but don't know
that I need them)?
I've been pulling my hair out for a week now, so many thanks for any help!
SteveSteven,
A common cause for the device off-line etc is that there is another process
accessing the backup device files. Normally this is when the backups are
going to disk and another process (BackupExec etc) is then copying them to
tape. The backups are done using the account that sql server runs under.
Has anyone changed that account of it's permissions? I would try deleting
the backup devices and starting from scratch with one db so you can
systematically rule out potential issues.
--
Andrew J. Kelly
SQL Server MVP
"Steven Kelly" <stevek@.metacase.com> wrote in message
news:Y7xnb.499$Xn1.494@.read3.inet.fi...
> Having worked fine for 4 years, all our SQL backups are failing:
> [Microsoft SQL-DMO (ODBC SQLState: 42000)] Error 3201: [Microsoft][ODBC
SQL
> Server Driver][SQL Server]Cannot open backup device '\\.\Tape0'. Device
> error or device off-line. See the SQL Server error log for more details.
> BackupTapeFile::OpenMedia: Backup device '\\.\Tape0' failed to open.
> Operating system error = 32(The process cannot access the file because it
> is being used by another process.).
> The first db in the maintenance plan claims to back up successfully in 1
> second (it used to take 10s), but its verify and all subsequent backups
> fail.
> I've rebooted and tried with a new tape to no avail. Using filemon from
> www.sysinternals.com shows no other process using the tape drive. Backing
up
> to disk fails for about half of the dbs: the OS error is 2(The system
cannot
> find the file specified.). Backing up individual dbs 'by hand' from
> Enterprise Manager gives the same OS error 32. If I use isql, I can seem
to
> backup OK.
> The day before backups started failing, there was an (apparently
> unsuccessful) attack on our server via IIS and MDAC, so I installed the
> latest MDAC and JET drivers. After that I reinstalled NT4 SP6a, the SRP
and
> subsequent security patches. I also changed some permissions, as part of
> adding a new user; I don't think I touched anything to do with SQLserver,
> and I've later increased the permissions of the SQLServer agent to no
> effect.
> The system is SBS 4.5, with latest service packs for all components. We
have
> four dbs, two <5MB and two <100MB, all used very lightly (mostly 0-2
users,
> say 5 transactions an hour, almost always no use during backup).
> I'd really need to find out what's going on here: could MDAC and/or JET
> break the backup (since the maintenance plan apparently uses ODBC
drivers)?
> What other cause might there be for this apparently erroneous OS error =32?
> What services need to be running for backup to work (other than
MSSQLServer
> and SQLServerAgent; I currently have MSDTC and MSSearch too, but don't
know
> that I need them)?
> I've been pulling my hair out for a week now, so many thanks for any help!
> Steve
>|||Thanks for the reply, Andrew,
There is no other process accessing the files or tape drive. SQLServer and
Agent are running under the LocalSystem account. Following your advice, I
created new disk backup devices, one for each db, and backed up each db by
hand from Enterprise Manager to them. I noticed that the smallest dbs backed
up and verified OK, but the large ones (60MB) came back with an error in a
few seconds. Looking at the file on the disk for that db, it was either
non-existent or <10MB. If I backed up by hand WITHOUT verify, it came back
with 'backup completed OK' in a few seconds, but clearly from the disk noise
and file size the backup was actually still being written to disk. If I left
it until the disk was quiet, the full file has been written, and indeed that
backup can be restored fine (fc /b shows its not binary identical to the
original db, even though no transactions occurred in the meantime, but I
suppose that's normal)
Thus my problem is that Enterprise Manager backups think they have finished
long before they actually have. The next step (normally verify) in the
maintenance plan fires, and finds that the file/tape is still being accessed
by 'another' process - this same backup process! Backups with isql seem not
to have this problem.
So, what course of action would people recommend? I guess somehow the JET
and MDAC messed up something, maybe reinstalling SQLServer SP4 would put
that right?
Thanks,
Steve
"Andrew J. Kelly" <sqlmvpnooospam@.shadhawk.com> wrote in message
news:O6PfVAanDHA.2000@.TK2MSFTNGP12.phx.gbl...
> Steven,
> A common cause for the device off-line etc is that there is another
process
> accessing the backup device files. Normally this is when the backups are
> going to disk and another process (BackupExec etc) is then copying them to
> tape. The backups are done using the account that sql server runs under.
> Has anyone changed that account of it's permissions? I would try deleting
> the backup devices and starting from scratch with one db so you can
> systematically rule out potential issues.
> --
> Andrew J. Kelly
> SQL Server MVP
>
> "Steven Kelly" <stevek@.metacase.com> wrote in message
> news:Y7xnb.499$Xn1.494@.read3.inet.fi...
> > Having worked fine for 4 years, all our SQL backups are failing:
> >
> > [Microsoft SQL-DMO (ODBC SQLState: 42000)] Error 3201: [Microsoft][ODBC
> SQL
> > Server Driver][SQL Server]Cannot open backup device '\\.\Tape0'. Device
> > error or device off-line. See the SQL Server error log for more details.
> >
> > BackupTapeFile::OpenMedia: Backup device '\\.\Tape0' failed to open.
> > Operating system error = 32(The process cannot access the file because
it
> > is being used by another process.).
> >
> > The first db in the maintenance plan claims to back up successfully in 1
> > second (it used to take 10s), but its verify and all subsequent backups
> > fail.
> >
> > I've rebooted and tried with a new tape to no avail. Using filemon from
> > www.sysinternals.com shows no other process using the tape drive.
Backing
> up
> > to disk fails for about half of the dbs: the OS error is 2(The system
> cannot
> > find the file specified.). Backing up individual dbs 'by hand' from
> > Enterprise Manager gives the same OS error 32. If I use isql, I can seem
> to
> > backup OK.
> >
> > The day before backups started failing, there was an (apparently
> > unsuccessful) attack on our server via IIS and MDAC, so I installed the
> > latest MDAC and JET drivers. After that I reinstalled NT4 SP6a, the SRP
> and
> > subsequent security patches. I also changed some permissions, as part of
> > adding a new user; I don't think I touched anything to do with
SQLserver,
> > and I've later increased the permissions of the SQLServer agent to no
> > effect.
> >
> > The system is SBS 4.5, with latest service packs for all components. We
> have
> > four dbs, two <5MB and two <100MB, all used very lightly (mostly 0-2
> users,
> > say 5 transactions an hour, almost always no use during backup).
> >
> > I'd really need to find out what's going on here: could MDAC and/or JET
> > break the backup (since the maintenance plan apparently uses ODBC
> drivers)?
> > What other cause might there be for this apparently erroneous OS error => 32?
> > What services need to be running for backup to work (other than
> MSSQLServer
> > and SQLServerAgent; I currently have MSDTC and MSSearch too, but don't
> know
> > that I need them)?
> >
> > I've been pulling my hair out for a week now, so many thanks for any
help!
> > Steve
> >
> >
>|||I have never seem a case where EM could not backup and verify correctly. It
sounds like a hardware or driver issue to me. Is the hardware on the HCL?
Do you have any 3rd party caching software being used on that machine? What
Edition and Version of the OS and SQL Server are you using and do you have
the latest service packs? It's odd and that is a new one on me as this is
usually a pretty simple process.
--
Andrew J. Kelly
SQL Server MVP
"Steven Kelly" <stevek@.metacase.com> wrote in message
news:AH3ob.55$gJ3.7@.read3.inet.fi...
> Thanks for the reply, Andrew,
> There is no other process accessing the files or tape drive. SQLServer and
> Agent are running under the LocalSystem account. Following your advice, I
> created new disk backup devices, one for each db, and backed up each db by
> hand from Enterprise Manager to them. I noticed that the smallest dbs
backed
> up and verified OK, but the large ones (60MB) came back with an error in a
> few seconds. Looking at the file on the disk for that db, it was either
> non-existent or <10MB. If I backed up by hand WITHOUT verify, it came back
> with 'backup completed OK' in a few seconds, but clearly from the disk
noise
> and file size the backup was actually still being written to disk. If I
left
> it until the disk was quiet, the full file has been written, and indeed
that
> backup can be restored fine (fc /b shows its not binary identical to the
> original db, even though no transactions occurred in the meantime, but I
> suppose that's normal)
> Thus my problem is that Enterprise Manager backups think they have
finished
> long before they actually have. The next step (normally verify) in the
> maintenance plan fires, and finds that the file/tape is still being
accessed
> by 'another' process - this same backup process! Backups with isql seem
not
> to have this problem.
> So, what course of action would people recommend? I guess somehow the JET
> and MDAC messed up something, maybe reinstalling SQLServer SP4 would put
> that right?
> Thanks,
> Steve
> "Andrew J. Kelly" <sqlmvpnooospam@.shadhawk.com> wrote in message
> news:O6PfVAanDHA.2000@.TK2MSFTNGP12.phx.gbl...
> > Steven,
> >
> > A common cause for the device off-line etc is that there is another
> process
> > accessing the backup device files. Normally this is when the backups
are
> > going to disk and another process (BackupExec etc) is then copying them
to
> > tape. The backups are done using the account that sql server runs
under.
> > Has anyone changed that account of it's permissions? I would try
deleting
> > the backup devices and starting from scratch with one db so you can
> > systematically rule out potential issues.
> >
> > --
> >
> > Andrew J. Kelly
> > SQL Server MVP
> >
> >
> > "Steven Kelly" <stevek@.metacase.com> wrote in message
> > news:Y7xnb.499$Xn1.494@.read3.inet.fi...
> > > Having worked fine for 4 years, all our SQL backups are failing:
> > >
> > > [Microsoft SQL-DMO (ODBC SQLState: 42000)] Error 3201:
[Microsoft][ODBC
> > SQL
> > > Server Driver][SQL Server]Cannot open backup device '\\.\Tape0'.
Device
> > > error or device off-line. See the SQL Server error log for more
details.
> > >
> > > BackupTapeFile::OpenMedia: Backup device '\\.\Tape0' failed to open.
> > > Operating system error = 32(The process cannot access the file because
> it
> > > is being used by another process.).
> > >
> > > The first db in the maintenance plan claims to back up successfully in
1
> > > second (it used to take 10s), but its verify and all subsequent
backups
> > > fail.
> > >
> > > I've rebooted and tried with a new tape to no avail. Using filemon
from
> > > www.sysinternals.com shows no other process using the tape drive.
> Backing
> > up
> > > to disk fails for about half of the dbs: the OS error is 2(The system
> > cannot
> > > find the file specified.). Backing up individual dbs 'by hand' from
> > > Enterprise Manager gives the same OS error 32. If I use isql, I can
seem
> > to
> > > backup OK.
> > >
> > > The day before backups started failing, there was an (apparently
> > > unsuccessful) attack on our server via IIS and MDAC, so I installed
the
> > > latest MDAC and JET drivers. After that I reinstalled NT4 SP6a, the
SRP
> > and
> > > subsequent security patches. I also changed some permissions, as part
of
> > > adding a new user; I don't think I touched anything to do with
> SQLserver,
> > > and I've later increased the permissions of the SQLServer agent to no
> > > effect.
> > >
> > > The system is SBS 4.5, with latest service packs for all components.
We
> > have
> > > four dbs, two <5MB and two <100MB, all used very lightly (mostly 0-2
> > users,
> > > say 5 transactions an hour, almost always no use during backup).
> > >
> > > I'd really need to find out what's going on here: could MDAC and/or
JET
> > > break the backup (since the maintenance plan apparently uses ODBC
> > drivers)?
> > > What other cause might there be for this apparently erroneous OS error
=> > 32?
> > > What services need to be running for backup to work (other than
> > MSSQLServer
> > > and SQLServerAgent; I currently have MSDTC and MSSearch too, but don't
> > know
> > > that I need them)?
> > >
> > > I've been pulling my hair out for a week now, so many thanks for any
> help!
> > > Steve
> > >
> > >
> >
> >
>|||The hardware, driver, OS and SQL server versions have not
changed at all, and the backups worked fine for 4 years
(over 1 year at current SP levels). No 3rd party caching
software. Non-SQL backups (NTBACKUP) work fine for files
and Exchange.
OS: SBS 4.5 (i.e. NT 4), SP6a + SRP + all security patches
SQL: 7.0 SP4
Since backing up with isql waits until the backup is
really done, could the problem be in the presumably extra
layers of database libraries through which EM runs its
commands (SQL-DMO & ODBC)? Could the ODBC library have
been changed by updating JET and MDAC, or does SQLServer
have its own copy?
Cheers,
Steve
>--Original Message--
>I have never seem a case where EM could not backup and
verify correctly. It
>sounds like a hardware or driver issue to me. Is the
hardware on the HCL?
>Do you have any 3rd party caching software being used on
that machine? What
>Edition and Version of the OS and SQL Server are you
using and do you have
>the latest service packs? It's odd and that is a new one
on me as this is
>usually a pretty simple process.
>--
>Andrew J. Kelly
>SQL Server MVP
>
>"Steven Kelly" <stevek@.metacase.com> wrote in message
>news:AH3ob.55$gJ3.7@.read3.inet.fi...
>> Thanks for the reply, Andrew,
>> There is no other process accessing the files or tape
drive. SQLServer and
>> Agent are running under the LocalSystem account.
Following your advice, I
>> created new disk backup devices, one for each db, and
backed up each db by
>> hand from Enterprise Manager to them. I noticed that
the smallest dbs
>backed
>> up and verified OK, but the large ones (60MB) came back
with an error in a
>> few seconds. Looking at the file on the disk for that
db, it was either
>> non-existent or <10MB. If I backed up by hand WITHOUT
verify, it came back
>> with 'backup completed OK' in a few seconds, but
clearly from the disk
>noise
>> and file size the backup was actually still being
written to disk. If I
>left
>> it until the disk was quiet, the full file has been
written, and indeed
>that
>> backup can be restored fine (fc /b shows its not binary
identical to the
>> original db, even though no transactions occurred in
the meantime, but I
>> suppose that's normal)
>> Thus my problem is that Enterprise Manager backups
think they have
>finished
>> long before they actually have. The next step (normally
verify) in the
>> maintenance plan fires, and finds that the file/tape is
still being
>accessed
>> by 'another' process - this same backup process!
Backups with isql seem
>not
>> to have this problem.
>> So, what course of action would people recommend? I
guess somehow the JET
>> and MDAC messed up something, maybe reinstalling
SQLServer SP4 would put
>> that right?
>> Thanks,
>> Steve
>> "Andrew J. Kelly" <sqlmvpnooospam@.shadhawk.com> wrote
in message
>> news:O6PfVAanDHA.2000@.TK2MSFTNGP12.phx.gbl...
>> > Steven,
>> >
>> > A common cause for the device off-line etc is that
there is another
>> process
>> > accessing the backup device files. Normally this is
when the backups
>are
>> > going to disk and another process (BackupExec etc) is
then copying them
>to
>> > tape. The backups are done using the account that
sql server runs
>under.
>> > Has anyone changed that account of it's permissions?
I would try
>deleting
>> > the backup devices and starting from scratch with one
db so you can
>> > systematically rule out potential issues.
>> >
>> > --
>> >
>> > Andrew J. Kelly
>> > SQL Server MVP
>> >
>> >
>> > "Steven Kelly" <stevek@.metacase.com> wrote in message
>> > news:Y7xnb.499$Xn1.494@.read3.inet.fi...
>> > > Having worked fine for 4 years, all our SQL backups
are failing:
>> > >
>> > > [Microsoft SQL-DMO (ODBC SQLState: 42000)] Error
3201:
>[Microsoft][ODBC
>> > SQL
>> > > Server Driver][SQL Server]Cannot open backup
device '\\.\Tape0'.
>Device
>> > > error or device off-line. See the SQL Server error
log for more
>details.
>> > >
>> > > BackupTapeFile::OpenMedia: Backup
device '\\.\Tape0' failed to open.
>> > > Operating system error = 32(The process cannot
access the file because
>> it
>> > > is being used by another process.).
>> > >
>> > > The first db in the maintenance plan claims to back
up successfully in
>1
>> > > second (it used to take 10s), but its verify and
all subsequent
>backups
>> > > fail.
>> > >
>> > > I've rebooted and tried with a new tape to no
avail. Using filemon
>from
>> > > www.sysinternals.com shows no other process using
the tape drive.
>> Backing
>> > up
>> > > to disk fails for about half of the dbs: the OS
error is 2(The system
>> > cannot
>> > > find the file specified.). Backing up individual
dbs 'by hand' from
>> > > Enterprise Manager gives the same OS error 32. If I
use isql, I can
>seem
>> > to
>> > > backup OK.
>> > >
>> > > The day before backups started failing, there was
an (apparently
>> > > unsuccessful) attack on our server via IIS and
MDAC, so I installed
>the
>> > > latest MDAC and JET drivers. After that I
reinstalled NT4 SP6a, the
>SRP
>> > and
>> > > subsequent security patches. I also changed some
permissions, as part
>of
>> > > adding a new user; I don't think I touched anything
to do with
>> SQLserver,
>> > > and I've later increased the permissions of the
SQLServer agent to no
>> > > effect.
>> > >
>> > > The system is SBS 4.5, with latest service packs
for all components.
>We
>> > have
>> > > four dbs, two <5MB and two <100MB, all used very
lightly (mostly 0-2
>> > users,
>> > > say 5 transactions an hour, almost always no use
during backup).
>> > >
>> > > I'd really need to find out what's going on here:
could MDAC and/or
>JET
>> > > break the backup (since the maintenance plan
apparently uses ODBC
>> > drivers)?
>> > > What other cause might there be for this apparently
erroneous OS error
>=>> > 32?
>> > > What services need to be running for backup to work
(other than
>> > MSSQLServer
>> > > and SQLServerAgent; I currently have MSDTC and
MSSearch too, but don't
>> > know
>> > > that I need them)?
>> > >
>> > > I've been pulling my hair out for a week now, so
many thanks for any
>> help!
>> > > Steve
>> > >
>> > >
>> >
>> >
>>
>
>.
>|||I guess there could be a problem with DMO but this would be the first I have
heard of this issue and backups are done a lot thru EM all the time. I
don't know what else to have you try. If no one else pipes in and it is
still an issue you can try giving MS PSS a call.
Andrew J. Kelly
SQL Server MVP
"Steven Kelly" <stevek@.metacase.com> wrote in message
news:034401c39f20$d294c1c0$a301280a@.phx.gbl...
> The hardware, driver, OS and SQL server versions have not
> changed at all, and the backups worked fine for 4 years
> (over 1 year at current SP levels). No 3rd party caching
> software. Non-SQL backups (NTBACKUP) work fine for files
> and Exchange.
> OS: SBS 4.5 (i.e. NT 4), SP6a + SRP + all security patches
> SQL: 7.0 SP4
> Since backing up with isql waits until the backup is
> really done, could the problem be in the presumably extra
> layers of database libraries through which EM runs its
> commands (SQL-DMO & ODBC)? Could the ODBC library have
> been changed by updating JET and MDAC, or does SQLServer
> have its own copy?
> Cheers,
> Steve
> >--Original Message--
> >I have never seem a case where EM could not backup and
> verify correctly. It
> >sounds like a hardware or driver issue to me. Is the
> hardware on the HCL?
> >Do you have any 3rd party caching software being used on
> that machine? What
> >Edition and Version of the OS and SQL Server are you
> using and do you have
> >the latest service packs? It's odd and that is a new one
> on me as this is
> >usually a pretty simple process.
> >
> >--
> >
> >Andrew J. Kelly
> >SQL Server MVP
> >
> >
> >"Steven Kelly" <stevek@.metacase.com> wrote in message
> >news:AH3ob.55$gJ3.7@.read3.inet.fi...
> >> Thanks for the reply, Andrew,
> >>
> >> There is no other process accessing the files or tape
> drive. SQLServer and
> >> Agent are running under the LocalSystem account.
> Following your advice, I
> >> created new disk backup devices, one for each db, and
> backed up each db by
> >> hand from Enterprise Manager to them. I noticed that
> the smallest dbs
> >backed
> >> up and verified OK, but the large ones (60MB) came back
> with an error in a
> >> few seconds. Looking at the file on the disk for that
> db, it was either
> >> non-existent or <10MB. If I backed up by hand WITHOUT
> verify, it came back
> >> with 'backup completed OK' in a few seconds, but
> clearly from the disk
> >noise
> >> and file size the backup was actually still being
> written to disk. If I
> >left
> >> it until the disk was quiet, the full file has been
> written, and indeed
> >that
> >> backup can be restored fine (fc /b shows its not binary
> identical to the
> >> original db, even though no transactions occurred in
> the meantime, but I
> >> suppose that's normal)
> >>
> >> Thus my problem is that Enterprise Manager backups
> think they have
> >finished
> >> long before they actually have. The next step (normally
> verify) in the
> >> maintenance plan fires, and finds that the file/tape is
> still being
> >accessed
> >> by 'another' process - this same backup process!
> Backups with isql seem
> >not
> >> to have this problem.
> >>
> >> So, what course of action would people recommend? I
> guess somehow the JET
> >> and MDAC messed up something, maybe reinstalling
> SQLServer SP4 would put
> >> that right?
> >>
> >> Thanks,
> >> Steve
> >>
> >> "Andrew J. Kelly" <sqlmvpnooospam@.shadhawk.com> wrote
> in message
> >> news:O6PfVAanDHA.2000@.TK2MSFTNGP12.phx.gbl...
> >> > Steven,
> >> >
> >> > A common cause for the device off-line etc is that
> there is another
> >> process
> >> > accessing the backup device files. Normally this is
> when the backups
> >are
> >> > going to disk and another process (BackupExec etc) is
> then copying them
> >to
> >> > tape. The backups are done using the account that
> sql server runs
> >under.
> >> > Has anyone changed that account of it's permissions?
> I would try
> >deleting
> >> > the backup devices and starting from scratch with one
> db so you can
> >> > systematically rule out potential issues.
> >> >
> >> > --
> >> >
> >> > Andrew J. Kelly
> >> > SQL Server MVP
> >> >
> >> >
> >> > "Steven Kelly" <stevek@.metacase.com> wrote in message
> >> > news:Y7xnb.499$Xn1.494@.read3.inet.fi...
> >> > > Having worked fine for 4 years, all our SQL backups
> are failing:
> >> > >
> >> > > [Microsoft SQL-DMO (ODBC SQLState: 42000)] Error
> 3201:
> >[Microsoft][ODBC
> >> > SQL
> >> > > Server Driver][SQL Server]Cannot open backup
> device '\\.\Tape0'.
> >Device
> >> > > error or device off-line. See the SQL Server error
> log for more
> >details.
> >> > >
> >> > > BackupTapeFile::OpenMedia: Backup
> device '\\.\Tape0' failed to open.
> >> > > Operating system error = 32(The process cannot
> access the file because
> >> it
> >> > > is being used by another process.).
> >> > >
> >> > > The first db in the maintenance plan claims to back
> up successfully in
> >1
> >> > > second (it used to take 10s), but its verify and
> all subsequent
> >backups
> >> > > fail.
> >> > >
> >> > > I've rebooted and tried with a new tape to no
> avail. Using filemon
> >from
> >> > > www.sysinternals.com shows no other process using
> the tape drive.
> >> Backing
> >> > up
> >> > > to disk fails for about half of the dbs: the OS
> error is 2(The system
> >> > cannot
> >> > > find the file specified.). Backing up individual
> dbs 'by hand' from
> >> > > Enterprise Manager gives the same OS error 32. If I
> use isql, I can
> >seem
> >> > to
> >> > > backup OK.
> >> > >
> >> > > The day before backups started failing, there was
> an (apparently
> >> > > unsuccessful) attack on our server via IIS and
> MDAC, so I installed
> >the
> >> > > latest MDAC and JET drivers. After that I
> reinstalled NT4 SP6a, the
> >SRP
> >> > and
> >> > > subsequent security patches. I also changed some
> permissions, as part
> >of
> >> > > adding a new user; I don't think I touched anything
> to do with
> >> SQLserver,
> >> > > and I've later increased the permissions of the
> SQLServer agent to no
> >> > > effect.
> >> > >
> >> > > The system is SBS 4.5, with latest service packs
> for all components.
> >We
> >> > have
> >> > > four dbs, two <5MB and two <100MB, all used very
> lightly (mostly 0-2
> >> > users,
> >> > > say 5 transactions an hour, almost always no use
> during backup).
> >> > >
> >> > > I'd really need to find out what's going on here:
> could MDAC and/or
> >JET
> >> > > break the backup (since the maintenance plan
> apparently uses ODBC
> >> > drivers)?
> >> > > What other cause might there be for this apparently
> erroneous OS error
> >=> >> > 32?
> >> > > What services need to be running for backup to work
> (other than
> >> > MSSQLServer
> >> > > and SQLServerAgent; I currently have MSDTC and
> MSSearch too, but don't
> >> > know
> >> > > that I need them)?
> >> > >
> >> > > I've been pulling my hair out for a week now, so
> many thanks for any
> >> help!
> >> > > Steve
> >> > >
> >> > >
> >> >
> >> >
> >>
> >>
> >
> >
> >.
> >