network backups fail - error 64

I've found a lot of people with this same problem, but not really any
solutions. We have scheduled backup jobs that run & backup DBs over
the network to a UNC path on our SAN/NAS. A few nights ago, most of
the larger database backups started failing pretty regularly with the
following errors present in the logs.
BackupDiskFile::RequestDurableMedia: failure on backup device
'\\UNCpath\msdb_dump.BAK'. Operating system error 64(The specified
network name is no longer available.)
BACKUP failed to complete the command BACKUP DATABASE msdb TO DISK = '\\UNCPath\msdb_dump.BAK', WITH INIT, Format
BackupMedium::ReportIoError: write failure on backup device
'\\UNCpath\DB_dump.BAK'. Operating system error 64(The specified
network name is no longer available.).
Internal I/O request 0x26485108: Op: Write, pBuffer: 0x28690000, Size:
983040, Position: 3191806464, UMS: Internal: 0x0, InternalHigh:
0xF0000, Offset: 0xBE3F1A00, OffsetHigh: 0x0, m_buf: 0x28690000, m_len:
983040, m_actualBytes: 0, m_errcode: 64, BackupFile:
\\UNCPath\DB_dump.BAK
I don't know of a good solution here - can anyone help? I don't have
access to the Event Logs on the systems, but I'm trying to get to them,
and I'll repost if I find anything.Hi,
The OS error 64 usually indicates the intermittent network connectivity
problems.
To ensure that the backups for the database are taken properly, I recommend
you backup to
the local hard disk drive and then create a OS scheduled job to move the
backup file
to the share upon completion of the SQL Backup.
In parallel Inform your network infrastructure team to investigate this
issu. As well as run the below command from command prompt during backup to
check the network issue.
ping <ipaddress_of_remote_machine> -t >pingtest.txt
Once the problem occur verify the pingtest.txt for any Request time Out.
Thanks
Hari
SQL Server MVP
<unc27932@.yahoo.com> wrote in message
news:1121174032.571685.322230@.o13g2000cwo.googlegroups.com...
> I've found a lot of people with this same problem, but not really any
> solutions. We have scheduled backup jobs that run & backup DBs over
> the network to a UNC path on our SAN/NAS. A few nights ago, most of
> the larger database backups started failing pretty regularly with the
> following errors present in the logs.
> BackupDiskFile::RequestDurableMedia: failure on backup device
> '\\UNCpath\msdb_dump.BAK'. Operating system error 64(The specified
> network name is no longer available.)
> BACKUP failed to complete the command BACKUP DATABASE msdb TO DISK => '\\UNCPath\msdb_dump.BAK', WITH INIT, Format
> BackupMedium::ReportIoError: write failure on backup device
> '\\UNCpath\DB_dump.BAK'. Operating system error 64(The specified
> network name is no longer available.).
> Internal I/O request 0x26485108: Op: Write, pBuffer: 0x28690000, Size:
> 983040, Position: 3191806464, UMS: Internal: 0x0, InternalHigh:
> 0xF0000, Offset: 0xBE3F1A00, OffsetHigh: 0x0, m_buf: 0x28690000, m_len:
> 983040, m_actualBytes: 0, m_errcode: 64, BackupFile:
> \\UNCPath\DB_dump.BAK
> I don't know of a good solution here - can anyone help? I don't have
> access to the Event Logs on the systems, but I'm trying to get to them,
> and I'll repost if I find anything.
>|||How large is your DB?
Do you have any problem when you copy large files, more than 1GB, to the
SAN/NAS?
<unc27932@.yahoo.com> wrote in message
news:1121174032.571685.322230@.o13g2000cwo.googlegroups.com...
> I've found a lot of people with this same problem, but not really any
> solutions. We have scheduled backup jobs that run & backup DBs over
> the network to a UNC path on our SAN/NAS. A few nights ago, most of
> the larger database backups started failing pretty regularly with the
> following errors present in the logs.
> BackupDiskFile::RequestDurableMedia: failure on backup device
> '\\UNCpath\msdb_dump.BAK'. Operating system error 64(The specified
> network name is no longer available.)
> BACKUP failed to complete the command BACKUP DATABASE msdb TO DISK => '\\UNCPath\msdb_dump.BAK', WITH INIT, Format
> BackupMedium::ReportIoError: write failure on backup device
> '\\UNCpath\DB_dump.BAK'. Operating system error 64(The specified
> network name is no longer available.).
> Internal I/O request 0x26485108: Op: Write, pBuffer: 0x28690000, Size:
> 983040, Position: 3191806464, UMS: Internal: 0x0, InternalHigh:
> 0xF0000, Offset: 0xBE3F1A00, OffsetHigh: 0x0, m_buf: 0x28690000, m_len:
> 983040, m_actualBytes: 0, m_errcode: 64, BackupFile:
> \\UNCPath\DB_dump.BAK
> I don't know of a good solution here - can anyone help? I don't have
> access to the Event Logs on the systems, but I'm trying to get to them,
> and I'll repost if I find anything.
>|||The failing DBs range from 100mb to 8-10gb. I can copy from my client
to the SAN via a mapped network drive to the same share that the DB is
pointing to, just in windows, no problem (it takes forever for the 8gb
file, but it works).|||Another thing - the backup appears to run OK - using STATS I can view
it progressing but right before it's done it blows up...
A good one says this:
96 percent backed up.
97 percent backed up.
98 percent backed up.
99 percent backed up.
Processed 37680 pages for database 'DB', file 'DB_dat' on file 1.
100 percent backed up.
Processed 1 pages for database 'DB', file 'DB_log' on file 1.
BACKUP DATABASE successfully processed 37681 pages in 22.736 seconds
(13.576 MB/sec).
My bad one says this:
96 percent backed up.
97 percent backed up.
98 percent backed up.
99 percent backed up.
Processed 37680 pages for database 'DB', file 'DB_dat' on file 1.
100 percent backed up.
Processed 1 pages for database 'DB', file 'DB_log' on file 1.
<Then it says something like error 3013, BACKUP DATABASE is terminating
abnormally>sql