ufsdump - incremental file system dump
/usr/sbin/ufsdump [options] [arguments] files_to_dump
ufsdump backs up all files specified by files_to_dump (usually either a whole file system or files within a file system changed after a certain date) to magnetic tape or disk file.
The ufsdump command can only be used on unmounted file systems, or those mounted read-only. Attempting to dump a mounted, read-write file system might result in a system disruption or the inability to restore files from the dump. Consider using the fssnap(8) command to create a file system snapshot if you need a point-in-time image of a file system that is mounted.
options is a single string of one-letter ufsdump options.
arguments may be multiple strings whose association with the options is determined by order. That is, the first argument goes with the first option that takes an argument; the second argument goes with the second option that takes an argument, and so on.
files_to_dump is required and must be the last argument on the command line. See OPERANDS for more information.
With most devices ufsdump can automatically detect the end-of-media. Consequently, the d, s, and t options are not necessary for multi-volume dumps, unless ufsdump does not understand the way the device detects the end-of-media, or the files are to be restored on a system with an older version of the restore command.
The following options are supported:
The “dump level.” All files specified by files_to_dump that have been modified since the last ufsdump at a lower dump level are copied to the dump_file destination (normally a magnetic tape device). For instance, if a “level 2” dump was done on Monday, followed by a “level 4” dump on Tuesday, a subsequent “level 3” dump on Wednesday would contain all files modified or added since the “level 2” (Monday) backup. A “level 0” dump copies the entire file system to the dump_file.
Archive file. Archive a dump table-of-contents in the specified archive_file to be used by ufsrestore(8) to determine whether a file is in the dump file that is being restored.
Blocking factor. Specify the blocking factor for tape writes. The default is 20 blocks per write for tapes of density less than 6250BPI (bytes-per-inch). The default blocking factor for tapes of density 6250BPI and greater is 64. The default blocking factor for cartridge tapes (c option) is 126. The highest blocking factor available with most tape drives is 126. Note: the blocking factor is specified in terms of 512-byte blocks, for compatibility with tar(1).
Cartridge. Set the defaults for cartridge instead of the standard half-inch reel. This sets the density to 1000BPI and the blocking factor to 126. Since ufsdump can automatically detect the end-of-media, only the blocking parameter normally has an effect. When cartridge tapes are used, and this option is not specified, ufsdump will slightly miscompute the size of the tape. If the b, d, s or t options are specified with this option, their values will override the defaults set by this option.
Tape density. Not normally required, as ufsdump can detect end-of-media. This parameter can be used to keep a running tab on the amount of tape used per reel. The default density is 6250BPI except when the c option is used for cartridge tape, in which case it is assumed to be 1000BPI per track. Typical values for tape devices are:
1000 BPI The tape densities and other options are documented in the st(4D) man page.
Diskette. Obsolete option.
Dump file. Use dump_file as the file to dump to, instead of /dev/rmt/0. If dump_file is specified as −, dump to standard output.
If the name of the file is of the form machine:device, the dump is done from the specified machine over the network using rmt(8). Since ufsdump is normally run by root, the name of the local machine must appear in the /.rhosts file of the remote machine. If the file is specified as user@ machine:device, ufsdump will attempt to execute as the specified user on the remote machine. The specified user must have a .rhosts file on the remote machine that allows the user invoking the command from the local machine to access the remote machine.
Autoload. When the end-of-tape is reached before the dump is complete, take the drive offline and wait up to two minutes for the tape drive to be ready again. This gives autoloading (stackloader) tape drives a chance to load a new tape. If the drive is ready within two minutes, continue. If it is not, prompt for another tape and wait.
Sets the tape label to string, instead of the default none. string may be no more than sixteen characters long. If it is longer, it is truncated and a warning printed; the dump will still be done. The tape label is specific to the ufsdump tape format, and bears no resemblance to IBM or ANSI-standard tape labels.
Notify all operators in the sys group that ufsdump requires attention by sending messages to their terminals, in a manner similar to that used by the wall(8) command. Otherwise, such messages are sent only to the terminals (such as the console) on which the user running ufsdump is logged in.
Use device_name when recording information in /etc/dumpdates (see the u option) and when comparing against information in /etc/dumpdates for incremental dumps. The device_name provided can contain no white space as defined in scanf(3C) and is case-sensitive.
Offline. Take the drive offline when the dump is complete or the end-of-media is reached and rewind the tape. In the case of some autoloading 8mm drives, the tape is removed from the drive automatically. This prevents another process which rushes in to use the drive, from inadvertently overwriting the media.
Specify the size of the volume being dumped to. Not normally required, as ufsdump can detect end-of-media. When the specified size is reached, ufsdump waits for you to change the volume. ufsdump interprets the specified size as the length in feet for tapes and cartridges. The values should be a little smaller than the actual physical size of the media (for example, 425 for a 450-foot cartridge). Typical values for tape devices depend on the c option for cartridge devices:
Size estimate. Determine the amount of space that is needed to perform the dump without actually doing it, and display the estimated number of bytes it will take. This is useful with incremental dumps to determine how many volumes of media will be needed.
Specify the number of tracks for a cartridge tape. Not normally required, as ufsdump can detect end-of-media. The default is 9 tracks. The t option is not compatible with the D option. Values for Oracle-supported tape devices are:
Sets the amount of time to wait for an autoload command to complete. This option is ignored unless the l option has also been specified. The default time period to wait is two minutes. Specify time units with a trailing h ( for hours), m (for minutes), or s (for seconds). The default unit is minutes.
Update the dump record. Add an entry to the file /etc/dumpdates, for each file system successfully dumped that includes the file system name (or device_name as specified with the N option), date, and dump level.
Verify. After each tape is written, verify the contents of the media against the source file system. If any discrepancies occur, prompt for new media, then repeat the dump/verification process. The file system must be unmounted. This option cannot be used to verify a dump to standard output.
Warning. List the file systems that have not been backed up within a day. This information is gleaned from the files /etc/dumpdates and /etc/vfstab. When the w option is used, all other options are ignored. After reporting, ufsdump exits immediately.
Warning with highlight. Similar to the w option, except that the W option includes all file systems that appear in /etc/dumpdates, along with information about their most recent dump dates and levels. File systems that have not been backed up within a day are highlighted.
The following operand is supported:
Specifies the files to dump. Usually it identifies a whole file system by its raw device name (for example, /dev/rdsk/c0t3d0s6). Incremental dumps (levels 1 to 9) of files changed after a certain date only apply to a whole file system. Alternatively, files_to_dump can identify individual files or directories. All named directories that may be examined by the user running ufsdump, as well as any explicitly-named files, are dumped. This dump is equivalent to a level 0 dump of the indicated portions of the filesystem, except that /etc/dumpdates is not updated even if the –u option has been specified. In all cases, the files must be contained in the same file system, and the file system must be local to the system where ufsdump is being run.
files_to_dump is required and must be the last argument on the command line.
If no options are given, the default is 9uf /dev/rmt/0 files_to_dump.
The following command makes a full dump of a root file system on c0t3d0, on a 150-MByte cartridge tape unit 0:
example# ufsdump 0cfu /dev/rmt/0 /dev/rdsk/c0t3d0s0
The following command makes and verifies an incremental dump at level 5 of the usr partition of c0t3d0, on a 1/2 inch reel tape unit 1,:
example# ufsdump 5fuv /dev/rmt/1 /dev/rdsk/c0t3d0s6
While running, ufsdump emits many verbose messages. ufsdump returns the following exit values:
Startup errors encountered.
Abort − no checkpoint attempted.
default unit to dump to
dump date record
to find group sys
to gain access to remote system with drive
list of file systems
See attributes(7) for descriptions of the following attributes:
Fewer than 32 read errors on the file system are ignored.
Because each reel requires a new process, parent processes for reels that are already written hang around until the entire tape is written.
ufsdump requires operator intervention on these conditions: end of volume, end of dump, volume write error, volume open error or disk read error (if there are more than a threshold of 32). In addition to alerting all operators implied by the n option, ufsdump interacts with the operator on ufsdump's control terminal at times when ufsdump can no longer proceed, or if something is grossly wrong. All questions ufsdump poses must be answered by typing yes or no, as appropriate.
Since backing up a disk can involve a lot of time and effort, ufsdump checkpoints at the start of each volume. If writing that volume fails for some reason, ufsdump will, with operator permission, restart itself from the checkpoint after a defective volume has been replaced.
It is vital to perform full, “level 0”, dumps at regular intervals. When performing a full dump, bring the machine down to single-user mode using shutdown(8). While preparing for a full dump, it is a good idea to clean the tape drive and heads. Incremental dumps should be performed with the system running in single-user mode.
Incremental dumps allow for convenient backup and recovery of active files on a more frequent basis, with a minimum of media and time. However, there are some tradeoffs. First, the interval between backups should be kept to a minimum (once a day at least). To guard against data loss as a result of a media failure (a rare, but possible occurrence), capture active files on (at least) two sets of dump volumes. Another consideration is the desire to keep unnecessary duplication of files to a minimum to save both operator time and media storage. A third consideration is the ease with which a particular backed-up version of a file can be located and restored. The following four-week schedule offers a reasonable tradeoff between these goals.
Sun Mon Tue Wed Thu Fri Week 1: Full 5 5 5 5 3 Week 2: 5 5 5 5 3 Week 3: 5 5 5 5 3 Week 4: 5 5 5 5 3
Although the Tuesday through Friday incrementals contain “extra copies” of files from Monday, this scheme assures that any file modified during the week can be recovered from the previous day's incremental dump.
ufsdump uses multiple processes to allow it to read from the disk and write to the media concurrently. Due to the way it synchronizes between these processes, any attempt to run dump with a nice (process priority) of ‘−5’ or better will likely make ufsdump run slower instead of faster.
Most disks contain one or more overlapping slices because slice 2 covers the entire disk. The other slices are of various sizes and usually do not overlap. For example, a common configuration places root on slice 0, swap on slice 1, /opt on slice 5 and /usr on slice 6.
It should be emphasized that ufsdump dumps one ufs file system at a time. Given the above scenario where slice 0 and slice 2 have the same starting offset, executing ufsdump on slice 2 with the intent of dumping the entire disk would instead dump only the root file system on slice 0. To dump the entire disk, the user must dump the file systems on each slice separately.
The /etc/vfstab file does not allow the desired frequency of backup for file systems to be specified (as /etc/fstab did). Consequently, the w and W options assume file systems should be backed up daily, which limits the usefulness of these options.