Content-type: text/html Man page of SAVEFS


Section: Maintenance Commands (8)
Updated: Dec 11, 01m
Index Return to Main Contents


savefs - save filesystem to a NetWorker server  


[ options ] filesystem
-p [ options ] [ filesystem ... ]
[ -BEFnpqRv ] [ -s server ] [ -N name ] [ -g group ] [ -c client ] [ -l level | -C schedule ] [ -e expiration ] [ -w browse ] [ -y retention ] [ -f filename ] [ -o save_operations ] [ -W width ] [ -t date ]


The savefs command saves a filesystem (using save(1m)) to a NetWorker server. Mount points are not crossed, and symbolic links are not followed. NOTE: running savefs directly is not recommended; use savegrp(1m) instead.

A level-based system (similar to dump(1m)) is used to save only those files which have been modified since some previous save (a partial save).

The nsr_schedule(5) for the local NetWorker client is examined to determine the proper level of save for the current date.

The set of files saved depends on when, and at what level, previous saves have been performed, in addition to the effects of the default directives (see nsr_directive(5)), and the various directive files (see nsr(5)) which are encountered while processing the filesystem.  


The savefs command may also be used to probe a client for its filesystems and recent save times. When probing, savefs does not save data, but instead produces a machine-parsable report describing the layout of the client's filesystems. When used with the -p probe option, the local NetWorker client's nsr_client(5) resources are examined, and the filesystems listed in the save set attribute are probed (if no filesystems are listed on the command line). If the save set list consists of the keyword All, then the /etc/fstab file (/etc/vfstab on Solaris, /etc/mnttab on SCO, and a kernel table on AIX) are examined to determine which filesystems should be saved, making sure to save only local, mounted filesystems.

Note that metadevices within the Sun Solaris Online DiskSuite and Logical Volumes within the HP-UX Logical Volume Manager are treated like independent disks. This approach allows each to be saved in its own session, assuming sufficient parallelism.

Care should be taken when the NSR client resource explicitly lists the save sets, for two primary reasons. First, this list must be manually updated when new filesystems are added which need saving. Second, since savefs only stops at the end of a path or a mount point, if you list two save sets in the same filesystem, and one is a subdirectory of the other, the subdirectory will be saved twice.

Filesystem arguments can be specified to limit the filesystem saves to only those specified, but the specified filesystems must appear on some Save Set list for this client (see the -F option).

Probes are also useful when testing how NetWorker will behave in a clustered environment. In this setup ownership of shared filesystems must be determined, and performing a probe with the verbose option set allows one to examine the default ownership rules. Refer to pathownerignore(5) for a description of path-ownership rules.  


Force save of all connecting directory information from root (``/'') down to the point of invocation. This option is used by savegrp(1m), for example, when saving the server's bootstrap information.
-c client
The name of the client whose filesystem needs to be saved. This option is especially needed in a cluster environment where a physical host can represent its own hostname as well as hostnames of any virtual (also known as "logical") hosts that exist in this physical host. Without this option, the hostname of the physical host is assumed by default. This option is required if a filesystem that belongs to any of the virtual hosts needs to be saved.
-C schedule
The name of the schedule (see nsr_schedule(5)) to use when automatically determining the save level. If this option is not specified, savefs uses the schedule named by the NSR client resource for the specified filesystem.
-e expiration
Set the date (in nsr_getdate(3) format) when the saved data will expire. When a save set has an explicit expiration date, the save set remains both browsable and non-recyclable until it expires. After it expires and it has passed its browse time, its state will become non-browsable. If it has expired and it has passed its retention time, the save set will become recyclable. The special value forever is used to indicate that a volume that never expires (i.e. an archive volume) must be used. By default, no explicit expiration date is used.
-w browse
Sets the date (in nsr_getdate(3) format) after which this save set will no longer be browsable. By default, the server determines the browse date for the save set based on the browse policies in effect. This option allows overriding the existing policies on a save by save basis.
-y retention
Sets the date (in nsr_getdate(3) format) when the saved data will become recyclable. By default, the server determines this date for the save set based on the retention policies in effect.
Estimate. Before saving any data, browse the filesystem trees to be saved and accurately estimate the amount of data that will be generated. Without this flag, the estimate size is zero. This flag consumes an amount of time proportional to the number of files in each filesystem. This is because the entire directory is browsed before any saving begins and browsed again when actually saving the directory, but the file data is only read from the disk the last time. In many cases, the overhead for using this flag is small and is well-justified.
-f filename
The file from which application specific modules (or ASMs) should take their directives (see nsr(5)). By default, these are taken from the NSR directive resource named by the directive attribute in the NSR client resource for each client (see nsr_directive(5)).
Force. Save every argument like a filesystem, even if it is not listed in fstab(5) or nsr_client(5).
-g group
Restrict the scope of the client to a particular group. If this option is not specified, save sets from all instances of the NSR client resource for this client will be used, regardless of the group. This value is also passed on to save(1m), which uses it to select a specific media pool.
-l level
The level of save to perform. There are 12 levels: full, levels 1 through 9, incr, and skip. Full specifies that all files are to be saved. It is analogous to a level 0 dump in dump(1m). Incr specifies incremental saves in which only those files that have been modified since the most recent save, at any level, are saved. This level has no exact analogue in dump(1m) since the last save at any level, including previous incremental saves, are considered when determining what to save. Skip causes no files to be saved. The levels 1 through 9 cause all files to be saved which have been modified since any lower level save was performed. As an example, if you did a full save on Monday, followed by a level 3 save on Tuesday, a subsequent level 3 save on Wednesday would contain all files modified or added since the Monday full save. By default, the save level is determined automatically from the NetWorker client's schedule (see nsr_schedule(5)). By using the history of previous saves maintained by nsrmmd(1m) on the NetWorker server, the needed time for the given level can correctly be computed. By using media information on the server, times computed for saves that are based on previous save levels will automatically be adjusted as required when tapes are deleted.
No save. Accurately estimates the amount of data that would be generated (as described for -E, but doesn't save any data.
-N name
The symbolic name this set of saves is to be known by. By default, the first filesystem argument is used as the name.
List the name of the filesystems, the level of save that would be performed, and the time since which files must have been modified to be saved, but don't actually do the save. This information is gleaned from the /etc/fstab file (or another operating system specific file, as described above) and the nsr_schedule(5).
Quiet. Display only summary information and error messages.
Really quiet. Display only error messages.
Cause savefs to report on its success or failure, by echoing a simple "succeeded" or "failed" message. This is used by savegrp(1m) when it is running savefs.
-s server
Specifies which machine to use as the NetWorker server. See nsr(1m) for the algorithm NetWorker uses to choose a server when none is specified.
-t date
The date (in nsr_getdate(3) format) from which to base schedule level calculations. If not specified, the current time is used.
-o save_operations
Save Operations of the form KEYWORD:TOKEN=STATE. It is used to configure VSS saves on Windows 2003. Examples:

"vss:*=off"                          Turn off VSS.

"vss:Microsoft Exchange Writer=off"  Disable a writer.

"vss:C:=off"                         Disable VSS for a drive.

Please see the Admin Guide for more details.
Verbose. Causes lots of debugging style output. This option is also used by savegrp(1m) when it is probing for the capabilities of the client's savefs, for supporting multiple versions.
-W width
The width used when formatting output or notification messages. By default, this is 80.


NSR client
These resources specify the client's save sets, default schedule, and directives to use when saving them.
NSR directive
A resource of this type is named by the directive attribute in each NSR client resource. These are the directives used for the save sets specified in the associated NSR client resource.
NSR schedule
A resource of this type is named by the schedule attribute in each NSR client resource. This is the schedule used for the save sets specified in the associated NSR client resource.


If All is specified in the save set attribute for a NSR client resource, then the list of local filesystems is taken from this file.
Solaris only. The same as /etc/fstab on other operating systems.
SCO only. The same as /etc/fstab on other operating systems.


nsr_getdate(3), fstab(5), mnttab(F) (SCO only), vfstab(5) (Solaris only), nsr(5), nsr_service(5), nsr_schedule(5), dump(1m), nsr(1m), nsrd(1m), nsrindexd(1m), nsrmmd(1m), recover(1m), save(1m), savegrp(1m), pathownerignore(5).  



Exit Codes

Normal exit.
Abnormal exit.



Exit Codes

This document was created by man2html, using the manual pages.
Time: 02:37:08 GMT, October 02, 2010