Forum Discussion

bobwanamaker's avatar
15 years ago
Solved

.f files in the image database are compressed and need uncompressed

[root@pa1pbks4 1241000000]# ls -lt

total 48

drwx--x--x 2 root sys  4096 May  9  2010 catstore

-rw-r--r-- 1 root sys 35810 May 16  2009 VM_Instances_1241223583_FULL

-rw------- 1 root sys    36 May  9  2009 VM_Instances_1241223583_FULL.f.Z

drwxr-xr-x 3 root sys  4096 May  9  2009 tmp

[root@pa1pbks4 1241000000]# gunzip VM_Instances_1241223583_FULL.f.Z

[root@pa1pbks4 1241000000]# ls -lt

total 48

drwx--x--x 2 root sys  4096 May  9  2010 catstore

-rw-r--r-- 1 root sys 35810 May 16  2009 VM_Instances_1241223583_FULL

-rw------- 1 root sys    72 May  9  2009 VM_Instances_1241223583_FULL.f

drwxr-xr-x 3 root sys  4096 May  9  2009 tmp

[root@pa1pbks4 1241000000]# ls

catstore  tmp  VM_Instances_1241223583_FULL

[root@pa1pbks4 1241000000]# ll -R

.:

total 44

drwx--x--x 2 root sys  4096 Nov  9 10:06 catstore

drwxr-xr-x 3 root sys  4096 May  9  2009 tmp

-rw-r--r-- 1 root sys 35810 May 16  2009 VM_Instances_1241223583_FULL

 

./catstore:

total 16176

-rw------- 1 root sys    67011 May  9  2009 VM_Instances_1241223583_FULL.f_imgDir0.Z

-rw------- 1 root sys        3 May  9  2009 VM_Instances_1241223583_FULL.f_imgExtraObj0.Z

-rw------- 1 root sys  3798383 May  9  2009 VM_Instances_1241223583_FULL.f_imgFile0.Z

-rw------- 1 root sys       81 May  9  2009 VM_Instances_1241223583_FULL.f_imgHeader0.Z

-rw------- 1 root sys        3 May  9  2009 VM_Instances_1241223583_FULL.f_imgNDMP0.Z

-rw------- 1 root sys 11091660 May  9  2009 VM_Instances_1241223583_FULL.f_imgRecord0.Z

-rw------- 1 root sys  1552278 May  9  2009 VM_Instances_1241223583_FULL.f_imgStrings0.Z

-rw------- 1 root sys       19 May  9  2009 VM_Instances_1241223583_FULL.f_imgUserGroupNames0.Z

-rw------- 1 root sys      836 May  2  2009 VM_Instances_1241223583_FULL.f-list

 

./tmp:

total 4

drwxr-xr-x 2 root sys 4096 May  9  2009 catstore

 

./tmp/catstore:

total 0

[root@pa1pbks4 1241000000]#

  • From the Admin Guide:

    "...Control image-catalog compression by setting the Global Attributes property, Compress Catalog Interval. Use this property to specify how old the backup information must be before it is compressed. Specify the number of days to defer compression information, thus users who restore files from recent backups are unaffected. By default, Compress Catalog Interval is set to 0 and image compression is not enabled. For more information, see “Global Attributes properties” on page 435.
    Caution: Symantec discourages manually compressing or decompressing catalog backups using bpimage-[de]compress or any other method. Manually compressing or decompressing a catalog backup while any backup (regular or catalog) is running results in inconsistent image-catalog entries. When users list and restore files, the results could be incorrect...."

    The likelihood is that you've got compression set via the Global Attributes property.

    Also from the Admin Guide:

    "...Uncompressing the image catalog
    You may find it necessary to uncompress all records temporarily that are associated with an individual client. Uncompress the records if you anticipate large or numerous restore requests, for example.
    Perform the following steps as root on the master server:
    To uncompress the NetBackup catalog

    1 Verify that the partition where the image catalog resides has enough space
    to uncompress the client’s image records.
    2 Stop the request daemon, bprd, by running: /usr/openv/netbackup/bin/admincmd/bprdreq -terminate
    3 Make sure that bpdbm is running: /usr/openv/netbackup/bin/bpps
    4 Expand Host Properties > Master Servers. Open the properties of a host. On the Global Attributes properties, clear the Compress Catalog Interval check box. For more information, see “Global Attributes properties” on page 435.
    5 Set the Compress Catalog Interval Global Attributes property to 0.
    6 Change your working directory to /usr/openv/netbackup/bin and run the command:
    admincmd/bpimage -decompress -client name
    7 Restart the request daemon, bprd, by running: /usr/openv/netbackup/bin/initbprd
    8 Perform the file restorations from the client.
    9 Set the Compress Catalog After Global Attributes property to its previous value. The records that were uncompressed for this client are compressed after the next backup schedule...."

5 Replies

  • From the Admin Guide:

    "...Control image-catalog compression by setting the Global Attributes property, Compress Catalog Interval. Use this property to specify how old the backup information must be before it is compressed. Specify the number of days to defer compression information, thus users who restore files from recent backups are unaffected. By default, Compress Catalog Interval is set to 0 and image compression is not enabled. For more information, see “Global Attributes properties” on page 435.
    Caution: Symantec discourages manually compressing or decompressing catalog backups using bpimage-[de]compress or any other method. Manually compressing or decompressing a catalog backup while any backup (regular or catalog) is running results in inconsistent image-catalog entries. When users list and restore files, the results could be incorrect...."

    The likelihood is that you've got compression set via the Global Attributes property.

    Also from the Admin Guide:

    "...Uncompressing the image catalog
    You may find it necessary to uncompress all records temporarily that are associated with an individual client. Uncompress the records if you anticipate large or numerous restore requests, for example.
    Perform the following steps as root on the master server:
    To uncompress the NetBackup catalog

    1 Verify that the partition where the image catalog resides has enough space
    to uncompress the client’s image records.
    2 Stop the request daemon, bprd, by running: /usr/openv/netbackup/bin/admincmd/bprdreq -terminate
    3 Make sure that bpdbm is running: /usr/openv/netbackup/bin/bpps
    4 Expand Host Properties > Master Servers. Open the properties of a host. On the Global Attributes properties, clear the Compress Catalog Interval check box. For more information, see “Global Attributes properties” on page 435.
    5 Set the Compress Catalog Interval Global Attributes property to 0.
    6 Change your working directory to /usr/openv/netbackup/bin and run the command:
    admincmd/bpimage -decompress -client name
    7 Restart the request daemon, bprd, by running: /usr/openv/netbackup/bin/initbprd
    8 Perform the file restorations from the client.
    9 Set the Compress Catalog After Global Attributes property to its previous value. The records that were uncompressed for this client are compressed after the next backup schedule...."

  • The Compress Catalog Interval Global Attributes is set to zero.

    This problem appears to have started when we when from a unix master to a linux master.

    Whatever procedure was used must have compressed the catalog but did not uncompress it

    after completion. What is the netbackup way of unzipping a file?

  • The Netbackup procedure for uncompressing the image catalog does not work in this case,

  • What exactly is not working? Are you getting errors?

    Do you perhaps know how the catalogs were migrated when you moved to the new master server? If it was done via catalog backup and restore, NetBackup would have restored the image files in the same format that it was backed up on the old server. Any other method of manually compressing files before manually copying them across might have rendered your catalogs unusable.

  • Are all your entries like this (i.e. compressed) or just a few?

    What version of NB are you running? There are a few T/N's around dealing with catalog compression that may be relevant.

    Maybe worth running a NBCC to check for database inconsistencies (& a possible call to Tech Support).

    "The Netbackup procedure for uncompressing the image catalog does not work in this case" - any errors?