storeBackupCheckBackup.pl

If you copy data from one disk to another, you might have unrecognized bit failures (e.g. in main memory, in the cpu, on the transfer lines) resulting in wrong data stored in your backup. This may happen especially over long time periods on the backup disk itself. Because storeBackup also stores check sums in addition to the real data it is possible to compare the stored data in the backup with the check sums generated from the source files to give you a better probability that your backup data is not broken.¹⁷ But don't panic, this doesn't happen all the time.

storeBackupCheckBackup.pl Verifies the consistency of one or more backups by using the md5 sums generated during the backup in .md5CheckSums and compares them with just calculated ones. It also checks all files in the backup and the files considered in .md5CheckSums for existence or non-existence.

	storeBackupCheckBackup.pl -c backupDir [-p number] [-i]
	      [-w filePrefix] [--lastOfEachSeries]
	      [--includeRenamedBackups] [-T tmpdir]
	      [--logFile
	       [--plusLogStdout] [--suppressTime] [-m maxFilelen]
	       [[-n noOfOldFiles] | [--saveLogs]]

--print

Print the configuration parameters and stop processing

--checkDir / -c

The repository/backup, where you want backups to search. You can set this option to your whole backup repository, to a series or to a single backup.

--wrongFileTables / -w

Write file names with detected faults in line based regular files for later (not automated) bug fixing. The parameter for this option is a file prefix, e.g. if the file prefix is /tmp/bugsB- the following files are generated:
/tmp/bugsB-files.missing.txt
/tmp/bugsB-md5sums.missing.txt
/tmp/bugsB-md5sums.wrong.txt

--lastOfEachSeries

Only check the last backup of each series found.

--verbose / -v

Verbose. Print some extra messages so you see what's happening.

--parJobs / -p

Maximum number of parallel search operations. The default value is chosen automatically as the number of cores plus 1.

--includeRenamedBackups / -i

By using this option, renamed backups are checked also If you have renamed backups specified with option --checkDir. Renamed backups have to follow the scheme backupDir$-$something, e.g., 2012.01.30_15.21.03$-$renamed. Also see section 7.3.

--tmpdir / -T /tmpdir

Directory for temporary files, the default value is picked from the environment variable $TMPDIR. If it does not exist, /tmp is set as the default value.

--logFile / -l

Name of the log file. Default is stdout.

--plusLogStdout / plusLogStdout

If option logFile is set, you can here configure storeBackup.pl to print the log messages to stdout.

--suppressTime

Suppress the output of the actual time in the log file.

--maxFilelen / -m

Maximal size of a log file. After reaching this size, the log file will be rotated (see option noOfOldFiles) or compressed (see option saveLogs).

--noOfOldFiles / -n

Number of old rotated log files, default is 5. With default values, it will look like this:

$ ls -l /tmp/storebackup.log*
-rw------- 1 hjc  root  328815 30. Aug 12:12 /tmp/storebackup.log
-rw------- 1 root root 1000087 27. Aug 21:18 /tmp/storebackup.log.1
-rw------- 1 root root 1000038 20. Aug 19:02 /tmp/storebackup.log.2
-rw------- 1 root root 1000094 11. Aug 18:51 /tmp/storebackup.log.3
-rw------- 1 root root 1000147 11. Aug 18:49 /tmp/storebackup.log.4
-rw------- 1 root root 1000030 11. Aug 18:49 /tmp/storebackup.log.5

Older log files than *.5 have been deleted automatically.

--saveLogs

Save the log files with a time and date stamp instead of deleting them after rotating. (Setting this option overwrites the default value of option noOfOldFiles.)

--compressWith

Specifies the program to compress the saved log files (e.g., with gzip -9). Default value is bzip2.
On the command line, the parameter to this option is parsed like a line in the configuration file and normally has to be quoted.