Copyright 2006 by Theodore Ts'o. All Rights Reserved.
This file may be copied under the terms of the GNU Public License.
The
e2fsck.conf file uses an INI-style format. Stanzas, or top-level sections, are
delimited by square braces: [ ]. Within each section, each line
defines a relation, which assigns tags to values, or to a subsection,
which contains further relations or subsections.
Tags can be assigned multiple values
An example of the INI-style format used by this configuration file
follows below:
[section1]
tag1 = value_a
tag1 = value_b
tag2 = value_c
[section 2]
tag3 = {
subtag1 = subtag_value_a
subtag1 = subtag_value_b
subtag2 = subtag_value_c
}
tag1 = value_d
tag2 = value_e
}
Comments are delimited by a semicolon (';') or a hash ('#') character at the beginning of the comment, and are terminated by the end of line character.
Tags and values must be quoted using double quotes if they contain spaces. Within a quoted string, the standard backslash interpretations apply: "\en" (for the newline character), "\et" (for the tab character), "\eb" (for the backspace character), and "\e\e" (for the backslash character).
The following stanzas are used in the e2fsck.conf file. They will be described in more detail in future sections of this document.
[options] This stanza contains general configuration parameters for e2fsck 's behavior.
[defaults] Contains relations which define the default parameters used by e2fsck (8). In general, these defaults may be overridden by command-line options provided by the user.
[problems] This stanza allows the administrator to reconfigure how e2fsck handles various file system inconsistencies. @TDB_MAN_COMMENT@.TP @TDB_MAN_COMMENT@.I [scratch_files] @TDB_MAN_COMMENT@This stanza controls when e2fsck will attempt to use @TDB_MAN_COMMENT@scratch files to reduce the need for memory.
allow_cancellation If this relation is set to a boolean value of true, then if the user interrupts e2fsck using ^C, and the file system is not explicitly flagged as containing errors, e2fsck will exit with an exit status of 0 instead of 32. This setting defaults to false.
accept_time_fudge Unfortunately, due to Windows' unfortunate design decision to configure the hardware clock to tick localtime, instead of the more proper and less error-prone UTC time, many users end up in the situation where the system clock is incorrectly set at the time when e2fsck is run.
Historically this was usually due to some distributions having buggy init scripts and/or installers that didn't correctly detect this case and take appropriate countermeasures. Unfortunately, this is occasionally true even today, usually due to a buggy or misconfigured virtualization manager or the installer not having access to a network time server during the installation process. So by default, we allow the superblock times to be fudged by up to 24 hours. This can be disabled by setting accept_time_fudge to the boolean value of false. This setting defaults to true.broken_system_clock The e2fsck (8) program has some heuristics that assume that the system clock is correct. In addition, many system programs make similar assumptions. For example, the UUID library depends on time not going backwards in order for it to be able to make its guarantees about issuing universally unique ID's. Systems with broken system clocks, are well, broken. However, broken system clocks, particularly in embedded systems, do exist. E2fsck will attempt to use heuristics to determine if the time can not be trusted; and to skip time-based checks if this is true. If this boolean is set to true, then e2fsck will always assume that the system clock can not be trusted.
buggy_init_scripts This boolean relation is an alias for accept_time_fudge for backwards compatibility; it used to be that the behavior defined by accept_time_fudge above defaulted to false, and buggy_init_scripts would enable superblock time field to be wrong by up to 24 hours. When we changed the default, we also renamed this boolean relation to accept_time_fudge.
clear_test_fs_flag This boolean relation controls whether or not e2fsck (8) will offer to clear the test_fs flag if the ext4 file system is available on the system. It defaults to true.
defer_check_on_battery This boolean relation controls whether or not the interval between file system checks (either based on time or number of mounts) should be doubled if the system is running on battery. This setting defaults to true.
indexed_dir_slack_percentage When e2fsck (8) repacks a indexed directory, reserve the specified percentage of empty space in each leaf nodes so that a few new entries can be added to the directory without splitting leaf nodes, so that the average fill ratio of directories can be maintained at a higher, more efficient level. This relation defaults to 20 percent.
inode_count_fullmap If this boolean relation is true, trade off using memory for speed when checking a file system with a large number of hard-linked files. The amount of memory required is proportional to the number of inodes in the file system. For large file systems, this can be gigabytes of memory. (For example a 40TB file system with 2.8 billion inodes will consume an additional 5.7 GB memory if this optimization is enabled.) This setting defaults to false.
log_dir If the log_filename or problem_log_filename relations contains a relative pathname, then the log file will be placed in the directory named by the log_dir relation.
log_dir_fallback This relation contains an alternate directory that will be used if the directory specified by log_dir is not available or is not writable.
log_dir_wait If this boolean relation is true, them if the directories specified by log_dir or log_dir_fallback are not available or are not yet writable, e2fsck will save the output in a memory buffer, and a child process will periodically test to see if the log directory has become available after the boot sequence has mounted the requested file system for reading/writing. This implements the functionality provided by logsave (8) for e2fsck log files.
log_filename This relation specifies the file name where a copy of e2fsck's output will be written. If certain problem reports are suppressed using the max_count_problems relation, (or on a per-problem basis using the max_count relation), the full set of problem reports will be written to the log file. The filename may contain various percent-expressions (%D, %T, %N, etc.) which will be expanded so that the file name for the log file can include things like date, time, device name, and other run-time parameters. See the LOGGING section for more details.
max_count_problems This relation specifies the maximum number of problem reports of a particular type will be printed to stdout before further problem reports of that type are squelched. This can be useful if the console is slow (i.e., connected to a serial port) and so a large amount of output could end up delaying the boot process for a long time (potentially hours).
no_optimize_extents If this boolean relation is true, do not offer to optimize the extent tree by reducing the tree's width or depth. This setting defaults to false.
problem_log_filename This relation specifies the file name where a log of problem codes found by e2fsck be written. The filename may contain various percent-expressions (%D, %T, %N, etc.) which will be expanded so that the file name for the log file can include things like date, time, device name, and other run-time parameters. See the LOGGING section for more details.
readahead_mem_pct Use this percentage of memory to try to read in metadata blocks ahead of the main e2fsck thread. This should reduce run times, depending on the speed of the underlying storage and the amount of free memory. There is no default, but see readahead_kb for more details.
readahead_kb Use this amount of memory to read in metadata blocks ahead of the main checking thread. Setting this value to zero disables readahead entirely. By default, this is set the size of two block groups' inode tables (typically 4MiB on a regular ext4 file system); if this amount is more than 1/50th of total physical memory, readahead is disabled.
report_features If this boolean relation is true, e2fsck will print the file system features as part of its verbose reporting (i.e., if the -v option is specified)
report_time If this boolean relation is true, e2fsck will run as if the options -tt are always specified. This will cause e2fsck to print timing statistics on a pass by pass basis for full file system checks.
report_verbose If this boolean relation is true, e2fsck will run as if the option -v is always specified. This will cause e2fsck to print some additional information at the end of each full file system check.
undo_dir This relation specifies the directory where the undo file should be stored. It can be overridden via the E2FSPROGS_UNDO_DIR environment variable. If the directory location is set to the value none , e2fsck will not create an undo file.
Note that inappropriate settings in this stanza may cause e2fsck to behave incorrectly, or even crash. Most system administrators should not be making changes to this section without referring to source code.
Within each problem code's subsection, the following tags may be used:
description This relation allows the message which is printed when this file system inconsistency is detected to be overridden.
preen_ok This boolean relation overrides the default behavior controlling whether this file system problem should be automatically fixed when e2fsck is running in preen mode.
max_count This integer relation overrides the max_count_problems parameter (set in the options section) for this particular problem.
no_ok This boolean relation overrides the default behavior determining whether or not the file system will be marked as inconsistent if the user declines to fix the reported problem.
no_default This boolean relation overrides whether the default answer for this problem (or question) should be "no".
preen_nomessage This boolean relation overrides the default behavior controlling whether or not the description for this file system problem should be suppressed when e2fsck is running in preen mode.
no_nomsg This boolean relation overrides the default behavior controlling whether or not the description for this file system problem should be suppressed when a problem forced not to be fixed, either because e2fsck is run with the -n option or because the force_no flag has been set for the problem.
force_no This boolean option, if set to true, forces a problem to never be fixed. That is, it will be as if the user problem responds 'no' to the question of 'should this problem be fixed?'. The force_no option even overrides the -y option given on the command-line (just for the specific problem, of course).
not_a_fix This boolean option, it set to true, marks the problem as one where if the user gives permission to make the requested change, it does not mean that the file system had a problem which has since been fixed. This is used for requests to optimize the file system's data structure, such as pruning an extent tree. @TDB_MAN_COMMENT@.SH THE [scratch_files] STANZA @TDB_MAN_COMMENT@The following relations are defined in the @TDB_MAN_COMMENT@.I [scratch_files] @TDB_MAN_COMMENT@stanza. @TDB_MAN_COMMENT@.TP @TDB_MAN_COMMENT@.I directory @TDB_MAN_COMMENT@If the directory named by this relation exists and is @TDB_MAN_COMMENT@writeable, then e2fsck will attempt to use this @TDB_MAN_COMMENT@directory to store scratch files instead of using @TDB_MAN_COMMENT@in-memory data structures. @TDB_MAN_COMMENT@.TP @TDB_MAN_COMMENT@.I numdirs_threshold @TDB_MAN_COMMENT@If this relation is set, then in-memory data structures @TDB_MAN_COMMENT@will be used if the number of directories in the file system @TDB_MAN_COMMENT@are fewer than amount specified. @TDB_MAN_COMMENT@.TP @TDB_MAN_COMMENT@.I dirinfo @TDB_MAN_COMMENT@This relation controls whether or not the scratch file @TDB_MAN_COMMENT@directory is used instead of an in-memory data @TDB_MAN_COMMENT@structure for directory information. It defaults to @TDB_MAN_COMMENT@true. @TDB_MAN_COMMENT@.TP @TDB_MAN_COMMENT@.I icount @TDB_MAN_COMMENT@This relation controls whether or not the scratch file @TDB_MAN_COMMENT@directory is used instead of an in-memory data @TDB_MAN_COMMENT@structure when tracking inode counts. It defaults to @TDB_MAN_COMMENT@true.
The filename in log_filename may contain the following percent-expressions that will be expanded as follows.
%d The current day of the month
%D The current date; this is a equivalent of %Y%m%d
%h The hostname of the system.
%H The current hour in 24-hour format (00..23)
%m The current month as a two-digit number (01..12)
%M The current minute (00..59)
%N The name of the block device containing the file system, with any directory pathname stripped off.
%p The pid of the e2fsck process
%s The current time expressed as the number of seconds since 1970-01-01 00:00:00 UTC
%S The current second (00..59)
%T The current time; this is equivalent of %H%M%S
%u The name of the user running e2fsck.
%U This percent expression does not expand to anything, but it signals that any following date or time expressions should be expressed in UTC time instead of the local timezone.
%y The last two digits of the current year (00..99)
%Y The current year (i.e., 2012).
[problems]
0x040002 = {
preen_ok = true
description = "@u @i %i. "
}
The following recipe will cause an e2fsck logfile to be written to the directory /var/log/e2fsck, with a filename that contains the device name, the hostname of the system, the date, and time: e.g., "e2fsck-sda3.server.INFO.20120314-112142". If the directory containing /var/log is located on the root file system which is initially mounted read-only, then the output will be saved in memory and written out once the root file system has been remounted read/write. To avoid too much detail from being written to the serial console (which could potentially slow down the boot sequence), only print no more than 16 instances of each type of file system corruption.
[options]
max_count_problems = 16
log_filename = e2fsck-%N.%h.INFO.%D-%T
log_dir_wait = true
/etc/e2fsck.conf The configuration file for e2fsck (8).