{ "mode": "man", "parameter": "BTRFS-BALANCE", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/BTRFS-BALANCE/8/json", "generated": "2026-09-16T04:27:08Z", "synopsis": "btrfs balance ", "sections": { "NAME": { "content": "btrfs-balance - balance block groups on a btrfs filesystem\n", "subsections": [] }, "SYNOPSIS": { "content": "btrfs balance \n", "subsections": [] }, "DESCRIPTION": { "content": "The primary purpose of the balance feature is to spread block groups across all devices so\nthey match constraints defined by the respective profiles. See mkfs.btrfs(8) section PROFILES\nfor more details. The scope of the balancing process can be further tuned by use of filters\nthat can select the block groups to process. Balance works only on a mounted filesystem. Ex‐\ntent sharing is preserved and reflinks are not broken. Files are not defragmented nor recom‐\npressed, file extents are preserved but the physical location on devices will change.\n\nThe balance operation is cancellable by the user. The on-disk state of the filesystem is al‐\nways consistent so an unexpected interruption (e.g. system crash, reboot) does not corrupt\nthe filesystem. The progress of the balance operation is temporarily stored as an internal\nstate and will be resumed upon mount, unless the mount option skipbalance is specified.\n", "subsections": [ { "name": "WARNING:", "content": "Running balance without filters will take a lot of time as it basically move data/metadata\nfrom the whole filesystem and needs to update all block pointers.\n\nThe filters can be used to perform following actions:\n\n• convert block group profiles (filter convert)\n\n• make block group usage more compact (filter usage)\n\n• perform actions only on a given device (filters devid, drange)\n\nThe filters can be applied to a combination of block group types (data, metadata, system).\nNote that changing only the system type needs the force option. Otherwise system gets auto‐\nmatically converted whenever metadata profile is converted.\n\nWhen metadata redundancy is reduced (e.g. from RAID1 to single) the force option is also re‐\nquired and it is noted in system log.\n" }, { "name": "NOTE:", "content": "The balance operation needs enough work space, i.e. space that is completely unused in the\nfilesystem, otherwise this may lead to ENOSPC reports. See the section ENOSPC for more\ndetails.\n" } ] }, "COMPATIBILITY": { "content": "", "subsections": [ { "name": "NOTE:", "content": "The balance subcommand also exists under the btrfs filesystem namespace. This still works\nfor backward compatibility but is deprecated and should not be used any more.\n" }, { "name": "NOTE:", "content": "A short syntax btrfs balance works due to backward compatibility but is deprecated\nand should not be used any more. Use btrfs balance start command instead.\n" } ] }, "PERFORMANCE IMPLICATIONS": { "content": "Balancing operations are very IO intensive and can also be quite CPU intensive, impacting\nother ongoing filesystem operations. Typically large amounts of data are copied from one lo‐\ncation to another, with corresponding metadata updates.\n\nDepending upon the block group layout, it can also be seek heavy. Performance on rotational\ndevices is noticeably worse compared to SSDs or fast arrays.\n", "subsections": [] }, "SUBCOMMAND": { "content": "", "subsections": [ { "name": "cancel ", "content": "cancels a running or paused balance, the command will block and wait until the current\nblock group being processed completes\n\nSince kernel 5.7 the response time of the cancellation is significantly improved, on\nolder kernels it might take a long time until currently processed chunk is completely\nfinished.\n" }, { "name": "pause ", "content": "pause running balance operation, this will store the state of the balance progress and\nused filters to the filesystem\n" }, { "name": "resume ", "content": "resume interrupted balance, the balance status must be stored on the filesystem from\nprevious run, e.g. after it was paused or forcibly interrupted and mounted again with\nskipbalance\n" }, { "name": "start [options] ", "content": "start the balance operation according to the specified filters, without any filters\nthe data and metadata from the whole filesystem are moved. The process runs in the\nforeground.\n\nNOTE:\nThe balance command without filters will basically move everything in the filesys‐\ntem to a new physical location on devices (i.e. it does not affect the logical\nproperties of file extents like offsets within files and extent sharing). The run\ntime is potentially very long, depending on the filesystem size. To prevent start‐\ning a full balance by accident, the user is warned and has a few seconds to cancel\nthe operation before it starts. The warning and delay can be skipped with\n--full-balance option.\n\nPlease note that the filters must be written together with the -d, -m and -s options,\nbecause they're optional and bare -d and -m also work and mean no filters.\n\nNOTE:\nWhen the target profile for conversion filter is raid5 or raid6, there's a safety\ntimeout of 10 seconds to warn users about the status of the feature\n\nOptions\n\n-d[]\nact on data block groups, see section FILTERS for details about filters\n\n-m[]\nact on metadata chunks, see FILTERS for details about filters\n\n-s[]\nact on system chunks (requires -f), see FILTERS for details about filters.\n\n-f force a reduction of metadata integrity, e.g. when going from raid1 to single,\nor skip safety timeout when the target conversion profile is raid5 or raid6\n\n--background|--bg\nrun the balance operation asynchronously in the background, uses fork(2) to\nstart the process that calls the kernel ioctl\n\n--enqueue\nwait if there's another exclusive operation running, otherwise continue\n" }, { "name": "-v (deprecated)", "content": "", "flag": "-v" }, { "name": "status [-v] ", "content": "Show status of running or paused balance.\n\nOptions\n" }, { "name": "-v (deprecated)", "content": "", "flag": "-v" } ] }, "FILTERS": { "content": "From kernel 3.3 onwards, BTRFS balance can limit its action to a subset of the whole filesys‐\ntem, and can be used to change the replication configuration (e.g. convert data from single\nto RAID1).\n\nBalance can be limited to a block group profile with the following options:\n\n• -d for data block groups\n\n• -m for metadata block groups (also implicitly applies to -s)\n\n• -s for system block groups\n\nThe options have an optional parameter which means that the parameter must start right after\nthe option without a space (this is mandatory getopt syntax), like -dusage=10. Options for\nall block group types can be specified in one command.\n\nA filter has the following structure: filter[=params][,filter=...]\n\nTo combine multiple filters use ,, without spaces. Example: -dconvert=raid1,soft\n\nBTRFS can have different profiles on a single device or the same profile on multiple device.\n\nThe main reason why you want to have different profiles for data and metadata is to provide\nadditional protection of the filesystem's metadata when devices fail, since a single sector\nof unrecoverable metadata will break the filesystem, while a single sector of lost data can\nbe trivially recovered by deleting the broken file.\n\nBefore changing profiles, make sure there is enough unallocated space on existing drives to\ncreate new metadata block groups (for filesystems over 50GiB, this is 1GB * (numberofde‐\nvices + 2)).\n\nDefault profiles on BTRFS are:\n\n• data: single\n\n•\n\nmetadata:\n\n• single devices: dup\n\n• multiple devices: raid1\n\nThe available filter types are:\n", "subsections": [ { "name": "Filter types", "content": "" }, { "name": "profiles=", "content": "Balances only block groups with the given profiles. Parameters are a list of profile\nnames separated by | (pipe).\n" }, { "name": "usage=, usage=", "content": "Balances only block groups with usage under the given percentage. The value of 0 is\nallowed and will clean up completely unused block groups, this should not require any\nnew work space allocated. You may want to use usage=0 in case balance is returning\nENOSPC and your filesystem is not too full.\n\nThe argument may be a single value or a range. The single value N means at most N per‐\ncent used, equivalent to ..N range syntax. Kernels prior to 4.4 accept only the single\nvalue format. The minimum range boundary is inclusive, maximum is exclusive.\n" }, { "name": "devid=", "content": "Balances only block groups which have at least one chunk on the given device. To list\ndevices with ids use btrfs filesystem show.\n" }, { "name": "drange=", "content": "Balance only block groups which overlap with the given byte range on any device. Use\nin conjunction with devid to filter on a specific device. The parameter is a range\nspecified as start..end.\n" }, { "name": "vrange=", "content": "Balance only block groups which overlap with the given byte range in the filesystem's\ninternal virtual address space. This is the address space that most reports from btrfs\nin the kernel log use. The parameter is a range specified as start..end.\n" }, { "name": "convert=", "content": "Convert each selected block group to the given profile name identified by parameters.\n\nNOTE:\nStarting with kernel 4.5, the data chunks can be converted to/from the DUP profile\non a single device.\n\nNOTE:\nStarting with kernel 4.6, all profiles can be converted to/from DUP on multi-device\nfilesystems.\n" }, { "name": "limit=, limit=", "content": "Process only given number of chunks, after all filters are applied. This can be used\nto specifically target a chunk in connection with other filters (drange, vrange) or\njust simply limit the amount of work done by a single balance run.\n\nThe argument may be a single value or a range. The single value N means at most N\nchunks, equivalent to ..N range syntax. Kernels prior to 4.4 accept only the single\nvalue format. The range minimum and maximum are inclusive.\n" }, { "name": "stripes=", "content": "Balance only block groups which have the given number of stripes. The parameter is a\nrange specified as start..end. Makes sense for block group profiles that utilize\nstriping, i.e. RAID0/10/5/6. The range minimum and maximum are inclusive.\n\nsoft Takes no parameters. Only has meaning when converting between profiles, or When doing\nconvert from one profile to another and soft mode is on, chunks that already have the\ntarget profile are left untouched. This is useful e.g. when half of the filesystem\nwas converted earlier but got cancelled.\n\nThe soft mode switch is (like every other filter) per-type. For example, this means\nthat we can convert metadata chunks the \"hard\" way while converting data chunks selec‐\ntively with soft switch.\n\nProfile names, used in profiles and convert are one of:\n\n• raid0\n\n• raid1\n\n• raid1c3\n\n• raid1c4\n\n• raid10\n\n• raid5\n\n• raid6\n\n• dup\n\n• single\n\nThe mixed data/metadata profiles can be converted in the same way, but conversion between\nmixed and non-mixed is not implemented. For the constraints of the profiles please refer to\nmkfs.btrfs(8) section PROFILES.\n" } ] }, "ENOSPC": { "content": "The way balance operates, it usually needs to temporarily create a new block group and move\nthe old data there, before the old block group can be removed. For that it needs the work\nspace, otherwise it fails for ENOSPC reasons. This is not the same ENOSPC as if the free\nspace is exhausted. This refers to the space on the level of block groups, which are bigger\nparts of the filesystem that contain many file extents.\n\nThe free work space can be calculated from the output of the btrfs filesystem show command:\n\nLabel: 'BTRFS' uuid: 8a9d72cd-ead3-469d-b371-9c7203276265\nTotal devices 2 FS bytes used 77.03GiB\ndevid 1 size 53.90GiB used 51.90GiB path /dev/sdc2\ndevid 2 size 53.90GiB used 51.90GiB path /dev/sde1\n\nsize - used = free work space\n\n53.90GiB - 51.90GiB = 2.00GiB\n\nAn example of a filter that does not require workspace is usage=0. This will scan through all\nunused block groups of a given type and will reclaim the space. After that it might be possi‐\nble to run other filters.\n\nCONVERSIONS ON MULTIPLE DEVICES\n\nConversion to profiles based on striping (RAID0, RAID5/6) require the work space on each de‐\nvice. An interrupted balance may leave partially filled block groups that consume the work\nspace.\n", "subsections": [] }, "EXAMPLES": { "content": "A more comprehensive example when going from one to multiple devices, and back, can be found\nin section TYPICAL USECASES of btrfs-device(8).\n\nMAKING BLOCK GROUP LAYOUT MORE COMPACT\nThe layout of block groups is not normally visible; most tools report only summarized numbers\nof free or used space, but there are still some hints provided.\n\nLet's use the following real life example and start with the output:\n\n$ btrfs filesystem df /path\nData, single: total=75.81GiB, used=64.44GiB\nSystem, RAID1: total=32.00MiB, used=20.00KiB\nMetadata, RAID1: total=15.87GiB, used=8.84GiB\nGlobalReserve, single: total=512.00MiB, used=0.00B\n\nRoughly calculating for data, 75G - 64G = 11G, the used/total ratio is about 85%. How can we\ncan interpret that:\n\n• chunks are filled by 85% on average, i.e. the usage filter with anything smaller than 85\nwill likely not affect anything\n\n• in a more realistic scenario, the space is distributed unevenly, we can assume there are\ncompletely used chunks and the remaining are partially filled\n\nCompacting the layout could be used on both. In the former case it would spread data of a\ngiven chunk to the others and removing it. Here we can estimate that roughly 850 MiB of data\nhave to be moved (85% of a 1 GiB chunk).\n\nIn the latter case, targeting the partially used chunks will have to move less data and thus\nwill be faster. A typical filter command would look like:\n\n# btrfs balance start -dusage=50 /path\nDone, had to relocate 2 out of 97 chunks\n\n$ btrfs filesystem df /path\nData, single: total=74.03GiB, used=64.43GiB\nSystem, RAID1: total=32.00MiB, used=20.00KiB\nMetadata, RAID1: total=15.87GiB, used=8.84GiB\nGlobalReserve, single: total=512.00MiB, used=0.00B\n\nAs you can see, the total amount of data is decreased by just 1 GiB, which is an expected re‐\nsult. Let's see what will happen when we increase the estimated usage filter.\n\n# btrfs balance start -dusage=85 /path\nDone, had to relocate 13 out of 95 chunks\n\n$ btrfs filesystem df /path\nData, single: total=68.03GiB, used=64.43GiB\nSystem, RAID1: total=32.00MiB, used=20.00KiB\nMetadata, RAID1: total=15.87GiB, used=8.85GiB\nGlobalReserve, single: total=512.00MiB, used=0.00B\n\nNow the used/total ratio is about 94% and we moved about 74G - 68G = 6G of data to the re‐\nmaining block groups, i.e. the 6GiB are now free of filesystem structures, and can be reused\nfor new data or metadata block groups.\n\nWe can do a similar exercise with the metadata block groups, but this should not typically be\nnecessary, unless the used/total ratio is really off. Here the ratio is roughly 50% but the\ndifference as an absolute number is \"a few gigabytes\", which can be considered normal for a\nworkload with snapshots or reflinks updated frequently.\n\n# btrfs balance start -musage=50 /path\nDone, had to relocate 4 out of 89 chunks\n\n$ btrfs filesystem df /path\nData, single: total=68.03GiB, used=64.43GiB\nSystem, RAID1: total=32.00MiB, used=20.00KiB\nMetadata, RAID1: total=14.87GiB, used=8.85GiB\nGlobalReserve, single: total=512.00MiB, used=0.00B\n\nJust 1 GiB decrease, which possibly means there are block groups with good utilization. Mak‐\ning the metadata layout more compact would in turn require updating more metadata structures,\ni.e. lots of IO. As running out of metadata space is a more severe problem, it's not neces‐\nsary to keep the utilization ratio too high. For the purpose of this example, let's see the\neffects of further compaction:\n\n# btrfs balance start -musage=70 /path\nDone, had to relocate 13 out of 88 chunks\n\n$ btrfs filesystem df .\nData, single: total=68.03GiB, used=64.43GiB\nSystem, RAID1: total=32.00MiB, used=20.00KiB\nMetadata, RAID1: total=11.97GiB, used=8.83GiB\nGlobalReserve, single: total=512.00MiB, used=0.00B\n\nGETTING RID OF COMPLETELY UNUSED BLOCK GROUPS\nNormally the balance operation needs a work space, to temporarily move the data before the\nold block groups gets removed. If there's no work space, it ends with no space left.\n\nThere's a special case when the block groups are completely unused, possibly left after re‐\nmoving lots of files or deleting snapshots. Removing empty block groups is automatic since\n3.18. The same can be achieved manually with a notable exception that this operation does not\nrequire the work space. Thus it can be used to reclaim unused block groups to make it avail‐\nable.\n\n# btrfs balance start -dusage=0 /path\n\nThis should lead to decrease in the total numbers in the btrfs filesystem df output.\n", "subsections": [] }, "EXIT STATUS": { "content": "Unless indicated otherwise below, all btrfs balance subcommands return a zero exit status if\nthey succeed, and non zero in case of failure.\n\nThe pause, cancel, and resume subcommands exit with a status of 2 if they fail because a bal‐\nance operation was not running.\n\nThe status subcommand exits with a status of 0 if a balance operation is not running, 1 if\nthe command-line usage is incorrect or a balance operation is still running, and 2 on other\nerrors.\n", "subsections": [] }, "AVAILABILITY": { "content": "btrfs is part of btrfs-progs. Please refer to the documentation at\nhttps://btrfs.readthedocs.io.\n", "subsections": [] }, "SEE ALSO": { "content": "mkfs.btrfs(8), btrfs-device(8)\n\n\n6.6.3 Mar 31, 2024 BTRFS-BALANCE(8)", "subsections": [] } }, "summary": "btrfs-balance - balance block groups on a btrfs filesystem", "flags": [], "examples": [ "A more comprehensive example when going from one to multiple devices, and back, can be found", "in section TYPICAL USECASES of btrfs-device(8).", "MAKING BLOCK GROUP LAYOUT MORE COMPACT", "The layout of block groups is not normally visible; most tools report only summarized numbers", "of free or used space, but there are still some hints provided.", "Let's use the following real life example and start with the output:", "$ btrfs filesystem df /path", "Data, single: total=75.81GiB, used=64.44GiB", "System, RAID1: total=32.00MiB, used=20.00KiB", "Metadata, RAID1: total=15.87GiB, used=8.84GiB", "GlobalReserve, single: total=512.00MiB, used=0.00B", "Roughly calculating for data, 75G - 64G = 11G, the used/total ratio is about 85%. How can we", "can interpret that:", "• chunks are filled by 85% on average, i.e. the usage filter with anything smaller than 85", "will likely not affect anything", "• in a more realistic scenario, the space is distributed unevenly, we can assume there are", "completely used chunks and the remaining are partially filled", "Compacting the layout could be used on both. In the former case it would spread data of a", "given chunk to the others and removing it. Here we can estimate that roughly 850 MiB of data", "have to be moved (85% of a 1 GiB chunk).", "In the latter case, targeting the partially used chunks will have to move less data and thus", "will be faster. A typical filter command would look like:", "# btrfs balance start -dusage=50 /path", "Done, had to relocate 2 out of 97 chunks", "$ btrfs filesystem df /path", "Data, single: total=74.03GiB, used=64.43GiB", "System, RAID1: total=32.00MiB, used=20.00KiB", "Metadata, RAID1: total=15.87GiB, used=8.84GiB", "GlobalReserve, single: total=512.00MiB, used=0.00B", "As you can see, the total amount of data is decreased by just 1 GiB, which is an expected re‐", "sult. Let's see what will happen when we increase the estimated usage filter.", "# btrfs balance start -dusage=85 /path", "Done, had to relocate 13 out of 95 chunks", "$ btrfs filesystem df /path", "Data, single: total=68.03GiB, used=64.43GiB", "System, RAID1: total=32.00MiB, used=20.00KiB", "Metadata, RAID1: total=15.87GiB, used=8.85GiB", "GlobalReserve, single: total=512.00MiB, used=0.00B", "Now the used/total ratio is about 94% and we moved about 74G - 68G = 6G of data to the re‐", "maining block groups, i.e. the 6GiB are now free of filesystem structures, and can be reused", "for new data or metadata block groups.", "We can do a similar exercise with the metadata block groups, but this should not typically be", "necessary, unless the used/total ratio is really off. Here the ratio is roughly 50% but the", "difference as an absolute number is \"a few gigabytes\", which can be considered normal for a", "workload with snapshots or reflinks updated frequently.", "# btrfs balance start -musage=50 /path", "Done, had to relocate 4 out of 89 chunks", "$ btrfs filesystem df /path", "Data, single: total=68.03GiB, used=64.43GiB", "System, RAID1: total=32.00MiB, used=20.00KiB", "Metadata, RAID1: total=14.87GiB, used=8.85GiB", "GlobalReserve, single: total=512.00MiB, used=0.00B", "Just 1 GiB decrease, which possibly means there are block groups with good utilization. Mak‐", "ing the metadata layout more compact would in turn require updating more metadata structures,", "i.e. lots of IO. As running out of metadata space is a more severe problem, it's not neces‐", "sary to keep the utilization ratio too high. For the purpose of this example, let's see the", "effects of further compaction:", "# btrfs balance start -musage=70 /path", "Done, had to relocate 13 out of 88 chunks", "$ btrfs filesystem df .", "Data, single: total=68.03GiB, used=64.43GiB", "System, RAID1: total=32.00MiB, used=20.00KiB", "Metadata, RAID1: total=11.97GiB, used=8.83GiB", "GlobalReserve, single: total=512.00MiB, used=0.00B", "GETTING RID OF COMPLETELY UNUSED BLOCK GROUPS", "Normally the balance operation needs a work space, to temporarily move the data before the", "old block groups gets removed. If there's no work space, it ends with no space left.", "There's a special case when the block groups are completely unused, possibly left after re‐", "moving lots of files or deleting snapshots. Removing empty block groups is automatic since", "3.18. The same can be achieved manually with a notable exception that this operation does not", "require the work space. Thus it can be used to reclaim unused block groups to make it avail‐", "able.", "# btrfs balance start -dusage=0 /path", "This should lead to decrease in the total numbers in the btrfs filesystem df output." ], "see_also": [ { "name": "mkfs.btrfs", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/mkfs.btrfs/8/json" }, { "name": "btrfs-device", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/btrfs-device/8/json" } ] }