SHEEPDOG(8)
👑 SHEEPDOG(8) 👑
Bestie, welcome to the SHEEPDOG(8) page from the System Manager's Manual. This is the manual for dog, and it's giving iconic distributed-storage energy. 🦋
✨ NAME ✨
dog — command line utility for the sheep daemon. That's the whole vibe: one CLI, one daemon bestie, zero central drama. 💅
🌸 SYNOPSIS 🌸
dog <command> <subcommand> [options]💖 DESCRIPTION 💖
Okay so dog — Sheepdog is a distributed storage system for QEMU. It provides highly available block level storage volumes to virtual machines. Sheepdog supports advanced volume management features such as snapshot, cloning, and thin provisioning. The architecture of Sheepdog is fully symmetric; there is no central node such as a meta-data server.
The server daemon is called sheep(8). A command line utility is available via dog(8). QEMU virtual machines use the sheep daemon via a block driver available in qemu(1).
For more information, run dog <command> <subcommand> --help. No cap, that's the cheat code. ✨
🎀 COMMAND & SUBCOMMAND 🎀
🌈 vdi check 🌈
vdi check [-s snapshot] [-e] [-a address] [-p port] [-h] [-T] <vdiname>Options
-s,--snapshot— specify a snapshot id or tag name-e,--exist— only check objects exist or not, neither comparing nor repairing-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
check and repair image's consistency. It's the glow-up check for your image, bestie. 💖
🌈 vdi create 🌈
vdi create [-P] [-y] [-c copies] [-a address] [-p port] [-h] [-r] [-v] [-z block_size_shift] [-T] <vdiname> <size>Options
-P,--prealloc— preallocate all the data objects-y,--hyper— create a hyper volume-c,--copies— specify the data redundancy level-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes-v,--verbose— print more information than default-z,--block_size_shift— specify the bit shift num for data object size-T,--time— show elapsed time
Description
create an image. New image, new me, periodt. 🎀
🌈 vdi snapshot 🌈
vdi snapshot [-s snapshot] [-a address] [-p port] [-h] [-r] [-v] [-T] [-R] <vdiname>Options
-s,--snapshot— specify a snapshot id or tag name-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes-v,--verbose— print more information than default-T,--time— show elapsed time-R,--reduce-identical-snapshots— do not create snapshot if working VDI doesn't have its own objects
Description
create a snapshot. It's giving freeze-frame main character moment. 📸
🌈 vdi clone 🌈
vdi clone [-s snapshot] [-P] [-n] [-a address] [-p port] [-h] [-r] [-v] [-T] <src vdi> <dst vdi>Options
-s,--snapshot— specify a snapshot id or tag name-P,--prealloc— preallocate all the data objects-n,--no-share— share nothing with its parent-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes-v,--verbose— print more information than default-T,--time— show elapsed time
Description
clone an image. Twin era, but make it storage. 👯
🌈 vdi delete 🌈
vdi delete [-s snapshot] [-a address] [-p port] [-h] [-T] <vdiname>Options
-s,--snapshot— specify a snapshot id or tag name-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
delete an image. Say bye, but like, responsibly. 💀
🌈 vdi rollback 🌈
vdi rollback [-s snapshot] [-a address] [-p port] [-h] [-f] [-r] [-v] [-T] <vdiname>Options
-s,--snapshot— specify a snapshot id or tag name-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-f,--force— do operation forcibly-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes-v,--verbose— print more information than default-T,--time— show elapsed time
Description
rollback to a snapshot. Rewind time, bestie. ⏪
🌈 vdi list 🌈
vdi list [-a address] [-p port] [-r] [-h] [-o oid] [-T] [vdiname]Options
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes-h,--help— display this help and exit-o,--oid— specify the object id of the tracking object-T,--time— show elapsed time
Description
list images. It's the roll call, and everyone looks cute. 📋✨
🌈 vdi tree 🌈
vdi tree [-a address] [-p port] [-h] [-T]Options
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
show images in tree view format. Family tree, but for VDIs. 🌳
🌈 vdi graph 🌈
vdi graph [-a address] [-p port] [-h] [-T]Options
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
show images in Graphviz dot format. Graph girlies, this one's for you. 📊
🌈 vdi object 🌈
vdi object {location|map|dump-inode} [-i index] [-s snapshot] [-a address] [-p port] [-h] [-T] <vdiname>Available subcommands
location— show object location informationmap— show object map informationdump-inode— dump inode information
Options
-i,--index— specify the index of data objects-s,--snapshot— specify a snapshot id or tag name-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
show object information in the image. Object lore unlocked. 🔍
🌈 vdi track 🌈
vdi track [-i index] [-s snapshot] [-a address] [-p port] [-h] [-o oid] [-T] <vdiname>Options
-i,--index— specify the index of data objects-s,--snapshot— specify a snapshot id or tag name-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-o,--oid— specify the object id of the tracking object-T,--time— show elapsed time
Description
show the object epoch trace in the image. Tracking tea, served fresh. 🍵
🌈 vdi setattr 🌈
vdi setattr [-d] [-x] [-a address] [-p port] [-h] [-T] <vdiname> <key> [value]Options
-d,--delete— delete a key-x,--exclusive— write in an exclusive mode-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
set a VDI attribute. Customize your fave. 💅
🌈 vdi getattr 🌈
vdi getattr [-a address] [-p port] [-h] [-T] <vdiname> <key>Options
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
get a VDI attribute. Receipts, but for metadata. 🧾
🌈 vdi resize 🌈
vdi resize [-a address] [-p port] [-h] [-T] <vdiname> <new size>Options
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
resize an image. Glow-up, but make it capacity. 📏
🌈 vdi read 🌈
vdi read [-s snapshot] [-a address] [-p port] [-h] [-T] <vdiname> [<offset> [<len>]]Options
-s,--snapshot— specify a snapshot id or tag name-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description
read data from an image. Reading era, we love to see it. 📖
🌈 vdi write 🌈
vdi write [-a address] [-p port] [-w] [-h] [-T] <vdiname> [<offset> [<len>]]Options
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-w,--writeback— use writeback mode-h,--help— display this help and exit-T,--time— show elapsed time
Description
write data to an image. Pen to paper, bytes to block. ✍️
vdi backup 💾
vdi backup [-s snapshot] [-F from] [-a address] [-p port] [-h] [-T] <vdiname>Options:
-s,--snapshot— pick a snapshot id or tag name, bestie 🎀-F,--from— create a differential backup from the snapshot, like a glow-up from that point ✨-a,--address— specify the daemon address (default:localhost) — it's giving "where u at" 💅-p,--port— specify the daemon port, the door number basically 🚪-h,--help— display this help and exit, no cap 🙌-T,--time— show elapsed time, so u know how long u've been slaying ⏳
Description:
create an incremental backup between two snapshots and outputs to STDOUT — it's literally saving the diff between two snapshots and yeeting it to STDOUT 💖
vdi restore 🦋
vdi restore [-s snapshot] [-a address] [-p port] [-h] [-T] <vdiname>Options:
-s,--snapshot— specify a snapshot id or tag name, so u know which one to bring back 🦋-a,--address— specify the daemon address (default:localhost) — the addy of the daemon, iykyk 💜-p,--port— specify the daemon port, the port number, slay 🌸-h,--help— display this help and exit, we love a helpful queen 👑-T,--time— show elapsed time, for the time-conscious besties ⏰
Description:
restore snapshot images from a backup provided in STDIN — it takes the backup from STDIN and restores the snapshot images, no cap 💅
vdi alter-copy 🎀
vdi alter-copy [-c copies] [-a address] [-p port] [-h] [-T] [-f] <vdiname>Options:
-c,--copies— specify the data redundancy level, like how many copies u want, the more the merrier 🎀-a,--address— specify the daemon address (default:localhost) — where the daemon is at 🏠-p,--port— specify the daemon port, the port, periodt 🚪-h,--help— display this help and exit, a classic 🙌-T,--time— show elapsed time, so u can clock the tea ⏳-f,--force— do operation forcibly, like no questions asked, just do it 💥
Description:
set the vdi's redundancy level — it sets how many copies of the vdi u want, fr fr 💖
vdi lock 🔒
vdi lock {list|unlock} [-s snapshot] [-a address] [-p port] [-h] [-T]Available subcommands:
list— list locked VDIs, so u can see who's locked up 🔒unlock— unlock locked VDI forcibly, break them free, no cap 🔓
Options:
-s,--snapshot— specify a snapshot id or tag name, for the snapshot vibe 🎀-a,--address— specify the daemon address (default:localhost) — the daemon's addy 💜-p,--port— specify the daemon port, the port number, slay 🌸-h,--help— display this help and exit, we love to see it 👑-T,--time— show elapsed time, for the time girlies ⏰
Description:
See dog vdi lock for more information — so go check that out, bestie 💅
node kill 💀
node kill [-a address] [-p port] [-r] [-h] [-l] [-T] <node id>Options:
-a,--address— specify the daemon address (default:localhost) — where the daemon is at 🏠-p,--port— specify the daemon port, the port, periodt 🚪-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes — it's giving no filter, just the raw tea 💅-h,--help— display this help and exit, a classic 🙌-l,--local— issue request to local node, keep it local, bestie 🏡-T,--time— show elapsed time, so u can clock the tea ⏳
Description:
kill node — it kills the node, periodt 💀
node list 📝
node list [-a address] [-p port] [-r] [-h] [-T]Options:
-a,--address— specify the daemon address (default:localhost) — the daemon's addy 💜-p,--port— specify the daemon port, the port number, slay 🌸-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes — no filter, just facts ✨-h,--help— display this help and exit, we love a helpful queen 👑-T,--time— show elapsed time, for the time-conscious besties ⏰
Description:
list nodes — it lists all the nodes, no cap 📝
node info 🍵
node info [-a address] [-p port] [-r] [-h] [-T]Options:
-a,--address— specify the daemon address (default:localhost) — where the daemon is at 🏠-p,--port— specify the daemon port, the port, periodt 🚪-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes — raw and unfiltered, just how we like it 💅-h,--help— display this help and exit, a classic 🙌-T,--time— show elapsed time, so u can clock the tea ⏳
Description:
show information about each node — it spills the tea on each node, fr fr 🍵
node recovery 🏎️
node recovery {info|set-throttle|get-throttle} [-a address] [-p port] [-h] [-P] [-r] [-T] <max> <interval>Available subcommands:
info— show recovery information of nodes (default) — the default, so if u don't pick, this is it 💅set-throttle— set new throttling — set the throttle, like control the speed 🏎️get-throttle— get current throttling — get the current throttle, so u know what's up 📊
Options:
-a,--address— specify the daemon address (default:localhost) — the daemon's addy 💜-p,--port— specify the daemon port, the port number, slay 🌸-h,--help— display this help and exit, we love to see it 👑-P,--progress— show progress of recovery in the node — show the progress, so u can watch it slay 💪-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes — no filter, just the raw data ✨-T,--time— show elapsed time, for the time girlies ⏰
Description:
show recovery information or set/get recovery speed throttling of nodes — it shows recovery info or sets/gets the recovery speed throttle for nodes, no cap 💖
node md 💽
node md {info|plug|unplug} [-a address] [-p port] [-r] [-A] [-f] [-h] [-T] [disks]Available subcommands:
info— show multi-disk information — shows info about multiple disks, slay 💽plug— plug more disk(s) into node — plug in more disks, like adding more storage besties 🔌unplug— unplug disk(s) from node — unplug disks, bye Felicia 👋
Options:
-a,--address— specify the daemon address (default:localhost) — where the daemon is at 🏠-p,--port— specify the daemon port, the port, periodt 🚪-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes — raw and real 💅-A,--all— show md information of all the nodes — show md info for ALL the nodes, go big or go home 🌈-f,--force— ignore the confirmation — ignore the confirmation, just do it 💥-h,--help— display this help and exit, a classic 🙌-T,--time— show elapsed time, so u can clock the tea ⏳
Description:
See dog node md for more information — go check that out, bestie 💅
node stat 📊
node stat [-a address] [-p port] [-r] [-w] [-h] [-T]Options:
-a,--address— specify the daemon address (default:localhost) — the daemon's addy 💜-p,--port— specify the daemon port, the port number, slay 🌸-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes — no filter, just facts ✨-w,--watch— watch the stat every second — watch the stats every second, like a hawk 👀-h,--help— display this help and exit, we love a helpful queen 👑-T,--time— show elapsed time, for the time-conscious besties ⏰
Description:
show stat information about the node — it shows stat info about the node, fr fr 📊
node log level 🎚️
node log level [-a address] [-p port] [-h] [-T]Available subcommands:
level— manipulate loglevel — manipulate the log level, like turn it up or down 🎚️
Options:
-a,--address— specify the daemon address (default:localhost) — where the daemon is at 🏠-p,--port— specify the daemon port, the port, periodt 🚪-h,--help— display this help and exit, a classic 🙌-T,--time— show elapsed time, so u can clock the tea ⏳
Description:
show or set log level of the node — it shows or sets the log level of the node, no cap 💅
node vnodes set ✨
node vnodes set [-a address] [-p port] [-h] <num of vnodes>Available subcommands:
set— set new vnodes — set new vnodes, like a fresh start ✨
Options:
-a,--address— specify the daemon address (default:localhost) — the daemon's addy 💜-p,--port— specify the daemon port, the port number, slay 🌸-h,--help— display this help and exit, we love to see it 👑
Description:
set new vnodes — it sets new vnodes, periodt 💖
node format 💅
node format [-a address] [-p port] [-h] [-T] <directory of sheep> <a name of store format>Options:
-a,--address— specify the daemon address (default:localhost) — where the daemon is at 🏠-p,--port— specify the daemon port, the port, periodt 🚪-h,--help— display this help and exit, a classic 🙌-T,--time— show elapsed time, so u can clock the tea ⏳
Description:
initialize store format of the node — it initializes the store format of the node, fr fr 💅
cluster info 💖
cluster info [-a address] [-p port] [-r] [-h] [-v] [-T] [-d]Options:
-a,--address— specify the daemon address (default:localhost) — the daemon's addy 💜-p,--port— specify the daemon port, the port number, slay 🌸-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes — no filter, just facts ✨-h,--help— display this help and exit, we love a helpful queen 👑-v,--verbose— print more information than default — spill extra tea, verbose mode 🍵-T,--time— show elapsed time, for the time-conscious besties ⏰-d,--diff— just output the changes between the two adjacent epoches for cluster info — just the diff between adjacent epoches, like a before and after 💅
Description:
show cluster information — it shows cluster info, no cap 💖
cluster format 🛍️
cluster format [-b store] [-c copies] [-l] [-t] [-a address] [-p port] [-h] [-z block_size_shift] [-T] [-V] [-R] [-f] [-F]Options:
-b,--store— specify backend store — pick the backend store, like choosing ur vibe 🛍️-c,--copies— specify the default data redundancy (number of copies) — set the default number of copies, the more the merrier 🎀-l,--lock— Lock vdi to exclude multiple users — lock the vdi so multiple users can't touch it, exclusive vibes 🔒-t,--strict— do not serve write request if number of nodes is not sufficient — strict mode: no write requests if there aren't enough nodes, periodt 💅-a,--address— specify the daemon address (default:localhost) — where the daemon is at 🏠-p,--port— specify the daemon port, the port, periodt 🚪-h,--help— display this help and exit, a classic 🙌-z,--block_size_shift— specify the shift num of default data object size — specify the shift num for the default data object size, math queen 👑-T,--time— show elapsed time, so u can clock the tea ⏳-V,--fixedvnodes— disable automatic vnodes calculation — disable automatic vnodes calculation, manual mode ✋-R,--recyclevid— enable recycling of VID — enable VID recycling, sustainable queen ♻️-f,--force— do not prompt for confirmation — do not prompt for confirmation, just do it 💥-F,--avoid-diskfull— skip recovery if recovery process can cause disk full — skip recovery if it would cause a disk full, protecting ur disk space 🛡️
Description:
create a Sheepdog store — it creates a Sheepdog store, slay 💖
cluster shutdown 💀
cluster shutdown [-a address] [-p port] [-h] [-T]Options:
-a,--address— specify the daemon address (default:localhost) — the daemon's addy 💜-p,--port— specify the daemon port, the port number, slay 🌸-h,--help— display this help and exit, we love to see it 👑-T,--time— show elapsed time, for the time girlies ⏰
Description:
stop Sheepdog — it stops Sheepdog, periodt 💀
cluster snapshot 📸
cluster snapshot {save|list|load|show} [-a address] [-p port] [-h] [-T] [-m] <tag|idx> <path> [vdi1] [vdi2] ...okay bestie, so for cluster snapshot the available subcommands are giving:
save— save snapshot tolocalpathlist— list snapshot oflocalpathload— load snapshot fromlocalpathshow— show vdi list from snapshot
Options:
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time-m,--multithread— use multi-thread for 'cluster snapshot save'
Description: snapshot/restore the cluster
cluster recover 🛟
Usage:
cluster recover {force|enable|disable} [-a address] [-f] [-p port] [-h] [-T]Available subcommands:
force— force recover cluster immediatelyenable— enable automatic recovery and run once recover if necessarydisable— disable automatic recovery
Options:
-a,--address— specify the daemon address (default:localhost)-f,--force— do not prompt for confirmation-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description: See 'dog cluster recover' for more information
cluster reweight ⚖️
Usage:
cluster reweight [-a address] [-p port] [-h] [-T]Options:
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description: reweight the cluster
cluster check ✅
Usage:
cluster check [-a address] [-p port] [-h] [-T]Options:
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time
Description: check and repair cluster
cluster alter-copy 📦
Usage:
cluster alter-copy [-a address] [-p port] [-h] [-T] [-c copies] [-f]Options:
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-h,--help— display this help and exit-T,--time— show elapsed time-c,--copies— specify the default data redundancy (number of copies)-f,--force— do not prompt for confirmation
Description: set the cluster's redundancy level
upgrade inode-convert 🪄
Usage:
upgrade inode-convert [-h] [-T] [-o orig-version] <path of original inode file> <path of new inode file>Options:
-h,--help— display this help and exit-T,--time— show elapsed time-o,--orig-version— version of converting file
Description: upgrade inode object file
upgrade epoch-convert 🕰️
Usage:
upgrade epoch-convert [-h] [-T] [-o orig-version] <path of original epoch log file> <path of new epoch log file>Options:
-h,--help— display this help and exit-T,--time— show elapsed time-o,--orig-version— version of converting file
Description: upgrade epoch log file
benchmark io 🏎️
Usage:
benchmark io [-a address] [-p port] [-r] [-h] [-T] [-f] [-w workqueue] [-t total] [-n nr-threads] <vdiname>Options:
-a,--address— specify the daemon address (default:localhost)-p,--port— specify the daemon port-r,--raw— raw output mode: omit headers, separate fields with single spaces and print all sizes in decimal bytes-h,--help— display this help and exit-T,--time— show elapsed time-f,--force— do not prompt for confirmation-w,--workqueue— specify workqueue type-t,--total— a number of total operation (e.g. I/O request)-n,--nr-threads— a number of worker threads (only used for fixed workqueue)
Description: benchmark I/O performance
✨ DEPENDENCIES ✨
Sheepdog requires QEMU 0.13.z or later and Corosync 1.y.z or 2.y.z.
📁 FILES 📁
none
👀 SEE ALSO 👀
sheep(8), qemu(1), sheepfs(8), corosync_overview(8)
👑 AUTHORS 👑
This software is developed by the Sheepdog community which may be reached via mailing list at <sheepdog@lists.wpkg.org>.
that's the whole vibe, bestie! 💖 signing off from SHEEPDOG(8) on 2025-07-26. periodt. 🌈✨
📦 where this page comes from
- fedora 43: sheepdog 1.0.1-25.fc43 · GPL-2.0-only AND GPL-2.0-or-later
- 💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms