Add ZFS usage tips to freebsd-tips.
Add a bunch of examples on how to use ZFS features like: - listing available space, - setting and displaying a userquota, - displaying pool I/O statistics and pool history, - displaying the compression ratio for a dataset, - various list options (sorting, removing headers), - performing a dry-run of a snapshot delete, - removing a range of snapshots, - setting a custom property, - preventing removal of a snapshot with ZFS holds, - permission sets for zfs send/receive. Additionally, clarify the existing examples a bit when it comes to displaying space by mentioning UFS explicitly. Other examples include displaying I/O in top(1), querying sysctl(8) for active CPUs and available RAM. Mention systat(1) and its options, too. While here, reformat the example to upload a dmesg(8) a bit to wrap properly. Thanks to Allan Jude for his help with some of the ZFS examples. Reviewed by: dru,allanjude Approved by: allanjude (earlier version) MFC after: 3 days Relnotes: yes (ZFS examples in freebsd-tips) Differential Revision: https://reviews.freebsd.org/D18541
This commit is contained in:
Benedict Reuschling
parent
989321df11
commit
56d417fd5d
@ -270,12 +270,12 @@ To see how long it takes a command to run, type the word "time" before the
|
||||
command name.
|
||||
-- Dru <genesis@istar.ca>
|
||||
%
|
||||
To see how much disk space is left on your partitions, use
|
||||
To see how much disk space is left on your UFS partitions, use
|
||||
|
||||
df -h
|
||||
-- Dru <genesis@istar.ca>
|
||||
%
|
||||
To see the 10 largest files on a directory or partition, use
|
||||
To see the 10 largest files on a directory or UFS partition, use
|
||||
|
||||
du -h /partition_or_directory_name | sort -rh | head
|
||||
-- Dru <genesis@istar.ca>
|
||||
@ -554,9 +554,241 @@ Use "sysrc name=value" to add an entry and "sysrc -x name" to delete an entry.
|
||||
|
||||
-- Lars Engels <lme@FreeBSD.org>
|
||||
%
|
||||
You can upload the dmesg of your system to help developers get an overview of commonly used hardware and peripherals for FreeBSD.
|
||||
Use the curl package to upload it in one command:
|
||||
You can upload the dmesg of your system to help developers get an overview of commonly
|
||||
used hardware and peripherals for FreeBSD. Use the curl package to upload it like this:
|
||||
curl -v -d "nickname=$USER" -d "description=FreeBSD/$(uname -m) on \
|
||||
$(kenv smbios.system.maker) $(kenv smbios.system.product)" -d "do=addd" \
|
||||
--data-urlencode 'dmesg@/var/run/dmesg.boot' http://dmesgd.nycbug.org/index.cgi
|
||||
%
|
||||
Want to know how much memory (in bytes) your machine has available? Let
|
||||
sysctl(8) tell you with the following command:
|
||||
|
||||
sysctl hw.physmem
|
||||
|
||||
The number of active CPUs is displayed using this command:
|
||||
|
||||
sysctl hw.ncpu
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
When using ZFS as the file system the "df" command will display confusing
|
||||
values. Use the built-in "zfs list" command to get an overview of space usage:
|
||||
|
||||
zfs list -o space
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
To learn more about what your system is doing, take a look at systat(1). For
|
||||
example, to get an overview of I/O happening in the system, run:
|
||||
|
||||
systat -iostat
|
||||
|
||||
Other values are icmp, icmp6, ifstat, ip, ip6, netstat, pigs, sctp, swap, tcp,
|
||||
vmstat, or zarc. You can switch between displays using :<display> and exit
|
||||
back to your shell by typing
|
||||
|
||||
:quit
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
To set a quota of 10 GB for the user named foo on a ZFS dataset, run the
|
||||
following command:
|
||||
|
||||
# zfs set userquota@foo=10G pool/home/foo
|
||||
|
||||
The zfs userspace command can display the quota and current space usage:
|
||||
|
||||
# zfs userspace pool/home/foo
|
||||
|
||||
To unset a quota, assign "none" as the value.
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
ZFS can display I/O statistics for a given pool using the iostat subcommand.
|
||||
By default, it will display one line of current activity. To display stats
|
||||
every 5 seconds run the following command (cancel with CTRL+C):
|
||||
|
||||
zpool iostat 5
|
||||
|
||||
To view individual disk activities, specify the -v parameter:
|
||||
|
||||
zpool iostat -v
|
||||
|
||||
Of course, both can be combined. For more options, see zpool(8).
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
FreeBSD's top(1) utility displays CPU statistics by default.
|
||||
To display I/O activity for each process instead, run top like this:
|
||||
|
||||
top -m io
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
ZFS keeps a history of commands run against a specific pool using the
|
||||
history subcommand to zpool:
|
||||
|
||||
zpool history
|
||||
|
||||
More details are available using the -i and -l parameters. Note that ZFS
|
||||
will not keep the complete pool history forever and will remove older
|
||||
events in favor of never ones.
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
To display the compression ratio for the ZFS dataset /var/log on the pool
|
||||
mypool, run the following command:
|
||||
|
||||
zfs get refcompressratio mypool/var/log
|
||||
|
||||
The refcompressratio will only display the compression ratio for that specific
|
||||
dataset, not the descendant datasets. To include the child datasets, the
|
||||
command looks like this:
|
||||
|
||||
zfs get compressratio mypool/var
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
You can limit the depth of the displayed datasets in the "zfs list" output
|
||||
using the -d parameter. To display only the first level of datasets below
|
||||
mypool/usr and not the ones deeper than those, run this command:
|
||||
|
||||
zfs list -d 1 mypool/usr
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
The "zfs list" command can be filtered in multiple ways. To display just
|
||||
the dataset name, use the -o parameter:
|
||||
|
||||
zfs list -o name mypool/usr
|
||||
|
||||
More columns and their order can be defined by separating them with commas:
|
||||
|
||||
zfs list -o mountpoint,name,avail
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
The output of "zfs list" can be sorted by a specific column using -s. To
|
||||
sort the datasets by the "used" column in ascending order, run this command:
|
||||
|
||||
zfs list -s used
|
||||
|
||||
To sort in descending order instead, use -S:
|
||||
|
||||
zfs list -S used
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
To make the "zfs list" output more script-friendly, you can suppress the
|
||||
output of the headers for each column by passing the -H parameter:
|
||||
|
||||
zfs list -H
|
||||
|
||||
Another helpful option for script writers is -p, which displays the numbers
|
||||
in non-rounded, exact values:
|
||||
|
||||
zfs list -p
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
Before deleting a dataset or snapshot, perform a dry run using the -n
|
||||
parameter. This is to make sure you really want to delete just that
|
||||
dataset/snapshot and not any dependent ones. ZFS will display the resulting
|
||||
action when -n is combined with the -v option without actually performing
|
||||
it:
|
||||
|
||||
zfs destroy -rvn mypool@mysnap
|
||||
|
||||
Once you are sure this is exactly what you intend to do, remove the -n
|
||||
parameter to execute the destroy operation.
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
You can delete a range of ZFS snapshots (a-z) in multiple ways.
|
||||
The following will delete d and all earlier snapshots:
|
||||
|
||||
zfs destroy mypool/data@%d
|
||||
|
||||
To delete d and all later snapshots:
|
||||
|
||||
zfs destroy mypool/data@d%
|
||||
|
||||
To delete all dataset snapshots:
|
||||
|
||||
zfs destroy mypool/data@%
|
||||
|
||||
Make sure to let ZFS perform a dry run (-n option) first and display (-v) what
|
||||
it would do to confirm that the delete operation is removing exactly what you
|
||||
intended.
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
To set a custom ZFS property on the mypool pool, you need to provide it
|
||||
using the "key1:key2=value" syntax, where the colon (:) is used as the
|
||||
separator and identifier from the built-in ZFS properties:
|
||||
|
||||
# zfs set warranty:expires=2038-01-19 mypool
|
||||
|
||||
The custom property is applied to all datasets and can be queried like any
|
||||
built-in properties using zfs get:
|
||||
|
||||
zfs get warranty:expires mypool
|
||||
|
||||
To reset the value of a custom property, use the inherit subcommand:
|
||||
|
||||
# zfs inherit warranty:expires mypool
|
||||
|
||||
Removing a custom property from a pool is done using the -r flag to the
|
||||
"zfs inherit" command:
|
||||
|
||||
# zfs inherit -r warranty:expires mypool
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
To delete a range of ZFS snapshots, use the % (percent) character after the
|
||||
full path to the first snapshot that should be included. For example, to
|
||||
simulate deleting snapshots a through (including) d, use this command:
|
||||
|
||||
# zfs destroy -rvn mypool/tmp@a%d
|
||||
|
||||
Once you are sure that this is what you want, remove the -n option:
|
||||
|
||||
# zfs destroy -rv mypool/tmp@a%d
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
You can prevent the removal of a ZFS snapshot by using the hold subcommand.
|
||||
For example, to prevent the snapshot called milestone from deletion, run the
|
||||
following command:
|
||||
|
||||
# zfs hold milestone_hold mypool/projects@my_milestone
|
||||
|
||||
The "zfs holds" command will list all current snapshots that are protected
|
||||
this way (-r for a recursive list):
|
||||
|
||||
# zfs holds -r mypool
|
||||
|
||||
The TIMESTAMP column in the output of the above command is from when the
|
||||
hold was created, not the snapshot it holds. The "zfs destroy" command will
|
||||
echo a "dataset is busy" message on the console when it encounters a hold.
|
||||
Use "zfs release" to release the hold on the snapshot:
|
||||
|
||||
# zfs release milestone_hold mypool/projects@my_milestone
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
A user "sender" needs the following permissions set to send a ZFS dataset:
|
||||
|
||||
# zfs allow -u sender send,snapshot txpool
|
||||
|
||||
On the receiving side, the user "receiver" requires these permissions:
|
||||
|
||||
# zfs allow -u receiver compression,mountpoint,mount,create,receive rxpool
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
Don't let your zpool fill up completely by creating a dataset with
|
||||
reservation.
|
||||
|
||||
# zfs create -o refreservation=<5% of total pool space> <poolname>/reserved
|
||||
|
||||
You can always shrink the reserve if you need the space, but your pool will
|
||||
always have space left this way.
|
||||
|
||||
-- Benedict Reuschling <bcr@FreeBSD.org>
|
||||
%
|
||||
|
||||
Loading…
Reference in New Issue
Block a user