| Message ID | 20200806072906.358641-1-wqu@suse.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | btrfs-progs: docs: update the stability and performance status of quota | expand |
On Thu, Aug 6, 2020 at 3:32 AM Qu Wenruo <wqu@suse.com> wrote: > > There are a lot of enhancement to btrfs quota through v5.x releases. > > Now btrfs quota is more stable than it used to be. > > So update the man page to relect this. > > Signed-off-by: Qu Wenruo <wqu@suse.com> > --- > Documentation/btrfs-quota.asciidoc | 43 +++++++++++++++++++++++++----- > 1 file changed, 37 insertions(+), 6 deletions(-) > > diff --git a/Documentation/btrfs-quota.asciidoc b/Documentation/btrfs-quota.asciidoc > index 85ebf729c2fa..1c032f11d001 100644 > --- a/Documentation/btrfs-quota.asciidoc > +++ b/Documentation/btrfs-quota.asciidoc > @@ -23,16 +23,47 @@ PERFORMANCE IMPLICATIONS > ~~~~~~~~~~~~~~~~~~~~~~~~ > > When quotas are activated, they affect all extent processing, which takes a > -performance hit. Activation of qgroups is not recommended unless the user > -intends to actually use them. > +performance hit. > + > +Under most cases, the performance hit should be more or less acceptable for > +root fs usage. > + > +There used to be a huge performance hit for balance with quota enabled. > +That problem is solved since v5.4 kernel. > > STABILITY STATUS > ~~~~~~~~~~~~~~~~ > > -The qgroup implementation has turned out to be quite difficult as it affects > -the core of the filesystem operation. Qgroup users have hit various corner cases > -over time, such as incorrect accounting or system instability. The situation is > -gradually improving and issues found and fixed. > +Btrfs quota has different stablity for different functionality: > + > +Extent accounting > +^^^^^^^^^^^^^^^^^ > + > +Pretty stable, there aren't many bugs (if any) affecting the extent accounting > +through v5.x release cycles. > + > +Thus if users just want referenced/exclusive usage of each subvolume, it > +should be safe to use. > + > +Limit > +^^^^^ > + > +Should be near stable since v5.9. > + > +There used to be some bugs causing early EDQUOT errors before v5.9. > +But v5.9 should solve them quite well, along with extra safe nets catching any > +reserved space leakage. > + > +Corner cases and small fixes may pop up time by time, but the core limit > +functionality should be in good shape since v5.9. > + > +Multi-level qgroups > +^^^^^^^^^^^^^^^^^^^ > + > +Needs more testing. Although the core extent accounting should also work well > +for higher level qgroups, we don't have good enough test coverage yet. > + > +Thus extra testing and bug reports are welcomed. > > HIERARCHICAL QUOTA GROUP CONCEPTS > --------------------------------- > -- > 2.28.0 > LGTM. Reviewed-by: Neal Gompa <ngompa13@gmail.com>
On 8/6/20 9:29 AM, Qu Wenruo wrote: > There are a lot of enhancement to btrfs quota through v5.x releases. > > Now btrfs quota is more stable than it used to be. > > So update the man page to relect this. > > Signed-off-by: Qu Wenruo <wqu@suse.com> > --- > Documentation/btrfs-quota.asciidoc | 43 +++++++++++++++++++++++++----- > 1 file changed, 37 insertions(+), 6 deletions(-) > > diff --git a/Documentation/btrfs-quota.asciidoc b/Documentation/btrfs-quota.asciidoc > index 85ebf729c2fa..1c032f11d001 100644 > --- a/Documentation/btrfs-quota.asciidoc > +++ b/Documentation/btrfs-quota.asciidoc > @@ -23,16 +23,47 @@ PERFORMANCE IMPLICATIONS > ~~~~~~~~~~~~~~~~~~~~~~~~ > > When quotas are activated, they affect all extent processing, which takes a > -performance hit. Activation of qgroups is not recommended unless the user > -intends to actually use them. > +performance hit. > + > +Under most cases, the performance hit should be more or less acceptable for > +root fs usage. > + > +There used to be a huge performance hit for balance with quota enabled. > +That problem is solved since v5.4 kernel. > > STABILITY STATUS > ~~~~~~~~~~~~~~~~ > > -The qgroup implementation has turned out to be quite difficult as it affects > -the core of the filesystem operation. Qgroup users have hit various corner cases > -over time, such as incorrect accounting or system instability. The situation is > -gradually improving and issues found and fixed. > +Btrfs quota has different stablity for different functionality: > + > +Extent accounting > +^^^^^^^^^^^^^^^^^ > + > +Pretty stable, there aren't many bugs (if any) affecting the extent accounting > +through v5.x release cycles. > + > +Thus if users just want referenced/exclusive usage of each subvolume, it > +should be safe to use. > + > +Limit > +^^^^^ > + > +Should be near stable since v5.9. Is it correct v5.9 ? We are at v5.8.... > + > +There used to be some bugs causing early EDQUOT errors before v5.9. > +But v5.9 should solve them quite well, along with extra safe nets catching any > +reserved space leakage. > + > +Corner cases and small fixes may pop up time by time, but the core limit > +functionality should be in good shape since v5.9. > + > +Multi-level qgroups > +^^^^^^^^^^^^^^^^^^^ > + > +Needs more testing. Although the core extent accounting should also work well > +for higher level qgroups, we don't have good enough test coverage yet. > + > +Thus extra testing and bug reports are welcomed. > > HIERARCHICAL QUOTA GROUP CONCEPTS > --------------------------------- >
On 2020/8/7 上午1:47, Goffredo Baroncelli wrote: > On 8/6/20 9:29 AM, Qu Wenruo wrote: >> There are a lot of enhancement to btrfs quota through v5.x releases. >> >> Now btrfs quota is more stable than it used to be. >> >> So update the man page to relect this. >> >> Signed-off-by: Qu Wenruo <wqu@suse.com> >> --- >> Documentation/btrfs-quota.asciidoc | 43 +++++++++++++++++++++++++----- >> 1 file changed, 37 insertions(+), 6 deletions(-) >> >> diff --git a/Documentation/btrfs-quota.asciidoc >> b/Documentation/btrfs-quota.asciidoc >> index 85ebf729c2fa..1c032f11d001 100644 >> --- a/Documentation/btrfs-quota.asciidoc >> +++ b/Documentation/btrfs-quota.asciidoc >> @@ -23,16 +23,47 @@ PERFORMANCE IMPLICATIONS >> ~~~~~~~~~~~~~~~~~~~~~~~~ >> When quotas are activated, they affect all extent processing, >> which takes a >> -performance hit. Activation of qgroups is not recommended unless the >> user >> -intends to actually use them. >> +performance hit. >> + >> +Under most cases, the performance hit should be more or less >> acceptable for >> +root fs usage. >> + >> +There used to be a huge performance hit for balance with quota enabled. >> +That problem is solved since v5.4 kernel. >> STABILITY STATUS >> ~~~~~~~~~~~~~~~~ >> -The qgroup implementation has turned out to be quite difficult as >> it affects >> -the core of the filesystem operation. Qgroup users have hit various >> corner cases >> -over time, such as incorrect accounting or system instability. The >> situation is >> -gradually improving and issues found and fixed. >> +Btrfs quota has different stablity for different functionality: >> + >> +Extent accounting >> +^^^^^^^^^^^^^^^^^ >> + >> +Pretty stable, there aren't many bugs (if any) affecting the extent >> accounting >> +through v5.x release cycles. >> + >> +Thus if users just want referenced/exclusive usage of each subvolume, it >> +should be safe to use. >> + >> +Limit >> +^^^^^ >> + >> +Should be near stable since v5.9. > > Is it correct v5.9 ? We are at v5.8.... Yes. The fixes are in the latest torvalds' master branch, which will become v5.9-rc1 soon. Thanks, Qu > >> + >> +There used to be some bugs causing early EDQUOT errors before v5.9. >> +But v5.9 should solve them quite well, along with extra safe nets >> catching any >> +reserved space leakage. >> + >> +Corner cases and small fixes may pop up time by time, but the core limit >> +functionality should be in good shape since v5.9. >> + >> +Multi-level qgroups >> +^^^^^^^^^^^^^^^^^^^ >> + >> +Needs more testing. Although the core extent accounting should also >> work well >> +for higher level qgroups, we don't have good enough test coverage yet. >> + >> +Thus extra testing and bug reports are welcomed. >> HIERARCHICAL QUOTA GROUP CONCEPTS >> --------------------------------- >> > >
diff --git a/Documentation/btrfs-quota.asciidoc b/Documentation/btrfs-quota.asciidoc index 85ebf729c2fa..1c032f11d001 100644 --- a/Documentation/btrfs-quota.asciidoc +++ b/Documentation/btrfs-quota.asciidoc @@ -23,16 +23,47 @@ PERFORMANCE IMPLICATIONS ~~~~~~~~~~~~~~~~~~~~~~~~ When quotas are activated, they affect all extent processing, which takes a -performance hit. Activation of qgroups is not recommended unless the user -intends to actually use them. +performance hit. + +Under most cases, the performance hit should be more or less acceptable for +root fs usage. + +There used to be a huge performance hit for balance with quota enabled. +That problem is solved since v5.4 kernel. STABILITY STATUS ~~~~~~~~~~~~~~~~ -The qgroup implementation has turned out to be quite difficult as it affects -the core of the filesystem operation. Qgroup users have hit various corner cases -over time, such as incorrect accounting or system instability. The situation is -gradually improving and issues found and fixed. +Btrfs quota has different stablity for different functionality: + +Extent accounting +^^^^^^^^^^^^^^^^^ + +Pretty stable, there aren't many bugs (if any) affecting the extent accounting +through v5.x release cycles. + +Thus if users just want referenced/exclusive usage of each subvolume, it +should be safe to use. + +Limit +^^^^^ + +Should be near stable since v5.9. + +There used to be some bugs causing early EDQUOT errors before v5.9. +But v5.9 should solve them quite well, along with extra safe nets catching any +reserved space leakage. + +Corner cases and small fixes may pop up time by time, but the core limit +functionality should be in good shape since v5.9. + +Multi-level qgroups +^^^^^^^^^^^^^^^^^^^ + +Needs more testing. Although the core extent accounting should also work well +for higher level qgroups, we don't have good enough test coverage yet. + +Thus extra testing and bug reports are welcomed. HIERARCHICAL QUOTA GROUP CONCEPTS ---------------------------------
There are a lot of enhancement to btrfs quota through v5.x releases. Now btrfs quota is more stable than it used to be. So update the man page to relect this. Signed-off-by: Qu Wenruo <wqu@suse.com> --- Documentation/btrfs-quota.asciidoc | 43 +++++++++++++++++++++++++----- 1 file changed, 37 insertions(+), 6 deletions(-)