On 2020/6/19 下午5:39, David Sterba wrote: > On Fri, Jun 19, 2020 at 09:59:46AM +0800, Qu Wenruo wrote: >> This patch will add the following sysfs interface: >> /sys/fs/btrfs/<UUID>/qgroups/<qgroup_id>/rfer >> /sys/fs/btrfs/<UUID>/qgroups/<qgroup_id>/excl >> /sys/fs/btrfs/<UUID>/qgroups/<qgroup_id>/max_rfer >> /sys/fs/btrfs/<UUID>/qgroups/<qgroup_id>/max_excl >> /sys/fs/btrfs/<UUID>/qgroups/<qgroup_id>/lim_flags >> ^^^ Above are already in "btrfs qgroup show" command output ^^^ >> >> /sys/fs/btrfs/<UUID>/qgroups/<qgroup_id>/rsv_data >> /sys/fs/btrfs/<UUID>/qgroups/<qgroup_id>/rsv_meta_pertrans >> /sys/fs/btrfs/<UUID>/qgroups/<qgroup_id>/rsv_meta_prealloc >> >> The last 3 rsv related members are not visible to users, but can be very >> useful to debug qgroup limit related bugs. >> >> Also, to avoid '/' used in <qgroup_id>, the seperator between qgroup >> level and qgroup id is changed to '_'. >> >> The interface is not hidden behind 'debug' as I want this interface to >> be included into production build so we could have an easier life to >> debug qgroup rsv related bugs. > > But why do you want to export it to sysfs at all? > There is an internal report where user is not that co-operative to do more experiments, but insists on providing more debugging info. And since they don't want to unset qgroup limit, nor unmount their root fs to make sure the latest qgroup data rsv safenet catches leakage, the last method to debug strange early EDQUOT is to export rsv info to user space. And, for most users, the new interface won't bother anyone, but when things go wrong and the user is not cooperative, such interface can save us a lot of time. Thanks, Qu
Attachment:
signature.asc
Description: OpenPGP digital signature
