Hello! We are running distributed volume that contains 7 bricks. Volume is mounted using native fuse client. After an unexpected system reboot, some files are disappeared from fuse mount point but still available on the bricks. The way it disappeared confusing me a lot. I can't see certain directories using ls -la but, at the same time, can cd into the missed directory. I can rename the invisible directory and it becomes accessible. When I renamed it back to the original name, it becomes invisible. I also tried to mount the same volume into another location and run ls hoping that selfheal will fix the problem. Unfortunately, it did not. Is there a way to bring our storage to normal? glusterfs 3.8.8 built on Jan 11 2017 16:33:17 Serg Gulko -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20180403/7f9065af/attachment.html>
http://lists.gluster.org/pipermail/gluster-users/2018-April/033811.html On Tue, Apr 3, 2018 at 6:43 PM, Serg Gulko <s.gulko at gmail.com> wrote:> Hello! > > We are running distributed volume that contains 7 bricks. > Volume is mounted using native fuse client. > > After an unexpected system reboot, some files are disappeared from fuse > mount point but still available on the bricks. > > The way it disappeared confusing me a lot. I can't see certain directories > using ls -la but, at the same time, can cd into the missed directory. I > can rename the invisible directory and it becomes accessible. When I > renamed it back to the original name, it becomes invisible. > > I also tried to mount the same volume into another location and run ls > hoping that selfheal will fix the problem. Unfortunately, it did not. > > Is there a way to bring our storage to normal? > > glusterfs 3.8.8 built on Jan 11 2017 16:33:17 > > Serg Gulko > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20180403/2f53fe6a/attachment.html>
Thanks for the reply, Vlad! Now is the trickiest part: gluster volume get volume_1 cluster.readdir-optimize Option Value ------ ----- cluster.readdir-optimize off Serg On Tue, Apr 3, 2018 at 9:20 PM, Vlad Kopylov <vladkopy at gmail.com> wrote:> http://lists.gluster.org/pipermail/gluster-users/2018-April/033811.html > > On Tue, Apr 3, 2018 at 6:43 PM, Serg Gulko <s.gulko at gmail.com> wrote: > >> Hello! >> >> We are running distributed volume that contains 7 bricks. >> Volume is mounted using native fuse client. >> >> After an unexpected system reboot, some files are disappeared from fuse >> mount point but still available on the bricks. >> >> The way it disappeared confusing me a lot. I can't see certain >> directories using ls -la but, at the same time, can cd into the missed >> directory. I can rename the invisible directory and it becomes accessible. >> When I renamed it back to the original name, it becomes invisible. >> >> I also tried to mount the same volume into another location and run ls >> hoping that selfheal will fix the problem. Unfortunately, it did not. >> >> Is there a way to bring our storage to normal? >> >> glusterfs 3.8.8 built on Jan 11 2017 16:33:17 >> >> Serg Gulko >> >> _______________________________________________ >> Gluster-users mailing list >> Gluster-users at gluster.org >> http://lists.gluster.org/mailman/listinfo/gluster-users >> > >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20180404/a59e52fd/attachment.html>
Raghavendra Gowdappa
2018-Apr-04 04:56 UTC
[Gluster-users] Invisible files and directories
On Wed, Apr 4, 2018 at 4:13 AM, Serg Gulko <s.gulko at gmail.com> wrote:> Hello! > > We are running distributed volume that contains 7 bricks. > Volume is mounted using native fuse client. > > After an unexpected system reboot, some files are disappeared from fuse > mount point but still available on the bricks. > > The way it disappeared confusing me a lot. I can't see certain directories > using ls -la but, at the same time, can cd into the missed directory. I > can rename the invisible directory and it becomes accessible. When I > renamed it back to the original name, it becomes invisible. > > I also tried to mount the same volume into another location and run ls > hoping that selfheal will fix the problem. Unfortunately, it did not. > > Is there a way to bring our storage to normal? >Can you check whether turning off option performance.readdir-ahead helps?> glusterfs 3.8.8 built on Jan 11 2017 16:33:17 > > Serg Gulko > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20180404/1f12622d/attachment.html>
Hello! Unfortunately no. Directory still not listed using ls -la, but I can cd into. I can rename it and it becomes available when I rename it back to the original name it's disappeared again. On Wed, Apr 4, 2018 at 12:56 AM, Raghavendra Gowdappa <rgowdapp at redhat.com> wrote:> > > On Wed, Apr 4, 2018 at 4:13 AM, Serg Gulko <s.gulko at gmail.com> wrote: > >> Hello! >> >> We are running distributed volume that contains 7 bricks. >> Volume is mounted using native fuse client. >> >> After an unexpected system reboot, some files are disappeared from fuse >> mount point but still available on the bricks. >> >> The way it disappeared confusing me a lot. I can't see certain >> directories using ls -la but, at the same time, can cd into the missed >> directory. I can rename the invisible directory and it becomes accessible. >> When I renamed it back to the original name, it becomes invisible. >> >> I also tried to mount the same volume into another location and run ls >> hoping that selfheal will fix the problem. Unfortunately, it did not. >> >> Is there a way to bring our storage to normal? >> > > Can you check whether turning off option performance.readdir-ahead helps? > > >> glusterfs 3.8.8 built on Jan 11 2017 16:33:17 >> >> Serg Gulko >> >> _______________________________________________ >> Gluster-users mailing list >> Gluster-users at gluster.org >> http://lists.gluster.org/mailman/listinfo/gluster-users >> > >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20180404/5d4fa72a/attachment.html>