Hi, I'm trying to use the reset-brick command, but it's not completely clear to me> > Introducing reset-brick command > > /Notes for users:/ The reset-brick command provides support to > reformat/replace the disk(s) represented by a brick within a volume. > This is helpful when a disk goes bad etc >That's what I need, the use case is a disk goes bad on a disperse gluster node and we want to replace it with a new disk> > Start reset process - > > |gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH start |This works, I can see in gluster volume status the brick is not there anymore> > The above command kills the respective brick process. Now the brick > can be reformatted. > > To restart the brick after modifying configuration - > > |gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH > HOSTNAME:BRICKPATH commit | > > If the brick was killed to replace the brick with same brick path, > restart with following command - > > |gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH > HOSTNAME:BRICKPATH commit force |This fails, I unmounted the gluster path, formatted a fresh disk, mounted it on the old mount point and created the brick subdir on it. gluster volume reset-brick glustervol gluster1:/gluster/brick1/glusterbrick1 gluster1:/gluster/brick1/glusterbrick1 commit force volume reset-brick: failed: Source brick must be stopped. Please use gluster volume reset-brick <volname> <dst-brick> start. Why would I even need to specify the "|HOSTNAME:BRICKPATH|" twice? I just want to replace the disk and get it back into the volume. Met vriendelijke groet, With kind regards, Jorick Astrego Netbulae Virtualization Experts ---------------- Tel: 053 20 30 270 info at netbulae.eu Staalsteden 4-3A KvK 08198180 Fax: 053 20 30 271 www.netbulae.eu 7547 TA Enschede BTW NL821234584B01 ---------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20171211/5c2f948b/attachment.html>
Hi Jorick, 1 - Why would I even need to specify the " HOSTNAME:BRICKPATH " twice? I just want to replace the disk and get it back into the volume. Reset brick command can be used in different scenarios. One more case could be where you just want to change the host name to IP address of that node of bricks. In this case also you will follow the same steps but just have to provide IP address gluster volume reset-brick glustervol gluster1:/gluster/brick1/glusterbrick1 "gluster1 IP address" :/gluster/brick1/glusterbrick1 commit force Now as we have this command for different cases, to keep uniformity of the command, we chose to provide brick path twice. Coming to your case, I think you followed all the steps correctly and it should be successful. Please provide guster volume status of the volume and also try to use "commit force" and only "commit" and let us know the result. You may have to raise a bug if it does not work so be prepared to provide glusterd logs in /var/log/glusterfs/ -- Ashish ----- Original Message ----- From: "Jorick Astrego" <jorick at netbulae.eu> To: gluster-users at gluster.org Sent: Monday, December 11, 2017 7:32:53 PM Subject: [Gluster-users] reset-brick command questions Hi, I'm trying to use the reset-brick command, but it's not completely clear to me Introducing reset-brick command Notes for users: The reset-brick command provides support to reformat/replace the disk(s) represented by a brick within a volume. This is helpful when a disk goes bad etc That's what I need, the use case is a disk goes bad on a disperse gluster node and we want to replace it with a new disk <blockquote> Start reset process - gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH start </blockquote> This works, I can see in gluster volume status the brick is not there anymore <blockquote> The above command kills the respective brick process. Now the brick can be reformatted. To restart the brick after modifying configuration - gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH HOSTNAME:BRICKPATH commit </blockquote> <blockquote> If the brick was killed to replace the brick with same brick path, restart with following command - gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH HOSTNAME:BRICKPATH commit force </blockquote> This fails, I unmounted the gluster path, formatted a fresh disk, mounted it on the old mount point and created the brick subdir on it. <blockquote> gluster volume reset-brick glustervol gluster1:/gluster/brick1/glusterbrick1 gluster1:/gluster/brick1/glusterbrick1 commit force volume reset-brick: failed: Source brick must be stopped. Please use gluster volume reset-brick <volname> <dst-brick> start. </blockquote> Why would I even need to specify the " HOSTNAME:BRICKPATH " twice? I just want to replace the disk and get it back into the volume. <blockquote> </blockquote> Met vriendelijke groet, With kind regards, Jorick Astrego Netbulae Virtualization Experts Tel: 053 20 30 270 info at netbulae.eu Staalsteden 4-3A KvK 08198180 Fax: 053 20 30 271 www.netbulae.eu 7547 TA Enschede BTW NL821234584B01 _______________________________________________ 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/20171211/0803a17c/attachment.html>
Hi Ashish, Thanks for the information. Weird, I did another test and it works fine now. Maybe I did something wrong the first time... I will try it a couple more times to see if I can reproduce.. Regards, Jorick On 12/12/2017 03:00 AM, Ashish Pandey wrote:> Hi Jorick, > > 1 - Why would I even need to specify the "|HOSTNAME:BRICKPATH|" twice? > I just want to replace the disk and get it back into the volume. > > Reset brick command can be used in different scenarios. One more case > could be where you just want to change the host name to IP address of > that node of bricks. > In this case also you will follow the same steps but just have to > provide IP address > > gluster volume reset-brick glustervol > gluster1:/gluster/brick1/glusterbrick1 */"gluster1 IP > address"/*:/gluster/brick1/glusterbrick1 commit force > > Now as we have this command for different cases, to keep uniformity of > the command, we chose to provide brick path twice. > > Coming to your case, I think you followed all the steps correctly and > it should be successful. > Please provide guster volume status of the volume and also try to use > "commit force" and only "commit" and let us know the result. > You may have to raise a bug if it does not work so be prepared to > provide glusterd logs in /var/log/glusterfs/ > > -- > Ashish > > > ------------------------------------------------------------------------ > *From: *"Jorick Astrego" <jorick at netbulae.eu> > *To: *gluster-users at gluster.org > *Sent: *Monday, December 11, 2017 7:32:53 PM > *Subject: *[Gluster-users] reset-brick command questions > > Hi, > > I'm trying to use the reset-brick command, but it's not completely > clear to me > > > Introducing reset-brick command > > /Notes for users:/ The reset-brick command provides support to > reformat/replace the disk(s) represented by a brick within a > volume. This is helpful when a disk goes bad etc > > That's what I need, the use case is a disk goes bad on a disperse > gluster node and we want to replace it with a new disk > > Start reset process - > > |gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH start | > > This works, I can see in gluster volume status the brick is not there > anymore > > The above command kills the respective brick process. Now the > brick can be reformatted. > > To restart the brick after modifying configuration - > > |gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH > HOSTNAME:BRICKPATH commit | > > If the brick was killed to replace the brick with same brick path, > restart with following command - > > |gluster volume reset-brick VOLNAME HOSTNAME:BRICKPATH > HOSTNAME:BRICKPATH commit force | > > This fails, I unmounted the gluster path, formatted a fresh disk, > mounted it on the old mount point and created the brick subdir on it. > > gluster volume reset-brick glustervol > gluster1:/gluster/brick1/glusterbrick1 > gluster1:/gluster/brick1/glusterbrick1 commit force > > volume reset-brick: failed: Source brick must be stopped. Please > use gluster volume reset-brick <volname> <dst-brick> start. > > Why would I even need to specify the "|HOSTNAME:BRICKPATH|" twice? I > just want to replace the disk and get it back into the volume. > > > > > > > Met vriendelijke groet, With kind regards, > > Jorick Astrego > * > Netbulae Virtualization Experts * > ------------------------------------------------------------------------ > Tel: 053 20 30 270 info at netbulae.eu Staalsteden 4-3A KvK 08198180 > Fax: 053 20 30 271 www.netbulae.eu 7547 TA Enschede BTW NL821234584B01 > > > ------------------------------------------------------------------------ > > > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > http://lists.gluster.org/mailman/listinfo/gluster-users >Met vriendelijke groet, With kind regards, Jorick Astrego Netbulae Virtualization Experts ---------------- Tel: 053 20 30 270 info at netbulae.eu Staalsteden 4-3A KvK 08198180 Fax: 053 20 30 271 www.netbulae.eu 7547 TA Enschede BTW NL821234584B01 ---------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20171218/2f2c477e/attachment.html>