Re: Upgrade 3.2.7 to 3.4.2

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



 I followed those directions deleting the trusted.gfid and trusted.glusterfs.volume-id. I still was not able to set the attribute. This is what I ended up doing correct me if I am wrong. Luckily this is not a production volume, contains backups


This volume had 4 bricks, distributed volume. I removed 3 bricks
gluster>volume remove-brick data02 ip1:/brick1
gluster>volume remove-brick data02 ip1:/brick2
gluster>volume remove-brick data02 ip2:/brick3

wouldn't let me remove the last brick

force added the 3 bricks back
gluster>volume add-brick data02 ip1:/brick1 force
gluster>volume add-brick data02 ip1:/brick2 force
gluster>volume add-brick data02 ip2:/brick3 force

then remove the brick that I couldn't
gluster>volume remove-brick data02 ip2:/brick4

add it back
gluster>volume add-brick data02 ip2:/brick4 force

check that the bricks now have the extended attribute
getfattr -n trusted.glusterfs.volume-id /brick1
# file: brick1
trusted.glusterfs.volume-id=0skluoKzppQx6hE+md8/oVlA==

volume started now no problem



From: "Bernhard Glomm" <bernhard.glomm@xxxxxxxxxxx>
To: knasveschuk@xxxxxxx, gluster-users@xxxxxxxxxxx
Sent: Thursday, March 13, 2014 1:28:21 PM
Subject: Re: Upgrade 3.2.7 to 3.4.2

Did u see this?

http://joejulian.name/blog/glusterfs-path-or-a-prefix-of-it-is-already-part-of-a-volume/


Am 13.03.2014 17:39:10, schrieb Kent Nasveschuk:
CentOS 6.5, repo glusterfs-epel.repo

I used both of these as guide to upgrade

My problem seems to be with trusted.glusterfs.volume-id that the 3.2.7 version doesn't use

I can't seem to set the hex value using any recommendations from previous posts. There seems to be a bug with this.

For each brick:
ex. setfattr -n trusted.glusterfs.volume-id -v 0x925ba82b-3a69-431e-a113-e99df3fa1594 /data02a (and the others)

The above doesn't work "bad input encoding" also \0x

I can only get it to take ascii, at which point I can't start the volume, complains: "Failed to get extended attribute trusted.glusterfs.volume-id for ...... Numerical result out of range"

Is there a workaround for this? Delete volume add bricks back? Any help would be appreciated.



Kent

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://supercolony.gluster.org/mailman/listinfo/gluster-users



_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://supercolony.gluster.org/mailman/listinfo/gluster-users

[Index of Archives]     [Gluster Development]     [Linux Filesytems Development]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux