Re: [Gluster-users] Suggested method for replacing an entire node

2015-10-08 Thread Alastair Neil 
I think you should back up /var/lib/glusterd and then restore it after the
reinstall and installation of glusterfs packages.  Assuming the node will
have the same hostname and ip addresses and you are installing the same
version gluster bits, I think it should be fine.  I am assuming you are not
using ssl for the connections if so you will need to back up the keys for
that too.

-Alastair

On 8 October 2015 at 00:12, Atin Mukherjee  wrote:

>
>
> On 10/07/2015 10:28 PM, Gene Liverman wrote:
> > I want to replace my existing CentOS 6 nodes with CentOS 7 ones. Is
> > there a recommended way to go about this from the perspective of
> > Gluster? I am running a 3 node replicated cluster (3 servers each with 1
> > brick). In case it makes a difference, my bricks are on separate drives
> > formatted as XFS so it is possible that I can do my OS reinstall without
> > wiping out the data on two nodes (the third had a hardware failure so it
> > will be fresh from the ground up).
> That's possible. You could do the re-installation one at a time. Once
> the node comes back online self heal daemon will take care of healing
> the data. AFR team can correct me if I am wrong.
>
> Thanks,
> Atin
> >
> >
> >
> >
> > Thanks,
> > *Gene Liverman*
> > Systems Integration Architect
> > Information Technology Services
> > University of West Georgia
> > glive...@westga.edu 
> >
> > ITS: Making Technology Work for You!
> >
> >
> > ___
> > Gluster-users mailing list
> > Gluster-users@gluster.org
> > http://www.gluster.org/mailman/listinfo/gluster-users
> >
> ___
> Gluster-users mailing list
> Gluster-users@gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users
>
___
Gluster-users mailing list
Gluster-users@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-users

Re: [Gluster-users] Suggested method for replacing an entire node

2015-10-08 Thread Alastair Neil 
Ahh that is good to know.

On 8 October 2015 at 09:50, Atin Mukherjee 
wrote:

> -Atin
> Sent from one plus one
> On Oct 8, 2015 7:17 PM, "Alastair Neil"  wrote:
> >
> > I think you should back up /var/lib/glusterd and then restore it after
> the reinstall and installation of glusterfs packages.  Assuming the node
> will have the same hostname and ip addresses and you are installing the
> same version gluster bits, I think it should be fine.  I am assuming you
> are not using ssl for the connections if so you will need to back up the
> keys for that too.
> If the same machine is used with out hostname/ IP change, backing up
> glusterd configuration *is not* needed as syncing the configuration will be
> taken care peer handshaking.
>
> >
> > -Alastair
> >
> > On 8 October 2015 at 00:12, Atin Mukherjee  wrote:
> >>
> >>
> >>
> >> On 10/07/2015 10:28 PM, Gene Liverman wrote:
> >> > I want to replace my existing CentOS 6 nodes with CentOS 7 ones. Is
> >> > there a recommended way to go about this from the perspective of
> >> > Gluster? I am running a 3 node replicated cluster (3 servers each
> with 1
> >> > brick). In case it makes a difference, my bricks are on separate
> drives
> >> > formatted as XFS so it is possible that I can do my OS reinstall
> without
> >> > wiping out the data on two nodes (the third had a hardware failure so
> it
> >> > will be fresh from the ground up).
> >> That's possible. You could do the re-installation one at a time. Once
> >> the node comes back online self heal daemon will take care of healing
> >> the data. AFR team can correct me if I am wrong.
> >>
> >> Thanks,
> >> Atin
> >> >
> >> >
> >> >
> >> >
> >> > Thanks,
> >> > *Gene Liverman*
> >> > Systems Integration Architect
> >> > Information Technology Services
> >> > University of West Georgia
> >> > glive...@westga.edu 
> >> >
> >> > ITS: Making Technology Work for You!
> >> >
> >> >
> >> > ___
> >> > Gluster-users mailing list
> >> > Gluster-users@gluster.org
> >> > http://www.gluster.org/mailman/listinfo/gluster-users
> >> >
> >> ___
> >> Gluster-users mailing list
> >> Gluster-users@gluster.org
> >> http://www.gluster.org/mailman/listinfo/gluster-users
> >
> >
> >
> > ___
> > Gluster-users mailing list
> > Gluster-users@gluster.org
> > http://www.gluster.org/mailman/listinfo/gluster-users
>
>
___
Gluster-users mailing list
Gluster-users@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-users

Re: [Gluster-users] Suggested method for replacing an entire node

2015-10-08 Thread Atin Mukherjee 
-Atin
Sent from one plus one
On Oct 8, 2015 7:17 PM, "Alastair Neil"  wrote:
>
> I think you should back up /var/lib/glusterd and then restore it after
the reinstall and installation of glusterfs packages.  Assuming the node
will have the same hostname and ip addresses and you are installing the
same version gluster bits, I think it should be fine.  I am assuming you
are not using ssl for the connections if so you will need to back up the
keys for that too.
If the same machine is used with out hostname/ IP change, backing up
glusterd configuration *is not* needed as syncing the configuration will be
taken care peer handshaking.
>
> -Alastair
>
> On 8 October 2015 at 00:12, Atin Mukherjee  wrote:
>>
>>
>>
>> On 10/07/2015 10:28 PM, Gene Liverman wrote:
>> > I want to replace my existing CentOS 6 nodes with CentOS 7 ones. Is
>> > there a recommended way to go about this from the perspective of
>> > Gluster? I am running a 3 node replicated cluster (3 servers each with
1
>> > brick). In case it makes a difference, my bricks are on separate drives
>> > formatted as XFS so it is possible that I can do my OS reinstall
without
>> > wiping out the data on two nodes (the third had a hardware failure so
it
>> > will be fresh from the ground up).
>> That's possible. You could do the re-installation one at a time. Once
>> the node comes back online self heal daemon will take care of healing
>> the data. AFR team can correct me if I am wrong.
>>
>> Thanks,
>> Atin
>> >
>> >
>> >
>> >
>> > Thanks,
>> > *Gene Liverman*
>> > Systems Integration Architect
>> > Information Technology Services
>> > University of West Georgia
>> > glive...@westga.edu 
>> >
>> > ITS: Making Technology Work for You!
>> >
>> >
>> > ___
>> > Gluster-users mailing list
>> > Gluster-users@gluster.org
>> > http://www.gluster.org/mailman/listinfo/gluster-users
>> >
>> ___
>> Gluster-users mailing list
>> Gluster-users@gluster.org
>> http://www.gluster.org/mailman/listinfo/gluster-users
>
>
>
> ___
> Gluster-users mailing list
> Gluster-users@gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users
___
Gluster-users mailing list
Gluster-users@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-users

Re: [Gluster-users] Suggested method for replacing an entire node

2015-10-08 Thread Gene Liverman 
 Thanks for all the replies! Just to make sure I have this right, the
following should work for *both* machines with and machines without a
currently populated brick if the name and IP stay the same:

   - reinstall os
   - reinstall gluster software
   - start gluster

Do I need to do any peer probing or anything else? Do I need to do any
brick removal / adding (I'm thinking no but want to make sure)?




Thanks,
*Gene Liverman*
Systems Integration Architect
Information Technology Services
University of West Georgia
glive...@westga.edu

ITS: Making Technology Work for You!



On Thu, Oct 8, 2015 at 9:52 AM, Alastair Neil  wrote:

> Ahh that is good to know.
>
> On 8 October 2015 at 09:50, Atin Mukherjee 
> wrote:
>
>> -Atin
>> Sent from one plus one
>> On Oct 8, 2015 7:17 PM, "Alastair Neil"  wrote:
>> >
>> > I think you should back up /var/lib/glusterd and then restore it after
>> the reinstall and installation of glusterfs packages.  Assuming the node
>> will have the same hostname and ip addresses and you are installing the
>> same version gluster bits, I think it should be fine.  I am assuming you
>> are not using ssl for the connections if so you will need to back up the
>> keys for that too.
>> If the same machine is used with out hostname/ IP change, backing up
>> glusterd configuration *is not* needed as syncing the configuration will be
>> taken care peer handshaking.
>>
>> >
>> > -Alastair
>> >
>> > On 8 October 2015 at 00:12, Atin Mukherjee  wrote:
>> >>
>> >>
>> >>
>> >> On 10/07/2015 10:28 PM, Gene Liverman wrote:
>> >> > I want to replace my existing CentOS 6 nodes with CentOS 7 ones. Is
>> >> > there a recommended way to go about this from the perspective of
>> >> > Gluster? I am running a 3 node replicated cluster (3 servers each
>> with 1
>> >> > brick). In case it makes a difference, my bricks are on separate
>> drives
>> >> > formatted as XFS so it is possible that I can do my OS reinstall
>> without
>> >> > wiping out the data on two nodes (the third had a hardware failure
>> so it
>> >> > will be fresh from the ground up).
>> >> That's possible. You could do the re-installation one at a time. Once
>> >> the node comes back online self heal daemon will take care of healing
>> >> the data. AFR team can correct me if I am wrong.
>> >>
>> >> Thanks,
>> >> Atin
>> >> >
>> >> >
>> >> >
>> >> >
>> >> > Thanks,
>> >> > *Gene Liverman*
>> >> > Systems Integration Architect
>> >> > Information Technology Services
>> >> > University of West Georgia
>> >> > glive...@westga.edu 
>> >> >
>> >> > ITS: Making Technology Work for You!
>> >> >
>> >> >
>> >> > ___
>> >> > Gluster-users mailing list
>> >> > Gluster-users@gluster.org
>> >> > http://www.gluster.org/mailman/listinfo/gluster-users
>> >> >
>> >> ___
>> >> Gluster-users mailing list
>> >> Gluster-users@gluster.org
>> >> http://www.gluster.org/mailman/listinfo/gluster-users
>> >
>> >
>> >
>> > ___
>> > Gluster-users mailing list
>> > Gluster-users@gluster.org
>> > http://www.gluster.org/mailman/listinfo/gluster-users
>>
>>
>
> ___
> Gluster-users mailing list
> Gluster-users@gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users
>
___
Gluster-users mailing list
Gluster-users@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-users

Re: [Gluster-users] Suggested method for replacing an entire node

2015-10-08 Thread Atin Mukherjee 


On 10/08/2015 09:52 PM, Gene Liverman wrote:
>  Thanks for all the replies! Just to make sure I have this right, the
> following should work for *both* machines with and machines without a
> currently populated brick if the name and IP stay the same:
> 
>   * reinstall os
>   * reinstall gluster software
>   * start gluster
> 
> Do I need to do any peer probing or anything else? Do I need to do any
> brick removal / adding (I'm thinking no but want to make sure)?
No you don't.
> 
> 
> 
> 
> Thanks,
> *Gene Liverman*
> Systems Integration Architect
> Information Technology Services
> University of West Georgia
> glive...@westga.edu 
> 
> ITS: Making Technology Work for You!
> 
> 
> 
> On Thu, Oct 8, 2015 at 9:52 AM, Alastair Neil  > wrote:
> 
> Ahh that is good to know.
> 
> On 8 October 2015 at 09:50, Atin Mukherjee
> > wrote:
> 
> -Atin
> Sent from one plus one
> On Oct 8, 2015 7:17 PM, "Alastair Neil"  > wrote:
> >
> > I think you should back up /var/lib/glusterd and then restore it 
> after the reinstall and installation of glusterfs packages.  Assuming the 
> node will have the same hostname and ip addresses and you are installing the 
> same version gluster bits, I think it should be fine.  I am assuming you are 
> not using ssl for the connections if so you will need to back up the keys for 
> that too.
> If the same machine is used with out hostname/ IP change,
> backing up glusterd configuration *is not* needed as syncing the
> configuration will be taken care peer handshaking.
> 
> 
> >
> > -Alastair
> >
> > On 8 October 2015 at 00:12, Atin Mukherjee
> > wrote:
> >>
> >>
> >>
> >> On 10/07/2015 10:28 PM, Gene Liverman wrote:
> >> > I want to replace my existing CentOS 6 nodes with CentOS 7
> ones. Is
> >> > there a recommended way to go about this from the
> perspective of
> >> > Gluster? I am running a 3 node replicated cluster (3
> servers each with 1
> >> > brick). In case it makes a difference, my bricks are on
> separate drives
> >> > formatted as XFS so it is possible that I can do my OS
> reinstall without
> >> > wiping out the data on two nodes (the third had a hardware
> failure so it
> >> > will be fresh from the ground up).
> >> That's possible. You could do the re-installation one at a
> time. Once
> >> the node comes back online self heal daemon will take care of
> healing
> >> the data. AFR team can correct me if I am wrong.
> >>
> >> Thanks,
> >> Atin
> >> >
> >> >
> >> >
> >> >
> >> > Thanks,
> >> > *Gene Liverman*
> >> > Systems Integration Architect
> >> > Information Technology Services
> >> > University of West Georgia
> >> > glive...@westga.edu 
> >
> >> >
> >> > ITS: Making Technology Work for You!
> >> >
> >> >
> >> > ___
> >> > Gluster-users mailing list
> >> > Gluster-users@gluster.org 
> >> > http://www.gluster.org/mailman/listinfo/gluster-users
> >> >
> >> ___
> >> Gluster-users mailing list
> >> Gluster-users@gluster.org 
> >> http://www.gluster.org/mailman/listinfo/gluster-users
> >
> >
> >
> > ___
> > Gluster-users mailing list
> > Gluster-users@gluster.org 
> > http://www.gluster.org/mailman/listinfo/gluster-users
> 
> 
> 
> ___
> Gluster-users mailing list
> Gluster-users@gluster.org 
> http://www.gluster.org/mailman/listinfo/gluster-users
> 
> 
> 
> 
> ___
> Gluster-users mailing list
> Gluster-users@gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users
> 
___
Gluster-users mailing list
Gluster-users@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-users

Re: [Gluster-users] Suggested method for replacing an entire node

2015-10-07 Thread Atin Mukherjee 


On 10/07/2015 10:28 PM, Gene Liverman wrote:
> I want to replace my existing CentOS 6 nodes with CentOS 7 ones. Is
> there a recommended way to go about this from the perspective of
> Gluster? I am running a 3 node replicated cluster (3 servers each with 1
> brick). In case it makes a difference, my bricks are on separate drives
> formatted as XFS so it is possible that I can do my OS reinstall without
> wiping out the data on two nodes (the third had a hardware failure so it
> will be fresh from the ground up).
That's possible. You could do the re-installation one at a time. Once
the node comes back online self heal daemon will take care of healing
the data. AFR team can correct me if I am wrong.

Thanks,
Atin
> 
> 
> 
> 
> Thanks,
> *Gene Liverman*
> Systems Integration Architect
> Information Technology Services
> University of West Georgia
> glive...@westga.edu 
> 
> ITS: Making Technology Work for You!
> 
> 
> ___
> Gluster-users mailing list
> Gluster-users@gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users
> 
___
Gluster-users mailing list
Gluster-users@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-users

[Gluster-users] Suggested method for replacing an entire node

2015-10-07 Thread Gene Liverman 
I want to replace my existing CentOS 6 nodes with CentOS 7 ones. Is there a
recommended way to go about this from the perspective of Gluster? I am
running a 3 node replicated cluster (3 servers each with 1 brick). In case
it makes a difference, my bricks are on separate drives formatted as XFS so
it is possible that I can do my OS reinstall without wiping out the data on
two nodes (the third had a hardware failure so it will be fresh from the
ground up).




Thanks,
*Gene Liverman*
Systems Integration Architect
Information Technology Services
University of West Georgia
glive...@westga.edu

ITS: Making Technology Work for You!
___
Gluster-users mailing list
Gluster-users@gluster.org
http://www.gluster.org/mailman/listinfo/gluster-users