Hmmm.. thanks for the info, Emmanuel. 
I'm waiting on a NetBSD machine for debugging this issue. 
I will let you know what I find, when I'm done. 

-Krutika 

----- Original Message -----

> From: "Emmanuel Dreyfus" <[email protected]>
> To: "Krutika Dhananjay" <[email protected]>
> Cc: "Gluster Devel" <[email protected]>
> Sent: Saturday, December 6, 2014 10:23:34 PM
> Subject: Re: [Gluster-devel] question on glustershd

> Emmanuel Dreyfus <[email protected]> wrote:

> > Forcing the option to on in the sources does not help: it goes throuh
> > __afr_selfheal_metadata_finalize_source() without finding a source and
> > returns -1.

> Tracing further:

> afr_selfheal_find_direction() get the matrix:
> 0 1
> 1 0

> As a result, witness[] is all zero, there is no source.

> Could it be otherwise? The offending node is obtained this way (I
> stripped other operations), I do not see how it could be healed, after
> all:

> TEST mkdir spb_me_heal

> TEST kill_brick $V0 $H0 $B0/${V0}0
> TEST touch spb_me_heal/0

> $CLI volume stop $V0
> $CLI volume start $V0 force
> EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" afr_child_up_status $V0 0
> EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" afr_child_up_status $V0 1
> TEST kill_brick $V0 $H0 $B0/${V0}1

> TEST touch spb_me_heal/1

> $CLI volume stop $V0

> $CLI volume start $V0 force
> EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" afr_child_up_status $V0 1
> EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" afr_child_up_status $V0 0

> --
> Emmanuel Dreyfus
> http://hcpnet.free.fr/pubz
> [email protected]
_______________________________________________
Gluster-devel mailing list
[email protected]
http://supercolony.gluster.org/mailman/listinfo/gluster-devel

Reply via email to