Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 19 Jan 2005 13:57:56 -0500
From:      Brian McCann <bjmccann@gmail.com>
To:        Pawel Jakub Dawidek <pjd@freebsd.org>
Cc:        freebsd-geom@freebsd.org
Subject:   Re: ggatec reporting false info / not exiting correct?
Message-ID:  <2b5f066d0501191057549101c0@mail.gmail.com>
In-Reply-To: <20050119182641.GO795@darkness.comp.waw.pl>
References:  <2b5f066d05011910056e7a69e1@mail.gmail.com> <20050119182641.GO795@darkness.comp.waw.pl>

next in thread | previous in thread | raw e-mail | index | archive | help

Yes...it did.  Didn't quite understand what attach did. :)  Thanks!

--Brian

P.S.: ggate is really awesome...thanks again!


On Wed, 19 Jan 2005 19:26:41 +0100, Pawel Jakub Dawidek <pjd@freebsd.org> wrote:
> On Wed, Jan 19, 2005 at 01:05:22PM -0500, Brian McCann wrote:
> +>      Hi all.  I'm working on getting gmirror to work over ggate
> +> connections..and it's working well...just trying to break it and make
> +> recovery situations.  However, I've discovered what I believe to be a
> +> flaw in ggatec.  I had PC1 running ggatec and PC2 running ggated.  PC2
> +> was exporting /dev/da1s1d.  I had gmirror on PC1 setup with
> +> /dev/da1s1d and /dev/ggate0d.  Was working great.  I then simulated a
> +> failure in the network by killing ggated on PC2.  gmirror saw the
> +> problem, marked the array accordingly.  I then bought ggated back up
> +> on PC2.  To my disappointment, ggatec had apparantly closed once there
> +> was a failure (thought it may have tried to reconnect at a given
> +> interval or something like NFS).  However, when I ran "ggatec list",
> +> it showed "ggate0", and an ls of /dev showed a ggate0 as well....but
> +> no ggatec process was running.  It's like when ggatec could not talk
> +> to ggated anymore it just gave up and quit, not cleaning up after
> +> itself at all...this can't be right...at least....I wouldn't think it
> +> to be right.  Is it supposed to work like this (not cleaning up)?  If
> +> so, I'm curious as to why.  On a side note, it would be awesome if
> +> ggatec COULD have a background option...where if it can't contact the
> +> ggated host to wait and try again after a certain time.
> 
> And if you run:
> 
>         # ggatec attach -u 0 ...
> 
> Does it work again? It should.
> 
> --
> Pawel Jakub Dawidek                       http://www.wheel.pl
> pjd@FreeBSD.org                           http://www.FreeBSD.org
> FreeBSD committer                         Am I Evil? Yes, I Am!
> 
> 
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2b5f066d0501191057549101c0>