[Drbd-dev] FW: [DRBD-cvs] svn commit by phil - r2453 - in trunk: drbd drbd/linuxuser - * Removed some explicit calls to drbd_thread_ ... funct

[Drbd-dev] FW: [DRBD-cvs] svn commit by phil - r2453 - in trunk: drbd drbd/linuxuser - * Removed some explicit calls to drbd_thread_ ... funct

[Graham, Simon]

I think there may be a problem with this checkin (or I misunderstand) -- this checkin includes the following change to drbd_nl_net_conf:


Modified: trunk/drbd/drbd_nl.c
===================================================================
--- trunk/drbd/drbd_nl.c	2006-09-25 07:24:32 UTC (rev 2452)
+++ trunk/drbd/drbd_nl.c	2006-09-25 11:32:44 UTC (rev 2453)
@@ -934,8 +934,13 @@
 	struct crypto_tfm* tfm = NULL;
 	struct hlist_head *new_tl_hash = NULL;
 	struct hlist_head *new_ee_hash = NULL;
-	struct Drbd_Conf *odev;
+	drbd_dev *odev;
 
+	if (mdev->state.conn > StandAlone) {
+		retcode=HaveNetConfig;
+		goto fail;
+	}
+

This has the unfortunate side effect of blocking 'drbdadm connect' requests in any state other than Standalone - so I cant manually cause the network to be reconnected anymore -- is this supposed to be something I never have to do? Because right now my system is WFConnection on one side and Unconnected on the other...

Simon

Re: [Drbd-dev] FW: [DRBD-cvs] svn commit by phil - r2453 - in trunk: drbd drbd/linuxuser - * Removed some explicit calls to drbd_thread_ ... funct

[Philipp Reisner]

Am Montag, 25. September 2006 21:44 schrieb Graham, Simon:
> I think there may be a problem with this checkin (or I misunderstand) --
> this checkin includes the following change to drbd_nl_net_conf:
>
>
> Modified: trunk/drbd/drbd_nl.c
> ===================================================================
> --- trunk/drbd/drbd_nl.c	2006-09-25 07:24:32 UTC (rev 2452)
> +++ trunk/drbd/drbd_nl.c	2006-09-25 11:32:44 UTC (rev 2453)
> @@ -934,8 +934,13 @@
>  	struct crypto_tfm* tfm = NULL;
>  	struct hlist_head *new_tl_hash = NULL;
>  	struct hlist_head *new_ee_hash = NULL;
> -	struct Drbd_Conf *odev;
> +	drbd_dev *odev;
>
> +	if (mdev->state.conn > StandAlone) {
> +		retcode=HaveNetConfig;
> +		goto fail;
> +	}
> +
>
> This has the unfortunate side effect of blocking 'drbdadm connect' requests
> in any state other than Standalone - so I cant manually cause the network
> to be reconnected anymore -- is this supposed to be something I never have
> to do? Because right now my system is WFConnection on one side and
> Unconnected on the other...
>

Hi Simon,

This was a bug somwhere else, I fixed today. That particular
code means that it is no longer possible to simply do

drbdadm connect r0
drbdadm connect r0
...

to cause it to reconnect. Instead you have to do now:

drbdadm disconnect r0 ; drbdadm connect r0
drbdadm disconnect r0 ; drbdadm connect r0
...

-phil
-- 
: Dipl-Ing Philipp Reisner                      Tel +43-1-8178292-50 :
: LINBIT Information Technologies GmbH          Fax +43-1-8178292-82 :
: Schönbrunnerstr 244, 1120 Vienna, Austria    http://www.linbit.com :