Re: [HippoCMS-dev] RE: [Smals] Yet Another Clustering Problem

Bart van der Schans Thu, 05 Jun 2008 05:20:39 -0700

Hi David,

The locks and unlocks are not propagated through the (repository)cluster. I think I can fix it but it would require some coding andtesting. Untill an update is available you're probably better off bytemporary turning off locking in the CMS.


Regards,
Bart


[EMAIL PROTECTED] wrote:

Hi,

Here‘s update information about my previous mail. When we restart the two
repositories, the first time we lock (and only on the lock) a document, it
correctly propagated. But the unlock will not be clustered. Furthermore,
if we try again, any actions (edit and close) will never be propagated. So
it really confusing... So does somebody know how the 'lock' information is
clustered? It may help to understand this weird comportment.

Any help is welcome ;-)

David Calcus
Smals - F. Manini - Web Content Management
Rue du Prince Royal 102, 1050 Bruxelles
[EMAIL PROTECTED]
+32 (0)2 787 54 92
______________________________________________________________________________________________
Smals sluit elke aansprakelijkheid uit in verband met de juistheid, de
volledigheid of het tijdig toekomen van de informatie in deze e-mail. Aan
deze e-mail kunnen geen rechten worden ontleend en deze e-mail houdt in
geen geval een erkenning van welkdanige aansprakelijkheid in.
Dit bericht is alleen bestemd voor de geadresseerde. Indien dit bericht
niet voor u bestemd is, verzoeken wij u dit onmiddellijk aan ons te melden
en het bericht te vernietigen.

Smals décline toute responsabilité quant à l'exactitude, à l'exhaustivité
et au délai de transmission des informations contenues dans cet e-mail.
Aucun droit ne peut être revendiqué sur cet e-mail et cet e-mail
n'implique en aucun cas une reconnaissance de responsabilité, quelle
qu'elle soit.
Ce message s'adresse uniquement au destinataire. Si ce message ne vous est
pas destiné, nous vous prions de nous le signaler immédiatement et de
détruire le message.

David Calcus/SMALS-MVM
04/06/2008 13:37

To
Hippo CMS development public mailinglist<[email protected]>
cc

Subject
Re: [HippoCMS-dev] RE: [Smals] Yet Another Clustering Problem

Ard,

In fact we have got a webdav client connect on each repository. In one
case we see the lock set by the cms and no one in the second case. So the
lock seem not to be propagated.

We do another test. We lock the document with the cms with user ‘A’. And
with a webdav client connect on both repositories, we try to edit and save
this document with user ‘B’. In one case, we could do it but not in the
other one (where the lock was visible).

And what next is more critical: If we unlock a document, the cms may not
be aware. So we will not be able to edit the document because the WF box
will be empty...

Any suggestions ;-) ?

Ard Schrijvers<[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]
04/06/2008 12:46
Please respond to
Hippo CMS development public mailinglist<[email protected]>

To
[EMAIL PROTECTED], [email protected]
cc

Subject
[HippoCMS-dev] RE: [Smals] Yet Another Clustering Problem

Hello David,

Hi Bartosz and all !

Our clustering is working fine when we editing content or
properties but not at all with the "lock".

Actually if we lock a document using the cms, this
information is not propagate between the different clustered
repositories.


Correct

AFAIU a lock is not really a property and a lock on a


Exactly

document won't trigger a JMS message and this information
will not be correctly clustered.


Indeed

Nobody has an idea how to fix the issue which is not very
comfortable :-(


Yes I understand. I know that a lock won't trigger a JMS, but I am not
sure wether locks are not propagated between repositories. So, I hoped you
could do a test for me: if you make sure that in one cms, you get a lock.
Then, go to the other cms, and try to edit the document and save it. Does
this work or do you get a message that the document is locked? If it is
locked, locking is propagated between repositories, but the cms is just
not informed, because no JMS has been sent.

Pls let me know your findings,

-Ard

Many thanks,


David Calcus
Smals -  F. Manini - Web Content Management Rue du Prince
Royal 102, 1050 Bruxelles [EMAIL PROTECTED]
+32 (0)2 787 54 92
______________________________________________________________
________________________________
Smals sluit elke aansprakelijkheid uit in verband met de
juistheid, de volledigheid of het tijdig toekomen van de
informatie in deze e-mail. Aan deze e-mail kunnen geen
rechten worden ontleend en deze e-mail houdt in geen geval
een erkenning van welkdanige aansprakelijkheid in.
Dit bericht is alleen bestemd voor de geadresseerde. Indien
dit bericht niet voor u bestemd is, verzoeken wij u dit
onmiddellijk aan ons te melden en het bericht te vernietigen.

Smals décline toute responsabilité quant à l'exactitude, à
l'exhaustivité et au délai de transmission des informations
contenues dans cet e-mail. Aucun droit ne peut être
revendiqué sur cet e-mail et cet e-mail n'implique en aucun
cas une reconnaissance de responsabilité, quelle qu'elle soit.
Ce message s'adresse uniquement au destinataire. Si ce
message ne vous est pas destiné, nous vous prions de nous le
signaler immédiatement et de détruire le message.



Bartosz Oudekerk<[EMAIL PROTECTED]>  Sent by:
[EMAIL PROTECTED]

14/05/2008 16:51
Please respond to
[EMAIL PROTECTED]




To
[EMAIL PROTECTED]
cc
Subject
Re: [Smals] Yet Another Clustering Problem






[EMAIL PROTECTED] wrote:

Bartosz,

I think I found the problem with the clustering ! ;-)
I have added this line in the events.xml on the repository :

<listener role="java.util.EventListener/jms-cluster-propagator"/>

And now, it work correctly and we have our repository in

clustering, I

see in the slide.log file the publication of messages \o/

So, when I have made the repository, I based myself on the template
given with the repository 1.2.15 (J2EE-Template folder).

But this line

is not in events.xml !
It's a forget ?  Can maybe cause the same behaviour in the

future (for

others people) ?!?

Thanks, wel spotted! I will commit this immediately.

Regards,
--
Bartosz Oudekerk
.---------------------------------.---------------------------
--------.
| Hippo B.V.                      | Hippo USA Inc.
         |
| Oosteinde 11                    | 101 H Street, suite Q
Petaluma CA |
| 1017 WT  Amsterdam              | 94952-3329  San Francisco
         |
| The Netherlands                 | United States
         |
| Tel  +31 (0)20 5224466          | +1 (707) 773-4646
         |
+---------------------------------+---------------------------
--------+
|     [EMAIL PROTECTED]     |
http://www.onehippo.com      |
`---------------------------------^---------------------------
--------'
_______________________________________________
Smals mailing list
[EMAIL PROTECTED]
http://lists.hippo.nl/mailman/listinfo/smals

********************************************
Hippocms-dev: Hippo CMS development public mailinglist





------------------------------------------------------------------------

********************************************
Hippocms-dev: Hippo CMS development public mailinglist



--
Hippo B.V.  -  Amsterdam
Oosteinde 11, 1017 WT, Amsterdam, +31(0)20-5224466

Hippo USA Inc.  -  San Francisco
101 H Street, Suite Q, Petaluma CA, 94952-3329, +1 (707) 773-4646
-----------------------------------------------------------------
http://www.onehippo.com   -  [EMAIL PROTECTED]
-----------------------------------------------------------------
********************************************
Hippocms-dev: Hippo CMS development public mailinglist

Re: [HippoCMS-dev] RE: [Smals] Yet Another Clustering Problem

Reply via email to