Adobe Experience Manager Sites & More

ronnyfm · 7/4/16

Hello, I was installing a separate instance at 4504, but automatically AEM now creates a cluster and even offloading agents. Is there a way to prevent this default behaviour completely?

Back in 5.6 there was a file repository.xml that you could edit, but I now cannot find a way to disable this in /system/console/topology

Thanks.

Opkar_Gill · 7/4/16

Hi Ronny,

unfortunately the terminology used in the documentation and product is confusing. The cluster that you see, which is used for job offloading, is actually based on Sling and is not clustering at the JCR level, as one saw in 5.x or MongoDB clustering as in 6(Although Sling Topologies/clusters are used for job management with MongoDB)

From the documentation [0], the clustering you are seeing can be described as follows:

Topologies are loosely-coupled Experience Manager clusters that are participating in offloading. A cluster consists of one or more Experience Manager server instances (a single instance is considered as a cluster).

So technically this is clustering at the Sling level, so you can have loosely-coupled instances, in this way you can distribute jobs to instances in the "cluster", this is the idea behind job offloading. This setup is also required in MongoDB to ensure jobs only run on a single instance. As for the offloading agent, I just disable this when I install a new AEM instance.

Hope this helps.

Regards,

Opkar

[0] https://docs.adobe.com/docs/en/aem/6-2/deploy/configuring/offloading.html#Administering%20Topologies

[1]http://sling.apache.org/documentation/bundles/discovery-api-and-impl.html#topology-topologyview

View solution in original post

Tuhin_Ghosh · 7/4/16

Hi,

Try accessing http://localhost:4502/libs/granite/topology/content/view.html and see if it helps.

Thanks

Tuhin

AnkurAhlawat · 7/4/16

With the introduction of AEM 6.0, clustering configuration has been moved to storage level. So you wont be able to edit it now. This is the after affects of migrating to oak as now aem supports TARMK instead of TARPM.

Instead they have come up with a concept of cold standby for higher availability of author.

regards,

Ankur Ahlawat

AEM CQ5 Tutorials

Opkar_Gill · 7/4/16

Hi Ronny,

unfortunately the terminology used in the documentation and product is confusing. The cluster that you see, which is used for job offloading, is actually based on Sling and is not clustering at the JCR level, as one saw in 5.x or MongoDB clustering as in 6(Although Sling Topologies/clusters are used for job management with MongoDB)

From the documentation [0], the clustering you are seeing can be described as follows:

Topologies are loosely-coupled Experience Manager clusters that are participating in offloading. A cluster consists of one or more Experience Manager server instances (a single instance is considered as a cluster).

So technically this is clustering at the Sling level, so you can have loosely-coupled instances, in this way you can distribute jobs to instances in the "cluster", this is the idea behind job offloading. This setup is also required in MongoDB to ensure jobs only run on a single instance. As for the offloading agent, I just disable this when I install a new AEM instance.

Hope this helps.

Regards,

Opkar

[0] https://docs.adobe.com/docs/en/aem/6-2/deploy/configuring/offloading.html#Administering%20Topologies

[1]http://sling.apache.org/documentation/bundles/discovery-api-and-impl.html#topology-topologyview

ronnyfm · 7/5/16

Thanks all, yes, I saw that this is different than the clustering I was referring to in 5.6. But after reading a little the trick to avoid this automatically offloading setup is editing /system/console/configMgr/org.apache.sling.discovery.impl.Config and leave Topology Connector URLs empty, this way the author instance won't be in a discovery mode.

Regards,