Does anybody know of a way to do http session replication between web apps running in distributed apache karaf OSGi containers?
In this post, http://karaf.922171.n3.nabble.com/Pax-web-failover-LoadBalancing-td4029552.html, Jean-Baptiste Onofré says it's not available in apache cellar yet. Is this capability available anywhere yet?
I've been googling all day and haven't found any options -- Thanks for any help.
Steve
Karaf does not have it for pax-web as confirmed by JB in the above post. I am also waiting for him to provide us.
If you are in a hurry
You can use Hazelcast (already with Cellar) for session replication.
Hazelcast supports it and provide HazelcastWM for the same purpose.
Jetty does provide session replication and we use jetty as
servlet-container.You have to check possibility with it.
Try on forum for putting Tomcat instead of Jetty.
Related
We have an "cache" (javax.cache.Cache) implementation that is a wrapper of Hazelcast's IMap. We use a composite Object key.
We upgraded from version 3.12.5 to 5.1.1. When I deploy the system on a local Windows machine, all works well. But when I deploy the system into an Kubernetes environment, the map just "does not work". Values do not get persisted into the map (after a put operation). An Hazelcast cluster does get formed so it does not seem to be an auto discovery issue. I also have another K8S env in which it does work properly.
I enabled Hazelcast's diagnostic mode and it does not seem to show me anything useful. I do not get any error or warn messages from the com.hazelcast.* package. The same issue happened also when I tried version 4.x.
I am trying to explore ways which will help to the realise what is the issue here. Thank you.
Turns out it is a bug. Hazelcast recommends to use the value of 0 instead.
I had the same problem but i was migrating from hazelcast 3.11.1 to 5.1.2, I found the IMap in "com.hazelcast.map" not in "com.hazelcast.core"
Stuck with the below error while configuring dse address.yaml.
INFO [async-dispatch-1] 2021-04-17 07:50:06,487 Starting DynamicEnvironmentComponent
;;;;;
;;;;;
INFO [async-dispatch-1] 2021-04-17 07:50:06,503 Starting monitored database connection.
ERROR [async-dispatch-1] 2021-04-17 07:50:08,717 Can't connect to Cassandra, authentication error, please carefully check your Auth settings, retrying soon.
INFO [async-dispatch-1] 2021-04-17 07:50:08,720 Finished starting system.
Configured cassandra user and password in cluster-name.conf & address.yaml as well.
Any advice would be appreciated.
You've provided very little information about the issue you're running into so our ability to assist you is very limited.
In any case, my suspicion is that (a) you haven't configured the correct seed_hosts in cluster_name.conf, or (b) the seeds are unreachable from the agents.
If this is your first time installing OpsCenter, I highly recommend that you let OpsCenter install and configure the agents automatically. When you add the cluster to OpsCenter, you will get prompted with a choice on whether to do this automatically:
For details, see Installing DataStax Agents automatically.
As a side note, we don't recommend you set any properties in address.yaml unless it's a last resort. Wherever possible and in almost all cases, configure the agent properties in cluster_name.conf so it's managed centrally instead of individual nodes.
It's difficult to help you troubleshoot your problem in a Q&A forum. If you're still experiencing issues, my suggestion is to log a DataStax Support ticket so one of our engineers can assist you directly. Cheers!
I`m new with couchdb and I was configuring it for a new project, I had a doubt about what is the difference beetween httpd and chttpd in couch db main config:
vs:
when do you use one or the other?
From the documentation regarding chttpd
In CouchDB 2.x, the chttpd section refers to the standard, clustered
port. All use of CouchDB, aside from a few specific maintenance tasks
as described in this documentation, should be performed over this
port.
chttpd essentially is for address and port bindings.
httpd on the other hand is for configuring behaviors over a connection (i.e. chttpd), for example, to configure CORS.
I have tons of tomcat servers, they are all in a virtual machine. At the moment there is a need to develop a web panel, where I can track the statuses of servers, change their configurations, stop and restart. Actually the question itself: with what technologies can I do this. Previously, there was an idea to use the playbook ansible.
How can I at least display the names of my servers on the page?
There are two standard ways to monitor Tomcat:
you can use Tomcat Manager, especially its text interface,
you can use JMX directly or through the JMX Proxy Servlet. On Tomcat's webpage you can find a not so up-to-date list of MBeans. For some MBeans you'll have to fire up jconsole and explore the names yourself.
In the wildfly config file (standalone.xml), it is possible to enable statistics for the ejb and datasources susbsystems
I would like to enable them to get useful infos at runtime (invocations, execution times,...)
But is it something that can be activated on a production server ?
Is there any overhead ?
The answer to both question is yes :) you can activate it and there will be an overhead.
You can use the jboss-cli to enable/disable the stats without requiring a restart of the server.