Terracotta Discussion Forums (LEGACY READ-ONLY ARCHIVE)

Hi ,

We have just migrated to Bigmmeory Max. Ours is a clustered cache architecture where multiple clients join a Bigmemory cache server. Hence distributed.

One thing I noticed is that when the cache server starts up it looses all data. I read the 'Fastrestart' feature of Bigmemory where the cache data can be offloaded onto disk and reloaded on server start up. I am not clear of what to do for distributed caches.

Will persistenceStrategy="distributed" offload data onto disk across the cluster. There is no such mention. Or can clustered caches not be set up for FastRestart with data from disk.

How does persistenceStrategy="distributed" wprk as by default those caches are set with clustered true (So i do not need terracotta element added)

Regards,
Suranjana

Hi,
Do you have any update on the question that I posted earlier? Can someone please help me with the quartz jconsole issue?

Hi,

Thanks for the reply.

I launched jconsole and tried to run the operations on Quartz MBean(that shows up fine). However on trying to execute any operation like 'getJobDetail' or 'gettrigger' I am getting the following error message:

Problem invoking getJobDetail: java.lang.NullPointerException

I am doing the following:
Added org.quartz.scheduler.jmx.export=true in quartz.properties

have done everything that has been suggested online. Some Quartz jconsole operations work and some don't and this makes me believe that there could be some error on Quartz manager. I am listing down what I have done:

What I have done:

Setup: Set up Quartz jobs using Spring 3.1, Quartz 2.2.1 and big memory 4.0.4 as the job store (Bigmemory ships with Quartz 2.2.1)
The setup works fine and I can see the job run. I also see the jobs client joining as a client node in big memory server
I have used the following quartz properties:
org.quartz.scheduler.instanceName=ZongPayScheduler
org.quartz.scheduler.instanceId=AUTO
org.quartz.threadPool.class=org.quartz.simpl.SimpleThreadPool
org.quartz.threadPool.threadCount=5
org.quartz.threadPool.threadPriority=5
org.quartz.jobStore.misfireThreshold=60000
org.quartz.jobStore.class=org.terracotta.quartz.EnterpriseTerracottaJobStore
org.quartz.jobStore.tcConfigUrl=localhost:9510
org.quartz.scheduler.jmx.export=true
org.quartz.scheduler.jmx.objectName:quartz:type=QuartzScheduler,name=ZongPayScheduler, instanceId=NONE_CLUSTER

Viewing and triggering Quartz jobs: As suggested online, I have added the following properties to my quartz.properties
org.quartz.scheduler.jmx.export=true
org.quartz.scheduler.jmx.objectName:quartz:type=QuartzScheduler,name=ZongPayScheduler, instanceId=NONE_CLUSTER

Observation:
I am seeing the Quartz management show up as an Mbean and I can see the operations. I can also see the Triggers in 'AllTriggers' attribute in the Mbean. I can also see AllJobDetails values correctly. I can also execute operations like 'start', or 'shutdown' or standby from the console.

Error
I am unable to execute any operation on the mBean like 'scheduleJob', 'unscheduleJob', 'getJobDetail' or 'getTrigger'
When I click on getTrigger operation with the same Trigger name that shows up in the Alltriggers attribute, I am getting the null pointer exception:
Problem invoking getTrigger : java.lang.NullPointerException (screenshot attached)
'scheduleJob' and 'unscheduleJob' always returns false

As some attributes and operations work, I am assuming that operations like 'scheduleJob' or 'unscheduleJob' are not working because of some error on Quartz manager.

Can you please provide me any guidance on how I can get the console working for operations like 'scheduleJob' which is extremely important for our QA team.

Hi,
We are trying to upgrade to bigmemory, and as a part of it also upgrade to the latest Quartz. As a part of upgrading Quartz we are using the EnterpriseTerrcotaJobStore that comes with bigmemory. However we are unable to view the job store on TMC or any console.

However, we were earlier able to see all jobs on dev-concole.sh(terracotta 3.5.2) and we could also trigger jobs from Quartz tab in the console (for testing).

How can we see the jobstore and trigger jobs from bigmemory and quartz 2.2. Bigmemory does not have dev-console.sh and TMC does not seem to have Quartz.

Please help, and this is a major functionality that we will be loosing by upgrading to bigmemory. How can we view the Job store and trigger jobs after upgrading to Bigmemory?

Regards,
Suranjana

The issue can be solved by deleting
$HOME/.tc/mgmt directory

Hi,

I am facing the same issue with bigmemory version 4.0.2. I tried downloading the latest 4.0.3 but I am still seeing the same issue. Can you please let us know what todo. However I can start the console fine from bigmemory 4.0.0.

Below is the error that i am seeing while starting tmc console from 4.0.3

Code:

java.lang.NullPointerException
         at com.terracotta.management.config.Config.returnsAgentWithSameLocation(Config.java:178)
         at com.terracotta.management.config.Config.addAgent(Config.java:150)
         at com.terracotta.management.config.Config.<init>(Config.java:43)
         at com.terracotta.management.config.Config.loadAndValidate(Config.java:271)
         at com.terracotta.management.services.impl.FileConfigService.loadConfigFromInputStream(FileConfigService.java:232)
         at com.terracotta.management.services.impl.FileConfigService.loadConfigFromFile(FileConfigService.java:225)
         at com.terracotta.management.services.impl.FileConfigService.loadConfig(FileConfigService.java:106)
         at com.terracotta.management.services.impl.FileConfigService.<init>(FileConfigService.java:56)
         at com.terracotta.management.security.web.shiro.TMSEnvironmentLoaderListener.loadUnlicensedCtxt(TMSEnvironmentLoaderListener.java:228)

Hi,

We are trying to migrate to bigmemory 4.0.2 and I am facing issues with tmc console. I am listing the setup and the issues that I am facing below. Currently I am starting the server and client locally.

Setup:
I have started a bigmemory TC server. I have an application deployed in tomcat that has 'ehcache' implemented. I have a couple of CacheManagers where my caches are defined. I have implemented websessions and my client is successfully able to establish connection with the terracota server.

Issues:
Firstly, while trying to start up the tmc from bigmemory4.0.2/tools/management-console.. I am getting a HTTP 503 error. On looking at the error log I am seeing the following error:
2013-08-29 19:23:33.401:WARN:oejw.WebAppContext:Failed startup of context o.e.j.w.WebAppContext{/tmc,file:/private/tmp/jetty-0.0.0.0-9889-tmc.war-_tmc-any-/webapp/},/Users/subasu/Documents/Work/bigmemory-max-4.0.2/tools/management-console/webapps/tmc.war
java.lang.NullPointerException
at com.terracotta.management.config.Config.returnsAgentWithSameLocation(Config.java:178

I am therefore staring the tmc console from bigmemory 4.0.0. the console starts up fine and I can access the URL

http://localhost:9889/tmc

Here I see the local cluster with the main TC server and my webapp client.
However I want to see all the caches adn the cache Manager.
I am seeing the following in ApplicationData:

There are no CacheManagers

default-group
bigServer 10.244.201.223:9550 [Stats] [Config] Active
Connected Clients
localhost:64539 ClientID[1] [Stats] [Config]

Can you please tell me why I am not able to startup tmc from bigmemory4.0.2 which is the version I am using for development. Also, how can I see the cachemanagers and caches.