Re: Creating sample applications

Hi David, A good question. ASFAIK this is how Docker images work: - As you have mentioned each Docker image may contain several layers depending on the number of parents they have. - Each layer has an unique ID (I think this is a hash value) which changes when they are re-built. - If one of the pa

Re: [Discuss] [Issue] Python Cartridge Agent does not wait until Complete Topology Event at the Startup

Seems EXPORT_METADATA_KEYS are missing in the agent payload launch parameters. I will check why it is not there in payload. Thanks, Gayan On Mon, Feb 9, 2015 at 10:36 AM, Lakmal Warusawithana wrote: > I'm getting following error inside cartridge /tmp/agent.screen.log > > > [2015-02-09 04:49:50,

Re: [Discuss] [Issue] Python Cartridge Agent does not wait until Complete Topology Event at the Startup

I'm getting following error inside cartridge /tmp/agent.screen.log [2015-02-09 04:49:50,864] DEBUG {cartridgeagentconfiguration.py:__read_conf_file} - Config file path : '/mnt/apache-stratos-python-cartridge-agent-4.1.0-SNAPSHOT/agent.conf' [2015-02-09 04:49:50,865] DEBUG {cartridgeagentconfigur

Re: Creating sample applications

Hi, Imesh Thanks for your feedback. The second question is a bit complex. For example, when I build the wso2am minimum cluster mysql docker images, I may have three layer images: BaseImage ( RHEL7 ) ==> MysqlImage ( Dockerfile:: FROM BaseImage; RUN yum install mysql ) ==> WSO2MySQLImage ( Dockerf

Re: Creating sample applications

Hi David, It's nice to hear your feedback! - Actually we cannot access mock instances because they are just threads that simulate VM/container instances. - I didn't get your second question. Can you please rephrase it? Thanks On Mon, Feb 9, 2015 at 1:39 AM, david hbase wrote: > Hi, Team > > Th

Re: Could not connect to message broker

Hi Vishanth, When does this happen? Could you verify ActiveMQ is running on the given host and port (just to be confirmed) On Mon, Feb 9, 2015 at 12:54 AM, Vishanth Balasubramaniam < vishan...@wso2.com> wrote: > Hi devs, > > I have started ActiveMQ, but still getting the following error > recurs

Re: [Discuss] Error Response JSON Message Construction

I have fixed this [1] now in commit 35a2c1d374b590ae5d9b1822533518ec87371364 [1] https://issues.apache.org/jira/browse/STRATOS-1181 On Fri, Feb 6, 2015 at 4:00 PM, Sajith Kariyawasam wrote: > Yes Gayan, > In the sample I have used, the errorMessage is the one which comes from > the back end. >

Re: [Discuss] [Issue] Sample Applications do not Activate

Hi, The group-scaling-v1 can not be un-deployed, after failed undeployed, can not re-active again. Also after this failed, the single-cartridge-app can not active. David On Sun, Feb 8, 2015 at 1:54 PM, Reka Thirunavukkarasu wrote: > Hi > > After i have removed this particular check to start

Re: [Discuss] [Issue] Sample Applications do not Activate

According to my understanding, startup order is broken due to we start all the adder at once..Could you please check on this? Thanks, Reka On Sun, Feb 8, 2015 at 1:51 PM, Reka Thirunavukkarasu wrote: > Hi Gayan, > > I found below segment has been added when starting the monitor adder. > Could y

Re: [Discuss] [Issue] Sample Applications do not Activate

Hi Gayan, I found below segment has been added when starting the monitor adder. Could you please explain Why it has been added? We used to start the adder whenever needed as per earlier implementation by referring to dependency tree. Won't this break the dependency order of the Clusters? I found t

Re: Error log does not contains much information

I have fixed this in commit 35a2c1d374b590ae5d9b1822533518ec87371364. On Fri, Feb 6, 2015 at 4:19 PM, Sajith Kariyawasam wrote: > I did a remote debug and found that "faultMessage" contains the original > exception message sent from the backend. > So if we use > "e.getFaultMessage().getCartridge

[jira] [Commented] (STRATOS-1181) Add error response message body to API methods together with HTTP status codes (400, 409, etc)

[ https://issues.apache.org/jira/browse/STRATOS-1181?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14311551#comment-14311551 ] Sajith Kariyawasam commented on STRATOS-1181: - Fixed in commit 35a2c1d374b59

[jira] [Created] (STRATOS-1181) Add error response message body to API methods together with HTTP status codes (400, 409, etc)

Sajith Kariyawasam created STRATOS-1181: --- Summary: Add error response message body to API methods together with HTTP status codes (400, 409, etc) Key: STRATOS-1181 URL: https://issues.apache.org/jira/browse/

Fwd: Creating sample applications

Hi, Team Thanks for updated the UI and testing scripts. The mock testing procedures are more smooth now. Just found two questions: 1. How can we access the instance in mock? 2. For the Docker image in cartridge, if we have have the base/platform layer images updated, how can we trigger/hooks the

Could not connect to message broker

Hi devs, I have started ActiveMQ, but still getting the following error recursively. Anyone facing the same? [2015-02-09 00:50:01,030] ERROR {org.apache.stratos.messaging.broker.connect.amqp.AmqpTopicConnector} - Could not disconnect from message broker org.apache.activemq.ConnectionFailedExcept

Re: [Discuss] [Issue] Sample Applications do not Activate

Thanks Imesh for fixing the issues..I will continue testing with the other samples.. Thanks, Reka On Sun, Feb 8, 2015 at 10:59 AM, Imesh Gunaratne wrote: > May be this message processor is receiving an invalid message. Will debug > and see. > > On Sun, Feb 8, 2015 at 11:09 PM, Gayan Gunarathne

Re: [Discuss] [Issue] Sample Applications do not Activate

May be this message processor is receiving an invalid message. Will debug and see. On Sun, Feb 8, 2015 at 11:09 PM, Gayan Gunarathne wrote: > Hi Imesh, > > Seems "GroupStatusInactiveProcessor" is the last processor in the > "GroupStatusProcessorChain".We are printing the above warn message if th

Re: [Discuss] [Issue] Sample Applications do not Activate

Hi Imesh, Seems "GroupStatusInactiveProcessor" is the last processor in the "GroupStatusProcessorChain".We are printing the above warn message if the next processor is null. if (nextProcessor != null) { // ask the next processor to take care of the message. return n

Re: [Discuss] [Issue] Sample Applications do not Activate

I tested both group-scaling and group-scaling-v1 applications but they did not activate: I can see the following warning in the log: WARN {org.apache.stratos.autoscaler.status.processor.group.GroupStatusInactiveProcessor} - No possible state change found for [component] mygroup2 [instance] Than

Jenkins build is back to normal : Stratos-Master-Commit-Build #135

See

Jenkins build is back to normal : Stratos-Master-Commit-Build » Apache Stratos - Load Balancer - P2 Profile Generation #135

See

Re: [Discuss] [Issue] Python Cartridge Agent does not wait until Complete Topology Event at the Startup

Hi Lakmal, In my Kubernetes cluster I can see this image as follows: core@master ~ $ docker images REPOSITORY TAG IMAGE IDCREATED VIRTUAL SIZE stratos/php 4.1.0-alpha 44a4916713ea46 hours ago 417.8 MB stratos/php 4.1.0-

Re: [Discuss] [Issue] Python Cartridge Agent does not wait until Complete Topology Event at the Startup

Imesh, did you tried with uploaded image rc1? seems like it has not uploaded correctly. Here what I can see after docker pull. No name , no tag and it lesser that alpha tag. I have updates sample to rc1 docker images and it fails may be due to this. dbd9c1be

Build failed in Jenkins: Stratos-Master-Commit-Build #134

See Changes: [imesh] Introducing statistics publishing interfaces and factory classes -- [...truncated 9074 lines...] Compression is 0.0% Took 0.33 sec [JENKINS] Archiving

Build failed in Jenkins: Stratos-Master-Commit-Build » Apache Stratos - Load Balancer - P2 Profile Generation #134

See -- [INFO] [INFO] -

Re: DISCUSS: how stable is master / new snapshot ?

Yes for sure Lakmal, we will review all the issues found/reported so far and make sure we fix L1 and L2 issues for beta. Thanks On Sun, Feb 8, 2015 at 5:29 PM, Lakmal Warusawithana wrote: > IMO we should not have L1 and L2 issues for beta. > > On Sun, Feb 8, 2015 at 5:27 PM, Imesh Gunaratne wr

[Discuss] Introducing Interfaces and Factory Classes for Statistics Publishing

Hi Devs, I have introduced a set of interfaces and factory classes for statistics publishing: ​ Now we could introduce support for other Complex Event Processors without much effort. Thanks -- Imesh Gunaratne Technical Lead, WSO2 Committer & PMC Member, Apache Stratos

Re: DISCUSS: how stable is master / new snapshot ?

IMO we should not have L1 and L2 issues for beta. On Sun, Feb 8, 2015 at 5:27 PM, Imesh Gunaratne wrote: > Hi Lakmal/Martin, > > +1 Yes currently we are fixing some functional issues in grouping logic, > once this is done we should be able to create a beta developer preview tag. > > Thanks > > O

Re: DISCUSS: how stable is master / new snapshot ?

Hi Lakmal/Martin, +1 Yes currently we are fixing some functional issues in grouping logic, once this is done we should be able to create a beta developer preview tag. Thanks On Sun, Feb 8, 2015 at 4:34 PM, Lakmal Warusawithana wrote: > Shall we go for a Beta developer preview? > > On Sat, Feb

Re: [Discuss] [Issue] Sample Applications do not Activate

Great work Reka! Thanks for looking into this! On Sun, Feb 8, 2015 at 3:27 AM, Reka Thirunavukkarasu wrote: > Hi > > I have fixed one issue that i found when using more than one > groupInstances in the application. Now that, i could get all the > single-group-* samples became active without any

Re: DISCUSS: how stable is master / new snapshot ?

Shall we go for a Beta developer preview? On Sat, Feb 7, 2015 at 2:11 AM, Martin Eppel (meppel) wrote: > Hi, > > > > I am still using the alpha tagged version (4.1.0-alpha) but it seems like > quite a few fixes went into the master since then, do you guys think it > would be a good time to take

Jenkins build is back to stable : Stratos-Master-Nightly-Build #403

See

Jenkins build is back to stable : Stratos-Master-Nightly-Build » Apache Stratos - Integration Tests #403

See

DISCUSS: how stable is master / new snapshot ?

Hi, I am still using the alpha tagged version (4.1.0-alpha) but it seems like quite a few fixes went into the master since then, do you guys think it would be a good time to take a new snapshot / tag assuming the master is stable enough ? Thanks Martin

[GitHub] stratos pull request: [STRATOS-1164] - PCA - Git related logic ref...

Github user asfgit closed the pull request at: https://github.com/apache/stratos/pull/217 --- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is ena

Re: [Discuss] [Question] AutoscalerUtil.allClustersInitialized() Method

Sorry. I should have added the comment in that section. Will go through the code and add the relevant comments in the area where it has less visibility. Thanks, Reka On Sat, Feb 7, 2015 at 11:23 AM, Reka Thirunavukkarasu wrote: > Hi Imesh, > > As i explained in the other thread, all the cluster

Re: [Discuss] [Question] AutoscalerUtil.allClustersInitialized() Method

Hi Imesh, As i explained in the other thread, all the clusters are created at once in the ApplicationClustersCreatedEvent. I beleive that by checking whether one of the cluster exists, we can assume that all the clusters are there which will improve performance as well. Thanks, Reka On Sat, Feb

Re: [Discuss] [Question] Why did we introduce Application Clusters Created event while having Cluster Created event

Hi Imesh, Since we have introduced the Composite Application model where more than one clusters present, we are no longer using cluster created event in order to avoid unnecessary complication when registering the ApplicationMonitor. Let me explain the flow little bit here, 1. Deploy application

Re: [Discuss] [Issue] Sample Applications do not Activate

Hi I have fixed one issue that i found when using more than one groupInstances in the application. Now that, i could get all the single-group-* samples became active without any issue. Thanks, Reka On Sat, Feb 7, 2015 at 12:08 PM, Reka Thirunavukkarasu wrote: > Hi > > Do we have the current st

Re: Meta data service OAuth authentication

Hi Udara, It would be better if you can send the list of API methods in Metadata service and sample requests. I can see that we are invoking the SOAP based OAuth Admin Service to generate the Token in the application parser. Do we have a REST API for accessing the OAuth Admin Service? If so we mi

Re: [Discuss] [Issue] Sample Applications do not Activate

Hi Reka, I fixed subscribable information finding logic in default application parser, not sure about others (cannot see any other commits except for one from you). Moreover I found several problems in the samples and fixes those today. Did the group samples stay in Active state constantly? Than

Re: [Discuss] [Question] AutoscalerUtil.allClustersInitialized() Method

Great thanks Reka! Now it looks good. On Sun, Feb 8, 2015 at 12:03 AM, Reka Thirunavukkarasu wrote: > Added comment in 0b409e5cd2020322b82a9c472a9cee6d4c3e607f. > > On Sat, Feb 7, 2015 at 11:26 AM, Reka Thirunavukkarasu > wrote: > >> Sorry. I should have added the comment in that section. Will

Re: Meta data service OAuth authentication

On 8 Feb 2015 15:19, "Gayan Gunarathne" wrote: > > Hi All, > > AFAIU from the current code base following the way we have implemented to handle the OAuth token authentication. > > 1. We will generate the OAuth token when the instance spawn and pass it through the cartridge payload as launch param(

Re: [Discuss] [Issue] Sample Applications do not Activate

Hi Do we have the current state of the samples and application state? When i tried group samples in mock and it has become active..Did we fix this recently? Thanks, Reka On Sat, Feb 7, 2015 at 11:54 AM, Reka Thirunavukkarasu wrote: > Great job in organizing this samples..I just tried the simpl

Re: [Discuss] [Question] Why did we introduce Application Clusters Created event while having Cluster Created event

+1 to remove the cluster created event from the Topology. As we have implemented in a way to listen for all the clusters to be created at once, it may add more complexity now, if we change to cluster created event. I think that cartridge agent and others can use the ClusterInstanceCreated in order

Meta data service OAuth authentication

Hi All, AFAIU from the current code base following the way we have implemented to handle the OAuth token authentication. 1. We will generate the OAuth token when the instance spawn and pass it through the cartridge payload as launch param(TOKEN) to the cartridge. 2. From the cartridge payload we

Re: [Discuss] [Issue] Sample Applications do not Activate

Great job in organizing this samples..I just tried the simple sample. It saves lot of time in the testing phase.. On Fri, Feb 6, 2015 at 7:08 PM, Imesh Gunaratne wrote: > Hi Chamila, > > Now we have a common set of artifacts and deployment scripts for all the > IaaSes except for deployment pol

Re: [Discuss] [Question] AutoscalerUtil.allClustersInitialized() Method

+1. Will include comments for such code segment.. Thanks, Reka On Sat, Feb 7, 2015 at 6:43 PM, Imesh Gunaratne wrote: > As I found we have written above logic to check availability of one > cluster and assume that all the other clusters have been initialized > because of the Application Cluster

Re: [Discuss] [Question] AutoscalerUtil.allClustersInitialized() Method

Added comment in 0b409e5cd2020322b82a9c472a9cee6d4c3e607f. On Sat, Feb 7, 2015 at 11:26 AM, Reka Thirunavukkarasu wrote: > Sorry. I should have added the comment in that section. Will go through > the code and add the relevant comments in the area where it has less > visibility. > > Thanks, > Re