ESME Blog

In order to assure that ESME fits well into enterprise settings, we have started including Java Management Extensions (JMX) to our application.  This is to allow the monitoring of ESME at various levels (number of messages, number of users, etc.).  Most monitoring tools in the enterprises can tap into JMX-based measures; therefore, we decided to use this standard as well.

Currently, we have only added approximately 10 measures but plan to add more measures as the need arises.

If you’d like more information about the use of JMX in ESME, we’ve created an initial page in our wiki.

We are now using Apache Hudson for daily builds.   Every day, our code base is extracted from the subversion-based code repository, built and tests performed. This allows us to make sure that we have code base that is viable and assists in our quality control measures. JIRA items that were built also receive comments describing the successful build.

If you want to keep track of our builds, here is the RSS feed.

Siemens SIS has just released an official press release about ESME and about Siemens SIS’ involvement in the project.

The official press release is available here.

As we wrote yesterday here on our blog, you can now access ESME via existing Twitter clients.

I tried using an iPhone today against the Stax deplyoment with the latest Apache code drop.
On Laconi.ca’s app wiki, there are a few different apps to choose from. After trying out a few I eventually ended up using LaTwit. € 2,49 in the iTunes App Store.

Below is a screenshot of my timeline on the iPhone:



If you want to try it yourself, you need to do the following:
1. Create a user on the ESME server at Stax
2. Click on the manage token link and create a token. This is is your password for using the iPhone client

3. In the account settings in LaTwit:

Add the ESME API URL in the field “Server API URL” in this format: esmecloudserverapache.dickhirsch.staxapps.net/twitter

Add the username which you created in step 1.

Add the password (token) which you created in step 2.

4. Remember to enable the account before you finish.

5. You should now see the public timeline.

Please note that we have not yet done everything in the Twitter API, so some things still aren’t working (for example following users). Other functional characteristics of Twitter like DMs aren’t yet part of the ESME architecture.

Please also note that the Web UI at Stax doesn’t work properly, so bear with us. The Stax deployment is the latest code drop from Apache which means it might include bugs.

ESME has a REST API that is necessary for clients to access data / functionality. There was some degree  of debate / criticism – some of it rather critical  – of this initial interface. Ethan Jewett has blogged about a suggestion for a new REST API for ESME which not only provides an examination of main points of REST-related debate but also the foundation for a new REST API.  Ethan is planning on moving the discussion over to the Apache wiki for further discussion.

Please join us in creating the best possible REST API for ESME.

Note: I don’t currently have the link for the page in Apache wiki with the discussion but I will update this blog when I have the link.

One of the main tasks for the first sprints since we joined Apache has been support of the Twitter API .  The main reason for this functionality is the ability to use existing twitter clients to access ESME. Well, Vassil Dichev  has been working hard to this important task up and running. Using Twitter clients, it is now possible to send messages, receive the public and private timelines as well as view friends/followers.

The use of other Twitter clients to access ESME isn’t always possible, because the client’s configuration must be changed so that the application accesses the Esme server rather than the main Twitter server. Unfortunately, this switch isn’t always possible, because the server which is the basis for messages is usually hardwired.  Those applications that can also access laconi.ca  servers can also be made to access ESME servers

I’ve tried it using Twhirl against the Stax deplyoment with the latest Apache code drop and was pleased to see that it works.

If you want to try it yourself, then you to do the following steps.
1. Create a user on the ESME server at Stax
2. Click on the manage token link and create a token. This is is your password for using the Twitter client
3. Create a new account with twhirl using the laconi.ca type not the identica.ca type.

Use this synatx “apachedick@esmecloudserverapache.dickhirsch.staxapps.net”. Instead of “apachedick” use your own user_id.
4. Click on ”Connect” and set password using your token from step 2.

5. You should now see the public timeline.

I’ve only tried accessing ESME via twhirl. If anyone has success using other clients, then please add a comment.

Please note that we have not yet done everything in the Twitter API, so some things still aren’t working (for example following users). Other functional characteristics of Twitter like DMs aren’t yet part of the ESME architecture.

Please also note that the Web UI at Stax doesn’t work properly, so bear with us. The Stax deployment is the latest code drop from Apache which means it might include bugs.

Daniel Koller has released a few preliminary details on his latest load test based on the latest Apache-based source.  This time he focused on the REST API and compared a one-server configuration to a configuration with three server.

The first picture shows the environment with one server:

The second picture shows three servers:

What is interesting is the comparison of these results with the initial load tests that were based on an older version of Scala with a notorious memory bug. In the first tests, the performance curve was very chaotic. The new version of ESME is much more stable and shows more expected performance behavior.

We’ll publish more details on these latest performance test soon.

Darren Hague has moved the ESME Client based on the Java Api from the Google Code repository to our Apache repository. He also added additional methods including the critical ”getMessages” method which was previously missing. It is now possible to easily create bots that integrate with existing environments.  Darren has also provided a test application (with code) that shows how to use the API.

The JAVA API is still a work in progress but the most important parts are now present.

ESME got a nice mention in Mark Yolton’s (Senior Vice President of the SAP Community Network ) keynote at “Open Source Meets Business” (OSMB) conference. Take a look at Mark’s blog about his presentation (slides are here). The slides are interesting, because they describe the changing relationship between open- and closed-source software. This discussion is obviously relevant to the manner in which ESME evolves.

In a previous blog, I described our use of the Stax Cloud to perform load tests for ESME. Well, Daniel Koller has just finished his analysis of the results of this initial test series and has made a few interesting discoveries. For the ESME team, these initial tests were primarily focused on gaining experience with Stax as an environment to perform such tests (including the use of clusters) as well as creating a test bed for future ESME tests.  The tests used both the REST-API and the Web UI.

Now you may be thinking the main focus should be on measuring ESME performance. True. However, the ESME version that we used is based on a older Scala library with a known memory bug. Thus, the test results are not really representative of the current ESME code base.

If you look at other microblogging platforms (including Twitter) – irregardless of whether they are focused on the enterprise or not, you’ll be hard pressed to find any load test results that are published and available to potential users. However, such tests are usually mandatory for IT projects, especially for larger companies with many potential users.  Thus, we would like start publishing our results so that those interested in ESME can make better sizing decisions.

We will soon be starting a new round of load tests based on the Apache code base and a new Scala library, so expect a blog in the future with these results.

For those interested in performing the tests themselves, Daniel has also made the test scripts available