On Wednesday, May 30, 2012 at 12:32 PM, Frank Zhang wrote: > I have some different thoughts here > > > On Tuesday, May 29, 2012 at 10:29 PM, Alex Huang wrote: > > > > > CloudStack is built as a product with modularity enforced by interfaces. > > > It > > had its own component management via ComponentLocator. While that > > framework made it easy for a small group of developers to work together, it > > has outlived its usefulness. Now that we are expecting more developers to > > join Apache CloudStack, we're looking for the next component framework to > > scale code development. This framework needs to manage the following: > > > - life cycles of the components > > > - enforce boundaries between components > > > - easily allow components to be upgraded/downgraded > > > - database upgrades for the component > > > > > > The most successful of such an application is Eclipse, which uses > > > framework > > defined by OSGi. We're planning to use that for our component framework. > > > > > > I like to gather opinions about > > > > > > 1. Whether to use OSGi or there should be other frameworks to consider? > > Personally, I would absolutely love to see osgi explored. We have been using > > osgi with great success for ServiceMix for several years and have learned > > volumes about the ins and outs. For a controlled server environment like > > Cloudstack, I think it would be ideal. There are some learning curves, and > > some extra work involved, but the rewards are plenty. > > > > > First I don't think CloudStack needs to update or remove bundle on the fly. > This feature looks ideal, but considering a cluster management nodes, for > example four running mgmt nodes, how to propagate bundle upgrade on > one mgmt node to others? And how to guarantee all bundles get working only > after all mgmt nodes gets upgraded? I don't know if OSGI is capable of this > problem. > To me runtime upgrade introduces much complexity, at least to CloudStack. > >
Many ways to solve these issues if needed, but if you don't envision applying runtime updates why be concerned about it? In either case, this type of thing is addressed in the osgi spec, specifically parts regarding service references, bundle lifecycle, and maybe startlevel. > > Second I think the bundle boundary of OSGI is too tiny. I may be wrong here, > as I am not OSGI expert. I was thinking it is a good idea to treat every > interface as > a service? For example, logging, I hope the logging package used by every > bundles > is linked in it, instead of packaging it as a logging bundle. CloudStack > should focus on > its own bundle, like network bundle, storage bundle. As CloudStack is a > integration project > it uses a lot of thirdparty libraries, spending too much time in repackaging > libraries I thought > should be my linkage dependency to OSGI service distracts us from main > business. > I know there are some projects repackaging common java libraries, I just > simply suspect if > it is the best practice for plugin system. > > As time goes on, more and more projects are packaging jar files as osgi bundles anyway (things are much better than they were 3 years ago). However, there are quite a few resources for libraries that are not natively packaged as osgi such as the servicemix bundle packages (http://repo1.maven.org/maven2/org/apache/servicemix/bundles/) and the Springsource enterprise bundle repository (http://ebr.springsource.com/repository/app/), if all else fails, there are some fairly consistent methods for converting jars to bundles using maven. > > And about extender. Now CloudStack itself is responsible for loading all > components and starting them. > As I know in OSGI this is work of extender, does it mean we have to write our > extender? I know we can > use dependency in MANIFEST file, but I don't know how it guarantees loading > order. To me using scattering files > to describe inter-dependency to guarantee loading order makes people hard to > know the exact loading sequence. > And what I dislike MANIFEST file most is you have to recompile every time you > change it. > > If I understand your concerns correctly, osgi exists as a solution to these problems, not to create them. I think that is why Alex raised the possibility of using osgi in place of a project specific component model, to lower the bar for those wishing to contribute and to expand the community to the many projects (both inside and outside Apache) using osgi as a standard component model. I'm not sure it is worth diving into each one of your technical issues individually, but I will be happy to do so if it adds to the discussion. Instead, I recommend looking at some of the projects I mentioned previously for some great examples of osgi adding value and solving complex issues. Osgi gives you a standard component model and related services that are influenced by a diverse range of requirements, how and the extent to which you make use of it to assemble your systems is completely up to you. > > > > > > 2. If OSGi, whether to use Equinox or Apache Felix? > > > > I am biased having been involved with Felix and Karaf for many years, but I > > would encourage you to look at Karaf and Felix as a base for the container > > (eplained more below). You will get assistance from within the Apache > > community by way of Karaf, Felix, and several other projects that also use > > these frameworks (Geronimo, Camel, Sling, probably many others now) and > > those projects will mutually benefit from input and contributions via > > Cloudstack. > > > > > Does Apache have any OSGI runtime to ease developing? I am not sure if I use > correct notion. I mean something like Eclipse Virgo that builds on OSGI > engine. > > Apache Karaf is a convenient packaging of value added bundles, services, etc that sits on top of Apache Felix (OSGi framework implementation) or Equinox. I think Karaf is what you are looking for. > > > > > > 3. The tool to help integrate the OSGi framework into tomcat (Sling?) > > > > I recommend considering an inversion of this and instead look at hosting > > your > > web server inside the osgi container (or something along these lines). In > > this > > way, agents and servers could be modularized similarly. You can certainly > > use > > an osgi framework to provide modularity embedded in a standalone java app > > or Tomcat, but I think you will find a lot of rewards in bootstrapping your > > jvm > > with osgi as a container (see Karaf, ServiceMix et al). > > > > > I agree. Making CloudStack standalone allows us to integrate with other web > server like > Jetty. However, I know Eclipse Virgo builds a Tomcat in its OSGI runtime. > However it's > based on Spring that doesn't fit in current CloudStack. > > In Karaf, you can install the Jetty bundle (it is a native osgi bundle) and deploy war files that have access to osgi services and bundles. Karaf also includes spring dm support (spring osgi extensions) as well as the osgi blueprint spec, which is a spec for xml based dependency injection specific to osgi. The point of all of this is that you can have all of these coexist at the same time because they are abstracted by the osgi component model. The main point I want to make is that the issues you raise here are implementation details that will affect you no matter what component model you choose. Your choice of component model should be based on your current and future needs. Alex has stated the requirements in the initial email and there are many of us working on other Apache projects with the same needs, also using osgi. As Alex mentioned in his email, adopting a standard component model such as osgi will help you scale project development, as well as the community. > > > > > > 4. Anything we can use to upgrade/downgrade database versions? > > > > Not exactly sure what you are asking here, but if you mean DB migration, I > > can;t think of anything osgi related that applies here so I think you will > > have > > to roll your own solution. > > > > > > > > Please let me know your feedbacks. > > > > > > --Alex
