Is Java done?

Latest updates from all distro vendors. Java plugins no longer work on any browsers. Updated from Oracle, or the OpenJDK stack, or …

Doesn’t matter. Can’t get it to work anywhere.

This is a positive development … right? We can call this “experiment” over? Maybe all the nice folks who’ve been coding their IPMI/iLOM tools for years as Java clients will now … please … switch to HTML5 so we can drop this anachronism from our machines for once and for all? Please?

Heck, even Adobe has seen the light with Flash. Which arguably worked better for me than Java ever did as a browser plugin. They are going html5 going forward in order to “just work” everywhere.

/annoyed at borked plugins, and amazed that they are borked on so many different machines, all at once, in the same way.

Viewed 59495 times by 7320 viewers


7 thoughts on “Is Java done?

  1. Works fine on Debian (mostly unstable) with OpenJDK. I haven’t figured out the fuss, but I rarely run into applets. The Java Web Start goo used by certain IPMI folks, however, doesn’t work. And really cannot considering the various levels of network tunneling I need to reach the management ports.

  2. @Jason

    Unfortunately for us, if this is the only way we can reach a console remotely, it gets painful to do, but we have to do it.

    This is literally the only reason we have Java anywhere right now. Yeah, we are going to be doing Hadoop clusters, so we will be forced to use it there. And possibly for some of the dedup bits (leveraging existing code base, sadly in Java). But apart from these, the client side is pretty much toast for us for Java. No reason to consider it.

    The javaws stuff works for me, but the older iLOM/IPMI use the applet rather than the javaws. Javaws itself is a big steaming pile of bits. But it works better than the applet/plugin.

  3. My experience w/javaws v. plugins is the opposite. The plugin at least seems to inherit the SOCKS settings I use to tunnel (via ssh) into the host from which I can connect to the management interfaces. Instead I end up copying the downloaded jnlp over and running under VNC because Java hates my window manager. weee….

    The idea of *simplicity* in emergency tools has been lost. The command-line IPMI tools work ok for most things, but the serial console on our Intel boxes is flaky compared to the Java goo. bizarre. (And I shouldn’t be dealing with this at all, but there are too many arguments about who funds whom for which support. Needs to get done or else there are no funds for squabbling. sigh.)

  4. I’ve got to admit that with xCAT I very rarely have to mess with those nasty Java IPMI console tools – rcons/wcons just seems to work for us via SOL. Plus the conserver daemon that does that legwork logs everything on the screen to a per machine logfile, dead useful when a machine bounces itself and you want to go back and see whether the kernel (or BIOS) logged anything first..

  5. Oh, and I’ve seen folks at Oracle say that if it doesn’t work in OpenJDK then report it as a bug. They seem to be treating it (or want to be seen to be treating it – depending on your level of cynicism) as a first class citizen.

  6. @all

    The official Java plugins (e.g. Oracle) don’t work. The OpenJDK Iced Tea, which, in the past years I’ve tried it has never worked, suddenly have started working.

    Oh. Joy.

    To reiterate my request … can we all please just rewrite the console apps so that there is no Java requirement? Please?

  7. @Joe I can sympathise – I know that there is a reason that every Java app I’ve come across bundles its own JRE with it.. 🙁

    I don’t think xCAT needs Java at all for its remote console stuff.

Comments are closed.