Howdy, What every person defines as reasonable for their needs if subjective by definition. For example,
>Many people on the struts mailinglist and in private have told me >that resin is smart enough to be able to unpack a new version of a >war and delete any files which are no longer needed. For $500/server, now more expensive than Weblogic 8.1 express on one CPU, it better be able to do that ;) ;) Seriously though: I doubt Resin's claim, but I don't care to turn this into yet another Resin vs. Tomcat (vs. anything else) debate. That's been discussed many times on this and other lists, with the obvious conclusion that everyone uses what they like or what their organization/boss likes the best. ;) >After reading the first response on how to redeploy an app in tomcat >and then this one I just have to ask, does anyone else think that it >is kind of ridiculous what you have to do? Unless you run from the I don't. >war file, you have to remove the old war, remove the directory and >then remove the work directory. First, you are assuming that the >person deploying the app has permission to all of those directories >(which is a stretch). I don't think that's a stretch. In fact I think the opposite is a stretch: for the server administrator to not have access to the webapps or work directory seems unreasonable to me. But then again, see above about subjectivness. >Then, you are assuming that the application >can go down for more than 30 seconds while redeploying, which is >understandable, but still a stretch. I think this is a reasonable assumption as well. Moreover, it's proven more than reasonable in my experience. Again, your mileage may vary. Obviously some applications have operations which routinely exceed 30sec in length, so 30sec to release a new version is negligible. >I just hope that somebody stands back and look at how the developer >has to practically break his/her back to get a few changes made on a You've already done that. >running application. This really sucks. Imagine I change a few Then feel free to inspect and if needed rewrite the autodeploy code. Enhancing it to look for and remove the existing webapp directory shouldn't be very difficult. Same thing goes for the work directory. If enough people want this, it will happen eventually. >Except for the initial use of a war file to move an application from >one server to another, war files become more of a pain than a >convenience. I completely disagree. I find deploying and running from packed WARs easy and convenient. Moreover, it's proven easy and useful for times when I migrate containers. We routinely benchmark our app on several servers (including Resin 3.0 beta by the way) and find this to be the easiest way to deploy and run on multiple heterogeneous servers. >from the point. The point is that the way war files are handled >right now makes absolutely no sense. Strong words ;) As illustrated above, I don't agree with most of what you said, and I find the current autoDeploy to not only work well, but work conveniently. But your situation may be completely different and I have no doubt some of what you suggest other people will find useful as well. Contributions are always welcome ;) Yoav Shapira Millennium ChemInformatics This e-mail, including any attachments, is a confidential business communication, and may contain information that is confidential, proprietary and/or privileged. This e-mail is intended only for the individual(s) to whom it is addressed, and may not be saved, copied, printed, disclosed or used by anyone else. If you are not the(an) intended recipient, please immediately delete this e-mail from your computer system and notify the sender. Thank you. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
