I agree with Keith. prstat is well integrated to OpenSolaris and to me it would be definitely worth to learn what functionality prstat misses comparing to top. I am not convinced that the best solution is just to bring top to SFW because SA from other NIXES uses it. Maybe that much easier way would be just to provide enough information for these SA that prstat is there and how to use it. This may be also much cheaper solution considering long term future. I do not think that because such a such tool is available in such a such OS then this tool has to be integrated to OpenSolaris as well.
Besides, if someone wants to use top just now then there is one in CCD (3.5.1 as mentioned by David Comay) and another can be downloaded from Blastwave (version 3.6 -- I just installed it to check its version). I do not think that SFW future was decided yet and because SFW is now still considered as rather core part of OpenSolaris then we should be careful concerning what tool to bring in it. (And yes, I read all the discussions concerning future of SFW x CCD x Blastwave.) Lukas Keith M Wesolowski wrote: > On Fri, May 11, 2007 at 08:23:59PM -0400, brian.gupta at gmail.com wrote: > >> I would bet that a majority of SAs from all the other NIXES all use top. > > I'm sure they do, since they don't have anything better. > >> I vote to put it in unless it has a namespace collision. > > If this is the path we're going to start down, I'd like to understand > the criteria we're going to use in deciding that the comfort of > immigrants trumps coherence of strategy. For example, if we learned > that immigrants from GNU/Linux "just want SystemTap" would we > integrate it despite already having a superior tool in DTrace? > > It seems worth spending some time now to understand how we will choose > between helping immigrants find and learn the OpenSolaris Way of > achieving their goals and giving them the familiar, but often inferior > and certainly less well-integrated, tools. It would also be worth > understanding whether we'd like to learn why people prefer other tools > (when they have reasons other than finger memory) and improve our own. > That is, if we recognise a certain set of circumstances in which we'd > prefer to integrate third-party tools rather than enhance our own, is > that a set of circumstances in which we should also remove our > existing tools (since we presumably have no interest in improving them > to meet people's needs)? > > This, I think, is really the core of my fear for OpenSolaris's future: > what value are we providing if instead of learning what people want > from their tools and then providing solutions in a well-integrated and > architectually sane way, we simply integrate a set of tools we believe > are familiar to users of other operating systems? I simply don't > believe a sufficient reason has been offered for integrating top (all > of the things people have mentioned liking about top could easily be > implemented in prstat, either by default or as a compatibility mode), > if one posits that a desire to have a tool simply because it's the de > facto standard on other systems is sufficient reason to overlook > architectural consistency and completeness of integration, I don't see > any effort to establish a clear limit to the applicability of that > desire. >
