On 25/03/09 11:10 AM, Michael Schuster wrote: > On 03/25/09 09:27, Darren Reed wrote: >> On 25/03/09 09:19 AM, Michael Schuster wrote: > >>>> No, but it is easy, informative and much less prone to error. >>> >>> I think that's up for debate - making a mistake with vi is too easy, >>> for my liking. >> >> But the command line options being pursued offer *NOTHING* that is >> anything better. >> *NOTHING* > > I beg to differ - for one thing, the command line is validated as it's > entered, and an error is printed if necessary. That's much better > feedback than fumbling a change in some file and being told about it > sometime later.
So why do many people alias rm to "rm -i"? To prevent removing too many things when they do "rm * .o" (etc.) And people still do that, even though they have command line editing. The problem isn't the *validation* of input, it is the inability to undo anything once "Enter" is hit. The command line is a place where mistakes *often* get made. > (I can vividly remember a case where several people from Sun had to > listen to a customer complain bitterly and vehemently about SunCluster > (during a meeting with other providers, etc) - there'd been a switch > and not everthing had gone as expected. Once the meeting had concluded > its actual purpose, their admin slunk up to us and admitted he'd > forgotten to reverse a change he'd made in some config file. You don't > forget these "changes" on a commandline) I can't see this being any different for the command line. In your story he's forgot to undo some edit. The command line equivalent is forgetting to run some "undo" variation of a command you ran. >>>> A month or so ago, a few of us sat down with a group that does >>>> professional system administration and they were quite in favour >>>> of having configuration files that can be validated before loading >>>> rather than having to do everything on the command line. Why? >>> >>> >>>> Ever done "kill -9 - 1" by accident? >>>> Or "kill -9 % 1"? >>> >>> actually no :-) (you mean as root? that's the only case where it >>> would affect ilbd). I don't quite see how that pertains to this >>> discussion, please elaborate so I can understand the concern. >> >> The point is that recovering from finger trouble on the command line >> can be a whole lot harder than recovering from typing an address >> wrong and going "oops" as you review the configuration before >> applying it. > > I think we'll just have to disagree here - I find that it more natural > to use command-line editing to fix a command than to go back to a file. > >> The above isn't about whether those "kill" things affect ilb, it's >> whether or not you reboot the box by accident or kill init or not. > > if ilbadm had "powers" similar to kill, maybe I'd agree. as it is, I > don't think the comparison applies. Be creative and try and imagine ways in which a simple mistake could cause a big problem. What happens if you apply the load balancing to bge1 instead of bge2 and your ssh session stops responding because you were ssh'd in over bge1? Or you get the address wrong by 1 and instead of altering the flow of packets to a web server, you're altering packet flow to your box or a router? Darren
