At Wed, 16 May 2012 11:15:42 +0800, Liu Yuan wrote: > > On 05/16/2012 11:09 AM, MORITA Kazutaka wrote: > > > At Wed, 16 May 2012 10:37:24 +0800, > > Liu Yuan wrote: > >> > >> On 05/16/2012 09:54 AM, MORITA Kazutaka wrote: > >> > >>> -l, --loglevel specify the level of logging detail\n\ > >>> -p, --port specify the TCP port on which to listen\n\ > >>> -v, --vnodes specify the number of virtual nodes\n\ > >>> + -w, --enable-cache enable writecache (default)\n\ > >>> + -W, --disable-cache disable writecache\n\ > >>> -z, --zone specify the zone id\n\ > >> > >> > >> a single option --disable-cache isn't enough? > > > > Isn't it difficult to imagine that '-W' turns off the writecache > > without the '-w' option? > > > > > What do you mean? I don't get it. 'disable writecache' doesn't already > say it to disable writecache? And disables means it is enabled > otherwise. For e.g, '-f, --foreground make the program run in the > foreground' means without this option, we run sheep as background. Sorry, what I meant is that a single character option '-W' would make users imagine that the option enables write cache. Anyway, I don't mind to remove the '-w' option from this patch since it's doubtful that introducing '-w' makes '-W' more understandable. Thanks, Kazutaka |