From mboxrd@z Thu Jan 1 00:00:00 1970 To: 9fans@cse.psu.edu Subject: Re: [9fans] [reminder] pointer to Plan 9 FAQ In-Reply-To: <20010916224901.ED3821998A@mail.cse.psu.edu> References: <20010916224901.ED3821998A@mail.cse.psu.edu> From: Richard Message-Id: Date: Mon, 17 Sep 2001 21:33:54 -0700 Topicbox-Message-UUID: ed61ddba-eac9-11e9-9e20-41e7f4b1d025 jmk@plan9.bell-labs.com writes: >I'm not sure what you're getting at here. Are you saying that other systems >have ways to prevent you from losing your data without copying it somewhere >else? It is my understanding that kfs has received much less testing than the dedicated fileserver and indeed the other filesystems that a home user might be accustomed to. one can imagine slow or subtle forms of data corruption that escape notice till it has spread to all the backup copies assuming the usual practice of keeping around 3 to 5 backup copies. It is my understanding that the main reason for this lack of testing is that having a nice dedicated fileserver, Bell Labs has little need for kfs. If you are saying that kfs is tested enough or indeed you and your peers at Bell Labs have stared at the code long enough to convince yourself that it has a very very low probability of corrupting data in ways that can escape user notice, then I withdraw my suggestion. (And maybe it would have been better if I had asked a question rather than made a suggestion.)