From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dave Eckhardt To: 9fans@cse.psu.edu MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <10525.1120591352.1@piper.nectar.cs.cmu.edu> Date: Tue, 5 Jul 2005 15:22:32 -0400 Message-ID: <10526.1120591352@piper.nectar.cs.cmu.edu> Subject: [9fans] 423 missing clumps Topicbox-Message-UUID: 61c361c8-ead0-11e9-9d60-3106f5b1d025 I have an old venti (/bin/venti/venti is dated November 2003) whose bits I would like to transfer elsewhere as part of an upgrade. Because I'm paranoid I checked the venti index information available from http://127.1:8000/index against the output of rdarena and was disturbed to find that rdarena was seeing 423 fewer clumps than venti knew about. I tried venti/sync (followed by waiting) and also issuing "venti" on the fossil console to hang up and redial its connection, all to no avail. Then I read about the "clump info not on disk" issue discussed in . Sure enough, rebooting my fossi/venti machine caused the "missing" clumps to be visible to rdarena. Has venti been changed so that venti/sync forces the clump information out to disk? Should a warning be added to the ventiaux manual page, at least for people running old code? Dave Eckhardt