On 23-Jan-2004 Ian Freislich wrote: > I'm sorry that I didn't keep the ,v file that was corrupted. Don't worry. Hardly anybody does. > I'm > not sure that the OS buffers explain this because the machine had > been rebooted several times during the period that cvsup had not > refetched the corrupted file. Once the file has been written out to disk, the OS buffers aren't a factor any more. At that point, the data is corrupted on disk. But the metadata indicates that the file is OK. If the metadata says the file is up-to-date then CVSup won't update it. It would be waaaay too time-consuming to scan each entire file on every update run, so that is skipped if the metadata indicates the file is up-to-date. All file update packages operate that way. JohnReceived on Fri Jan 23 2004 - 08:17:23 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:39 UTC