On 18/11/2007 Dick Middleton wrote: >> I prepared cryptsetup packages with the proposed patch applied. could >> you give them a try? > > I just submitted this as a bug to udev (451702) completely forgetting the > workaround recently made to cryptsetup. So FYI: > > -- > [...] > > What seems to be happening is that cryptsetup is creating a new > device. As a result udev runs udevsettle but that doesn't exit until > timeout. Once udevsettle completes the run rule completes. > -- > > I suspect that udevsettle is the one you added to cryptsetup. Maybe a > timeout would fix this?
udevsettle indeed supports a --timeout= option, and this would workaround your delay. but still the invokation of udevsettle is a workaround for a race condition somewhere in device-mapper (if i got it right at all). udevsettle ensures that the temporary-cryptsetup-$$ devicefiles disappear after the real dm-crypt device has been setup. if we now add a timeout to udevsettle, this might be in conflict with the original goal, waiting until all udev events have finished. maybe the reason for your 3 minutes delay is that udevsettle has a builtin timeout of 3 minutes. this would mean that my doubts above would be baseless. in that case i have no problems with lowering the timeout of 30 seconds or something like that. greetings, jonas -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]