Summary: | /etc/init.d/ndpd fails when ntpd has been terminated abnormally | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Enrico 'nekrad' Weigelt <weigelt> |
Component: | Current packages | Assignee: | Gentoo Linux bug wranglers <bug-wranglers> |
Status: | RESOLVED INVALID | ||
Severity: | normal | ||
Priority: | High | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Enrico 'nekrad' Weigelt
2006-07-04 14:19:33 UTC
As I understand it, this is what the zap argument for init.d scripts is for. For example, you would want to run: $ /etc/init.d/ndpd zap start And your service would start again. This may not be ideal behavior, but it seems to be the behavior it is designed to do. I'm sure there are developers who would know a better answer (not being one myself). /etc/init.d/ntpd zap start (In reply to comment #2) > /etc/init.d/ntpd zap start > This doesn't really solve the problem, just a workaround. It requires manual intervention to explicitly call w/ "zap". If automatic stuff would use the "zap" option, it will force to start the service over and over, even if its really running. We should find a better way for it, ie. checks against pid-files, etc. Isn't there any (easy) way for using some pidfile instead (or additional to) the /var/lib/init.d/* files ? In baselayout-1.13 you can do this if ! /etc/init.d/ntpd --quiet status ; then /etc/init.d/ntpd --quiet start fi The status call checks that all daemons started by s-s-d are still running. If any are not then the service is "stopped" automatically and 1 is returned so the call to start works. Closing again, no idea why is this reopened. |