> I've been testing a scenario with a time shift of an hour to
the past,
> and even though the watchdog detects the shift and restarts, the
> scheduled events are still stuck until the time passes.
What kind of scheduled events? The periodical tasks like enumeration or
dyndns updates?
In this case, the retry after a failure connecting to an LDAP server
Kind regards,
Victor
On Thu, Feb 9, 2017 at 9:37 PM, Jakub Hrozek <jhrozek(a)redhat.com> wrote:
On Thu, Feb 09, 2017 at 05:55:57PM +0100, Victor Tapia wrote:
> Hi list,
>
> I've been testing a scenario with a time shift of an hour to the past,
> and even though the watchdog detects the shift and restarts, the
> scheduled events are still stuck until the time passes.
What kind of scheduled events? The periodical tasks like enumeration or
dyndns updates?
>
> Would it make sense to use the same resetOffline/SIGUSR2 mechanism used
> for networking status changes to reschedule those events after the
> watchdog reset[1]?
>
> Kind regards,
>
> Victor
>
>
> [1]
>
> diff --git i/src/util/util_watchdog.c w/src/util/util_watchdog.c
> index 0df85b7..f981516 100644
> --- i/src/util/util_watchdog.c
> +++ w/src/util/util_watchdog.c
> @@ -158,6 +158,9 @@ static void watchdog_fd_read_handler(struct
> tevent_context *ev,
> "[%d]: %s\n", ret, sss_strerror(ret));
> orderly_shutdown(1);
> }
> + if (getpid() == getpgrp()) {
> + kill(-getpgrp(), SIGUSR2);
> + }
> }
>
> int setup_watchdog(struct tevent_context *ev, int interval)
> _______________________________________________
> sssd-devel mailing list -- sssd-devel(a)lists.fedorahosted.org
> To unsubscribe send an email to sssd-devel-leave(a)lists.fedorahosted.org
_______________________________________________
sssd-devel mailing list -- sssd-devel(a)lists.fedorahosted.org
To unsubscribe send an email to sssd-devel-leave(a)lists.fedorahosted.org