When something is wrong in continuity

Sign-in does not work

The most common case, and it almost always has the same reason: sign-in runs through Microsoft, and Microsoft is what has just failed. The sign-in link by mail helps just as little, because it would go to the failed mailbox.

Anyone who has set a password signs in with it. Anyone who has not set one cannot catch up on that now, because resetting also runs through email. In that case only support can help.

That is the reason why the overview page constantly shows how many members still have no independent sign-in path. See preparation.

The continuity menu entry is missing

Three possible reasons.

Continuity is not activated, in which case no copies are created and there is nothing to show.

Visibility is set to only in an emergency and no outage is currently recognised. The entry then appears by itself as soon as one is recognised.

Visibility is set to hidden. Then only the administrators see the entry.

The view is empty although the outage is running

Nothing is waiting yet. Copies are created from the moment the outage is recognised, not retroactively. What came before was delivered normally and sits in the mailbox.

It is also possible that nothing was simply written to this person. The view shows only your own messages, never those of colleagues.

A sender says their mail comes back

At level 1 that is normal and not a fault: the delivering server receives a temporary rejection and tries again. Some servers tell their user after a few hours that delivery is delayed. The message is not lost in the process.

At level 2 that should not happen, because Conbool accepts there. If it happens nonetheless, two causes are likely: take-over is not switched on, or the recipient is not in the user directory. For unknown recipients no acceptance takes place.

A final non-delivery report only appears after five days without a successful delivery.

The mail server is running again but the mail does not arrive

After recovery all waiting messages are due immediately, and the catch-up keeps working without a pause until nothing is left. At usual volumes that takes minutes.

If it stays empty longer, it is worth looking at the continuity overview: it states how many messages are still open. If that number does not move, it is a case for support.

The outage notification never arrived

It goes to everyone who may configure continuity. If their mailboxes all sit in the failed domain, it lands in the continuity store, which you only find if you already know it exists.

The emergency contact address helps against that: an address outside your own domains. It is preferred when sending, because it is reachable at that moment.

The notification comes once per outage, not repeatedly. A new outage of the same domain notifies again.

A brief hiccup should not trigger continuity

It does not. Three consecutive unsuccessful delivery attempts are needed, and the first of them has to be at least five minutes old. A target system that stalls for thirty seconds does not meet that condition.

If a known permanent problem at the target system triggers it repeatedly nonetheless, the suspended mode is the right switch until the problem is fixed.

What support needs to know

Helpful are the period, the affected domain and whether take-over is switched on. Every outage is recorded with start, end, duration and the number of messages caught, so the details are in the overview.