-
New Feature
-
Resolution: Fixed
-
Minor
-
None
-
-
3.30
Whenever there is a problem with a swarm slave that causes its disconnection the slave just disappears. This prevents on first acknowledging there is a problem with that particular slave and secondly finding out more from logs.
One possible solution is to have a "sticky" option in slave configuration so to make it persistent in case it goes offline.
Another possible solution would be to add in general configuration some time based options for "swarm" specific slaves:
- connection timeout (ex 15 min)
- removal timeout (ex: 1 day)
The basic idea is the same, to provide more transparency of the status of slaves instead of hiding them when something goes wrong.
- links to
Another use case for persisting swarm slave info is when you configure the slave (eg add tool locations). When a network hiccup occurs the slave automatically reconnects, but lost all its configuration.