Supporting connected standby
Posted Jan 27, 2014 9:49 UTC (Mon) by iq-0 (subscriber, #36655)
The other case would be where you use SSH to login on a server and attach a screen session which contains your IRC client. In that case you'd probably want to enable protocol keepalive and the ssh client should be fixed to set an appropriate timer slack (or a workaround should be set in place until it's fixed). And of course data received from the server should still be handled normally since that wouldn't be subject to the infinite timer slack.
Although the timer slack could be considered infinite in most cases that means that it will probably be coalesced with another timer expiring around that time. One could even imagine a sort of master "max timer slack" setting that would ensure that the default "infinite timer slack" for a certains system would ensure that it would at still be processed in X time after the first such timer expired.
Many (unfixed) programs wouldn't necessarily degrade in functionality if they'd wake up at least within 1 minute or even after 5 minutes. This global max timer slack would thus enable the user/distribution to set an acceptable timer slack depending on the type of device and current status.
Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds