From 63cc03e228e0fe40aa3b7a7f973b18b734fcd338 Mon Sep 17 00:00:00 2001 From: Sean Whitton Date: Sat, 27 Feb 2021 13:38:48 -0700 Subject: note about forking before SAVE-LISP-AND-DIE Signed-off-by: Sean Whitton --- doc/ideas.rst | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/doc/ideas.rst b/doc/ideas.rst index 02e1d95..4d6944d 100644 --- a/doc/ideas.rst +++ b/doc/ideas.rst @@ -21,14 +21,14 @@ Connections Then DEPLOY* can call that and error out if establishing the next hop requires LISP- but we only have POSIX-. -- :DEBIAN-SBCL could SAVE-LISP-AND-DIE and then immediately reinvoke the saved - image. That way, we have something that a cronjob can call to re-run the - deployment to ensure that all properties remain applied. Need to - think about how the property which sets up the cronjob will be specified in - consfigs -- does it make sense to allow passing in arbitrary deployments, or - do we only allow re-running exactly the same thing? If the former, the - saved image will need to take some sort of command line input telling it - what arguments to pass to DEPLOY*. +- :DEBIAN-SBCL could (fork and) SAVE-LISP-AND-DIE. That way, we have + something that a cronjob can call to re-run the deployment to ensure that + all properties remain applied. Need to think about how the property which + sets up the cronjob will be specified in consfigs -- does it make sense to + allow passing in arbitrary deployments, or do we only allow re-running + exactly the same thing? If the former, the saved image will need to take + some sort of command line input telling it what arguments to pass to + DEPLOY*. Data sources ------------ -- cgit v1.2.3