On 04/06/2016 08:14, Eric Vidal wrote:
> The test is simple, i create a database with all i need. I create a
> second database with the same service less one. When i update the
> live database from first to second, all is good, s6-rc-update bring
> down the service which doesn't exit. But when i update the live
> database from second to first, s6-rc-update do not bring up the new
> service, but it's appear on the supervision tree.
Yes, that is working as intended. s6-rc-update tries to match the machine
states as closely as possible, so a new service will not be brought up
since it does not exist in the old state (so it is considered down in the
old state, and will remain down after s6-rc-update).
s6-rc-update has no memory of previous databases. It only knows the "old"
one (the current one you're replacing) and the "new" one. It doesn't know
that there was an "old old" database similar to the "new" one, with the
extra service up, so it cannot guess that you want that extra service up.
> According to
> your website, if a service that did not previously exist,
> s6-rc-update bring up automaticaly this new service.
Um, no. What part of the documentation made you think that? If the
documentation is unclear enough that you understood that, then I need to
fix it.
> I do not have any error message from s6-rc-update output with -v3,
> and i do not have any errors message on all-catch logger.
Everything is working as intended, so it's also intended that you have
no error message.
> I don't know how to make a race condition. Maybe you have an idea, or
> a way to make a race condition?
Why would you want to create a race condition? :)
--
Laurent
Received on Sat Jun 04 2016 - 09:46:59 UTC