Sergey Matveychuk wrote:
Russell Jackson wrote:
I'm debugging a problem with portupgrade being used as part of an
automated package
handler within puppet (http://puppet.reductivelabs.com).
The problem is that when the handler executes portupgrade as an
external process from
ruby, portupgrade hangs with a script process taking up 100% of the
cpu. I suspected that
script isn't designed to be run without a controlling tty. So, I
executed portupgrade like
so to test:
portupgrade --new --batch --yes <port> <&- >&- 2>&- &
To my surprise, portupgrade exits immediately with a status of 0. I
found an old pr
(bin/56166) describing similar behavior. It looks like script was
fixed, but portupgrade
still isn't happy.
This happens with both ports-mgmt/portupgrade and
ports-mgmt/portupgrade-devel.
Suggestions?
portupgrade was fixed to work when tty is not a physical device but
file or so.
I think you don't want to run portupgade this way really. You have a
possibility to brake everything.
I'm not sure how well ruby's debug mode does compared to perl, or
how it works in comparison, but try preceding the portupgrade command
with '/usr/local/bin/ruby18 -d', to force portupgrade to run in debug mode.
-Garrett
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
