How could I increase "runaway" timer for package build cluster?
rkoberman at gmail.com
Mon Mar 9 23:11:31 UTC 2015
On Mon, Mar 9, 2015 at 4:23 AM, Matthew Seaman <
m.seaman at infracaninophile.co.uk> wrote:
> On 03/09/15 11:11, Lev Serebryakov wrote:
> > On 08.03.2015 12:40, Ben Woods wrote:
> >> Actually, taking a look at my /usr/local/etc/poudriere.conf
> >> configuration file, I see these settings have indeed been exposed
> >> for configuration there without having to touch the poudriere code
> >> itself. The appropriate settings are:
> >> # This defines the max time (in seconds) that a command may run for
> >> a build # before it is killed for taking too long. Default: 86400
> >> #MAX_EXECUTION_TIME=86400
> >> # This defines the time (in seconds) before a command is considered
> >> to # be in a runaway state for having no output on stdout. Default:
> >> 7200 #NOHANG_TIME=7200
> > The problem is, it is not my cluster, but FreeBSD official cluster. I
> > could not change configs of it, I could only edit Makefile of my port.
> I added a small change to make 'pkg create' emit a progress bar which
> will be in pkg-1.15 -- you have to tweak some settings in pkg.conf to
> enable it though. With this change I've managed to avoid poudriere
> timeouts while packaging stuff with large numbers of files --
> texlive-texmf being a classic example. You can try it now by installing
> net-mgmt/pkg-devel, but beware that will eventually cause all your
> client machines to end up running pkg-devel too.
Just slightly confused. Do you mean pkg-1.4.15 (now on 1.4.12) or pkg-1.5?
Any idea on time-frame?
Kevin Oberman, Network Engineer, Retired
E-mail: rkoberman at gmail.com
More information about the freebsd-ports