> On May 24, 2016, at 8:35 PM, Anders Darander <[email protected]> wrote: > > * Matthijs Vader <[email protected]> [160523 13:37]: > >> Hi again, fixed and found it myself already. There is an error while >> mounting that >> partition on the first boot because the mount point (/data) doesn't exist >> yet. >> Then, later during first boot, some application creates it. Which explains >> why >> it does work on the second boot. > >> Which brings me to a new question: > >> Is there are reason for mountall.sh to put all errors into /dev/null? See: >> https://github.com/openembedded/openembedded-core/blob/master/meta/recipes-core/initscripts/initscripts-1.0/mountall.sh#L20 > > Good question. To be honest, I don't know.
you need to ensure that wherever you want to route the logs to, should be mounted and ready for writing. /dev/null is easiest. > >> Or shall I send a -tiny- patch dropping that 2>/dev/null? > > As long as it doesn't produce too much noise (i.e. errors on normal > boots etc), sure do that. At least as long as you check that it goes > into the syslog / journal. > > That would probably help people debugging certain issues. > > Cheers, > Anders > -- > Anders Darander, Senior System Architect > ChargeStorm AB / eStorm AB > -- > _______________________________________________ > Openembedded-core mailing list > [email protected] > http://lists.openembedded.org/mailman/listinfo/openembedded-core
signature.asc
Description: Message signed with OpenPGP using GPGMail
-- _______________________________________________ Openembedded-core mailing list [email protected] http://lists.openembedded.org/mailman/listinfo/openembedded-core
