From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: multipart/mixed; boundary="===============5938259357454238048==" MIME-Version: 1.0 From: Denis Kenzior Subject: Re: [PATCH 2/2] Added SQLite history plugin Date: Wed, 07 Apr 2010 12:55:36 -0500 Message-ID: <201004071255.36407.denkenz@gmail.com> In-Reply-To: List-Id: To: ofono@ofono.org --===============5938259357454238048== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Hi Niko, > On Wed, Apr 7, 2010 at 7:33 PM, Denis Kenzior wrote: > [...] > = > >> A similar lack of robustness exists at the modem API side as well, if > >> oFono crashes/runs out of battery after receiving the message from the > >> modem driver the message is lost. There is no provision in the SMS mod= em > >> API to signal back to the modem driver that the message has been > >> successfully stored on the APE. > > > > If oFono crashes/runs out of battery here the driver is gone along with > > oFono. Notifying it of anything is not going to help ;) > = > Yes, anyway I'm thinking about the case when the storage system is > full, or an other error occours while trying to store the message. > In this case the driver should notify the network with a NACK, while a > signal should notify the user that the storage is full or there is > another problem. If your storage is full then you have other problems. Adding nack/ack = capability would increate the complexity of the driver/core about 10 fold, = so = this is not something we're willing to consider right now. If this is an = issue, then the user should be notified well in advance to rectify the = situation or face possible data loss. Regards, -Denis --===============5938259357454238048==--