[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Nokia 6234 + Gnokii 0.6.26 = data loss

From: Pawel Kot
Subject: Re: Nokia 6234 + Gnokii 0.6.26 = data loss
Date: Mon, 14 Jul 2008 13:14:02 +0200


On Sat, Jul 12, 2008 at 21:04, Gauthier <address@hidden> wrote:
> As root, I could successfully run "gnokii --identify", "gnokii
> --getphonebook SM 1 end", and "gnokii --monitor", which showed a line "SMS
> Messages: Unread 0, Number 1006".  But when I ran "gnokii
> --showsmsfolderstatus", I got the following output:
> No. Name                             Id #Msg
> ============================================
>  0 Inbox                            IN   36
>  1 Outbox                           OU   42
> With "gnokii --getsms IN 1 end" and "gnokii --getsms OU 1 end" I got indeed
> 36 and 42 messages, but these were very old messages which had been deleted
> a long time (two years!) ago and were not visible with the phone itself...
>  (Note that I did not change the phone or the SIM card for more than two
> years.)  I thought (don't know if I was wrong) that these messages were
> stored on the SIM card, and that the actual 1006 messages were stored in the
> phone memory (which, I thought, is not accessible with Gnokii, since you
> cannot retrieve the messages), so I ran "gnokii --deletesms IN 1 end" and
> "gnokii --deletesms OU 1 end" to clean up the messages in the SIM...  This
> deleted those 78 already deleted messages, but also deleted the 1006 "real"
> messages, without even warning about it!
> I don't know if I did something wrong (except issuing a "deletesms" command,
> but how could I guess that "deletesms" deletes messages that are not shown
> by "getsms"?), or if something went wrong.

I just can say we're very sorry about what happened. What we are doing
here is reverse engineering of proprietary Nokia protocol with no
documentation just with try-and-guess approach. Recently, with latest
series40 editions, Nokia changed frames for SMS handling. I did some
work on analyzing frames being sent. It indeed showed not the exact
messages as the phone did but it seemed good approximation. And I
admit I didn't test SMS deletion with that code (mostly because I
don't have such phone on my own and did testing remotely with not
large amount of time).

Thanks for the report -- I'll block deleting SMSes for these phones
for now. That's unfortunate that we found this out after causing data
loss. I again apologize for that and I assure you that we try as hard
as we can not to lead to such situations.

take care,
Pawel Kot

reply via email to

[Prev in Thread] Current Thread [Next in Thread]