12.11.2015 15:05, Alex Peshkoff wrote:
> No, it's not possible (at least I do not see such bug in current code).
I see.
IMHO, startCryptThread() is far from obvious place for initialization. May
be a comment
in the beginning, stating that this routine is called synchronously at every
sing
On 11/12/2015 04:52 PM, Dimitry Sibiryakov wrote:
> Hello, All.
>
> In code of CryptoManager I see only one place where variable "crypt" is
> initialized
> during normal operations. Is it possible that if encrypted database has no
> single
> encrypted page (yet), new pages will be written
Hello, All.
In code of CryptoManager I see only one place where variable "crypt" is
initialized
during normal operations. Is it possible that if encrypted database has no
single
encrypted page (yet), new pages will be written unencrypted as well? Say, if
initial
encryption thread faile
