Understood, thanks Andy.

Jonathan

On 18 December 2016 at 14:58, Andy LoPresto <[email protected]>
wrote:

> Simon's suggestion is the best out of the box solution. NiFi does not
> provide key storage, as that is not part of the core feature set of data
> flow.
>
> If you wish to integrate with an HSM or software-defined sensitive data
> vault, at this time, you'll need to write that connector yourself via
> whatever mechanism it exposes (Java API, PKCS#11, etc.).
>
> Andy LoPresto
> [email protected]
> [email protected]
> PGP Fingerprint: 70EC B3E5 98A6 5A3F D3C4  BACE 3C6E F65B 2F7D EF69
>
> On Dec 18, 2016, at 11:26, Jonathan Telfer <[email protected]> wrote:
>
> Hi Simon
>
> Thanks for the quick response. I was aware of the sensitive property
> (already using it for a password field) but I wasn’t aware it might be
> appropriate for storing the key material for the sha256 itself. Was
> expecting to have to use some kind of key store service.
>
> I’ll give that a try.
>
> Thanks
>
>
> On 18 Dec 2016, at 10:19, Simon Elliston Ball <[email protected]>
> wrote:
>
> This sounds like a job for sensitive properties to me. If you create a
> property on your custom processor and mark it with .senstive(true) in your
> builder, then the value stored in the flow conf will be available as a
> regular property, but encrypted on disk with a key specified in the
> nifi.properties. See https://github.com/apache/nifi/blob/master/nifi-nar-
> bundles/nifi-standard-bundle/nifi-standard-processors/src/
> main/java/org/apache/nifi/processors/standard/GetHTTP.java#L172 for an
> example of how this works.
>
> Simon
>
>
> On 18 Dec 2016, at 09:32, Jonathan Telfer <[email protected]> wrote:
>
> Hi all
>
> I'm currently writing a custom processor to access a non standard
> messaging protocol. The protocol requires each call to contain an
> authorization header signed by a sha256 hmac.
>
> I'm using commons HmacUtils to generate the signature, but I need advice
> on the best way to store the secret key and make it available to the
> processor.
>
> Thanks.
>
>
>
>

Reply via email to