Security/Reviews/SettingsAPI: Difference between revisions

From MozillaWiki
Jump to navigation Jump to search
(Created page with "{{SecReviewInfo |SecReview name=Settings API }} {{SecReview}} {{SecReviewActionStatus |SecReview action item status=None }}")
 
No edit summary
Line 1: Line 1:
{{SecReviewInfo
{{SecReviewInfo
|SecReview name=Settings API
|SecReview name=Settings API
|SecReview target=<bugzilla>
{
"id":"750992,678695"
}
</bugzilla>
}}
{{SecReview
|SecReview feature goal=* a way to store system wide settings to a backend database
** set:get for settings
** Listen for changes
* Only certified apps have to settings API
** only certified apps can *write*
* Mainly it will only the Settings API
* List of settings:
** https://etherpad.mozilla.org/LuzVJ1iGuO
* Might expose specific settings to untrusted/trusted applications, for them to read them.
* Currently there are two permisions settings:
** read settings
** write settings
* Design might be:
** App would declare in the manifest, what settings it would want access to, based on the type of app
** Only certified apps could set them
*** I think we should allow some settings to be settable by trusted app
* Other apps could only read them
|SecReview alt solutions=* indexdb - this is per app rather than global
|SecReview solution chosen=* effectively a shared database
* add a layer of control for permissions
|SecReview threat brainstorming=* What security model has been implemented for the settings API?
** The current  implementation is similar to the SMS API and has a readonly and  readwrite permission flag per principal.URI.
*  Certified app vulnerability would have access to read or change any setting
** Could be an explicit permission for read/write settings (or even an explicit subset of settings). I.e. the list of settings that the app has access to would be enumerated in manifest.
** Even certified apps should be limited to the settings explicitly listed in its manifest -- improves auditability and a line of defense against compromised apps.
* (future threat?) Untrusted app reading sensitive data from a setting
** Expose only a subset of settings for untrusted app
** Review security & privacy implications of settings prior to exposing settings
* Denial of service - change a setting which breaks a feature (e.g. change RIL settings to stop phone functionality working)
** Mitigation: only certified apps can change settings
** Testing a certified app should include review of what settings it changes
* Storing some passwords in this settings database
** Settings are currently stored in plaintext encoded format
** No plan for a keystore at the moment
* [future threat] If we expose reading/setting to untrusted/trusted app, we should make sure, those app will not be able to reach settings they should not be able to reach. Given that it is a DB in the backend.
}}
}}
{{SecReview}}
{{SecReviewActionStatus
{{SecReviewActionStatus
|SecReview action item status=None
|SecReview action item status=In Progress
|SecReview action items=* Who :: What :: By when (Keep in mind all these things will be bugs that block the reivew bug, that blocks the feature bug)
* jonas :: cjones/gal Do we want to expose anything at all right now to the web?
* jonas :: Ask the B2G team (consumers of the API) what settings they need to read (and store?)
*pauljt :: Code review (both of the settings API and the consumers)
*pauljt :: Make certified apps specify which settings
}}
}}

Revision as of 19:46, 30 May 2012

Please use "Edit with form" above to edit this page.

Item Reviewed

Settings API
Target
   
     Full Query    
   
ID Summary Priority Status
678695 Settings API -- RESOLVED
750992 [Security Review] Settings API P2 RESOLVED

2 Total; 0 Open (0%); 2 Resolved (100%); 0 Verified (0%);

{{#set:SecReview name=Settings API

|SecReview target=

Full Query
ID Summary Priority Status
678695 Settings API -- RESOLVED
750992 [Security Review] Settings API P2 RESOLVED

2 Total; 0 Open (0%); 2 Resolved (100%); 0 Verified (0%);

}}

Introduce the Feature

Goal of Feature, what is trying to be achieved (problem solved, use cases, etc)

  • a way to store system wide settings to a backend database
    • set:get for settings
    • Listen for changes
  • Only certified apps have to settings API
    • only certified apps can *write*
  • Mainly it will only the Settings API
  • List of settings:
  • Might expose specific settings to untrusted/trusted applications, for them to read them.
  • Currently there are two permisions settings:
    • read settings
    • write settings
  • Design might be:
    • App would declare in the manifest, what settings it would want access to, based on the type of app
    • Only certified apps could set them
      • I think we should allow some settings to be settable by trusted app
  • Other apps could only read them

What solutions/approaches were considered other than the proposed solution?

  • indexdb - this is per app rather than global

Why was this solution chosen?

  • effectively a shared database
  • add a layer of control for permissions

Any security threats already considered in the design and why?

`

Threat Brainstorming

  • What security model has been implemented for the settings API?
    • The current implementation is similar to the SMS API and has a readonly and readwrite permission flag per principal.URI.
  • Certified app vulnerability would have access to read or change any setting
    • Could be an explicit permission for read/write settings (or even an explicit subset of settings). I.e. the list of settings that the app has access to would be enumerated in manifest.
    • Even certified apps should be limited to the settings explicitly listed in its manifest -- improves auditability and a line of defense against compromised apps.
  • (future threat?) Untrusted app reading sensitive data from a setting
    • Expose only a subset of settings for untrusted app
    • Review security & privacy implications of settings prior to exposing settings
* Denial of service - change a setting which breaks a feature (e.g. change RIL settings to stop phone functionality working)
** Mitigation: only certified apps can change settings
** Testing a certified app should include review of what settings it changes

  • Storing some passwords in this settings database
    • Settings are currently stored in plaintext encoded format
    • No plan for a keystore at the moment
  • [future threat] If we expose reading/setting to untrusted/trusted app, we should make sure, those app will not be able to reach settings they should not be able to reach. Given that it is a DB in the backend.

{{#set: SecReview feature goal=* a way to store system wide settings to a backend database

    • set:get for settings
    • Listen for changes
  • Only certified apps have to settings API
    • only certified apps can *write*
  • Mainly it will only the Settings API
  • List of settings:
  • Might expose specific settings to untrusted/trusted applications, for them to read them.
  • Currently there are two permisions settings:
    • read settings
    • write settings
  • Design might be:
    • App would declare in the manifest, what settings it would want access to, based on the type of app
    • Only certified apps could set them
      • I think we should allow some settings to be settable by trusted app
  • Other apps could only read them

|SecReview alt solutions=* indexdb - this is per app rather than global |SecReview solution chosen=* effectively a shared database

  • add a layer of control for permissions

|SecReview threats considered=' |SecReview threat brainstorming=* What security model has been implemented for the settings API?

    • The current implementation is similar to the SMS API and has a readonly and readwrite permission flag per principal.URI.
  • Certified app vulnerability would have access to read or change any setting
    • Could be an explicit permission for read/write settings (or even an explicit subset of settings). I.e. the list of settings that the app has access to would be enumerated in manifest.
    • Even certified apps should be limited to the settings explicitly listed in its manifest -- improves auditability and a line of defense against compromised apps.
  • (future threat?) Untrusted app reading sensitive data from a setting
    • Expose only a subset of settings for untrusted app
    • Review security & privacy implications of settings prior to exposing settings
* Denial of service - change a setting which breaks a feature (e.g. change RIL settings to stop phone functionality working)
** Mitigation: only certified apps can change settings
** Testing a certified app should include review of what settings it changes

  • Storing some passwords in this settings database
    • Settings are currently stored in plaintext encoded format
    • No plan for a keystore at the moment
  • [future threat] If we expose reading/setting to untrusted/trusted app, we should make sure, those app will not be able to reach settings they should not be able to reach. Given that it is a DB in the backend.

}}

Action Items

Action Item Status In Progress
Release Target `
Action Items
* Who :: What :: By when (Keep in mind all these things will be bugs that block the reivew bug, that blocks the feature bug)
  • jonas :: cjones/gal Do we want to expose anything at all right now to the web?
  • jonas :: Ask the B2G team (consumers of the API) what settings they need to read (and store?)
  • pauljt :: Code review (both of the settings API and the consumers)
  • pauljt :: Make certified apps specify which settings

{{#set:|SecReview action item status=In Progress

|Feature version=` |SecReview action items=* Who :: What :: By when (Keep in mind all these things will be bugs that block the reivew bug, that blocks the feature bug)

  • jonas :: cjones/gal Do we want to expose anything at all right now to the web?
  • jonas :: Ask the B2G team (consumers of the API) what settings they need to read (and store?)
  • pauljt :: Code review (both of the settings API and the consumers)
  • pauljt :: Make certified apps specify which settings

}}