Re: [libvirt] [PATCHv2 2/6] event: server RPC protocol tweaks for domain lifecycle events

On 02/11/2014 08:57 AM, Daniel P. Berrange wrote: > On Wed, Jan 29, 2014 at 10:49:22AM -0700, Eric Blake wrote: >> This patch adds some new RPC call numbers, but for ease of review, >> they sit idle until a later patch adds the client counterpart to >> drive the new RPCs. Also for ease of review, I limited this patch > > ACK > Thanks for the review. > > >> @@ -5068,5 +5085,25 @@ enum remote_procedure { >> * @generate: both >> * @acl: none >> */ >> - REMOTE_PROC_NETWORK_EVENT_LIFECYCLE = 315 >> + REMOTE_PROC_NETWORK_EVENT_LIFECYCLE = 315, >> + >> + /** >> + * @generate: none >> + * @priority: high >> + * @acl: none >> + */ >> + REMOTE_PROC_CONNECT_DOMAIN_EVENT_CALLBACK_REGISTER_ANY = 316, >> + >> + /** >> + * @generate: none >> + * @priority: high >> + * @acl: none >> + */ >> + REMOTE_PROC_CONNECT_DOMAIN_EVENT_CALLBACK_DEREGISTER_ANY = 317, > > I believe these ACLs need to be non-none now The way 'make -C src check-aclrules' works is by correlating all RPC calls back into their API names - but I'm not inventing any new API names. These new RPC calls are already covered by existing APIs, and the ACL checks performed there are already sufficient. But it turns out that it doesn't hurt to make these ACLs match the other register RPC numbers, so I'm inclined to squash this in, unless you think that generating unused functions in src/access/viraccessapicheck.c is not worth the pollution: diff --git i/src/remote/remote_protocol.x w/src/remote/remote_protocol.x index 982ab1f..26abcdd 100644 --- i/src/remote/remote_protocol.x +++ w/src/remote/remote_protocol.x @@ -5090,14 +5090,15 @@ enum remote_procedure { /** * @generate: none * @priority: high - * @acl: none + * @acl: connect:search_domains + * @aclfilter: domain:getattr */ REMOTE_PROC_CONNECT_DOMAIN_EVENT_CALLBACK_REGISTER_ANY = 316, /** * @generate: none * @priority: high - * @acl: none + * @acl: connect:read */ REMOTE_PROC_CONNECT_DOMAIN_EVENT_CALLBACK_DEREGISTER_ANY = 317,