On 03/28/2011 01:34 PM, Michal Novotny wrote:
Hi Laine, thanks for your reply. Comments inline...
On 03/25/2011 10:02 PM, Laine Stump wrote:
I haven't had time yet to look at the code in detail, but thought I should send this preliminary commentary.
Hi, this is the patch to add DNS TXT record support to libvirt networking driver since this is feature that's supported by DNSMasq that's being used by the bridge driver.
Maybe you fail to understand the reasons why to implement such a feature however it's a good thing IMHO since user could provide some information in the DNS TXT record headers. As a matter of fact, I think that not only is this useful, but configuring other capabilities presented by dnsmasq would be good. I
On 03/24/2011 09:58 AM, Michal Novotny wrote: think you'll find a kindred spirit in Paweł Krześniak, who was also wanting some other dnsmasq capabilities exposed (I forget which now).
Well, I have to agree that there are some options/capabilities that would be good to be configurable.
It would be great to: 1) add <user-class> and <vendor-class> tags inside <dhcp> that allow filtering according to user/vendor classes 2) allow to specify <bootp> inside those as well as inside <range> or <host> elements. 3) add support for DHCP options besides bootp, with a tag like <option force="yes|no" name="..." value="...">. For example, my router's DHCP configuration would look like this: <dhcp> <range ...> <user-class prefix="iPXE"> <bootp file="http://playground.usersys.redhat.com/pxe/boot.ipxe"> </user-class> <bootp file="undionly.kpxe"> </dhcp>
The headers are, of course, configurable in the network XML description and the idea got to me when I was reading an article about DKIM (DomainKeys Identified Mail) since it's using TXT records in the DNS to provide the public keys. This inspired me to implement the DNS TXT record support to libvirt bridge driver to allow users expose some information to the guest if they want to do so etc.
Limitations: - Records names and values containing space (' ') arguments are altered to change spaces to underscores ('_'). This is because of proper argument handling when spawning dnsmasq.
Is this really necessary? We're not talking about a shell commandline here, but an array of null terminated strings. If it's a restriction placed by dnsmasq itself, then we should just disallow ' ' during parsing rather than silently changing it, to avoid surprises.
Well, that's the reason since if we quote this we can't use dig to dig it correctly without the quotes. This was simply the way to disallow ' ' by changing it to underscores since I didn't want the definition to fail because of this.
It must be possible to use record values containing a space. $ dig TXT gmail.com [...] ;; QUESTION SECTION: ;gmail.com. IN TXT ;; ANSWER SECTION: gmail.com. 300 IN TXT "v=spf1 redirect=_spf.google.com" Paolo