Re: generalising oxenstored for unix nodes too

[Dave Scott <Dave.Scott@xxxxxxxxxxxxx>]

On Feb 15, 2012, at 10:45 AM, "Anil Madhavapeddy" <anil@xxxxxxxxxx> wrote:

> On 15 Feb 2012, at 17:37, Thomas Gazagnaire wrote:
> 
>>> Ok, just going with making it Lwt+functor for now then, as is
>>> the quickest. I had a vague idea that we might be able to hang
>>> arbitrary continuations as the value for an entry, rather than just
>>> strings, so they could be woken up by the scheduler.
>>> However, thats step 2 :)  Just having it working in a stub domain
>>> for now would be most useful.
>> 
>> You don't need lwt to make it work as stub donain, as the scheduler is 
>> hand-made and do not use threads. But indeed, that would be a nice 
>> improvement (but do you really need that?)
>> 
> 
> It's hard to compose two blocking events without it. So strictly
> speaking you could do a single oxenstored without Lwt, but even a
> blocking console ring would introduce problems then.
> 
> I suspect in production that an oxenstored will also need a management
> netfront or a logging ring of some kind, and so Lwt will be needed for
> that.

I agree - logging to a console ring would be essential. We can then tell 
xenconsoled to log the console output directly to syslog. We'd want that to be 
non-blocking or the system might grind to a halt under load (it's probably 
better to drop some logs and report you've done so and to continue, if the 
logging buffer fills up)

Cheers,
Dave