> On Mon, 21 May 2007, Ravi Pratap wrote:
>
> > Here's a patch for a crash in c-ares on init, on a host that has no
> > network adapter, or has its cable unplugged. The crash really is
> > related to the status variable not being initialized
> properly in that code path.
>
> That patch puzzles me. If 'status' wasn't set properly
> previously, then won't the if line on line 740 be weird? I
> mean if the init is the problem, shouldn't the fix be in the init?

I just checked with the engineer who made this fix and I agree with you.
The fix should ideally go either in the init, or in the if block that's
failing to update status.

I'll get back to you with an updated patch.

Thanks,

Ravi.
Received on 2007-05-23