ctlinnd reload all reason kills innd

[William F. Maton]

On Fri, 1 Mar 2002, Jeffrey M. Vinocur wrote:

> On Fri, 1 Mar 2002, William F. Maton wrote:
>
> > 	Executing ctlinnd reload all reason kills innd....whoops!
>
> Ooh, not again.  I can't reproduce this -- can you try issuing the
> different sorts of reload one by one (they're listed in the ctlinnd
> manpage) and track down which one it is?

It's:

				``history''

> We were seeing something like this in the early days of the IPv6 patch,
> and it turned out that some part of the original patch that Marco hadn't
> finished was returning a static string, and reloading was causing us to
> attempt to free() static data.  But we fixed that...

Mar  1 20:13:01 mcast innd: SERVER can't close history
/var/news/db/history: Interrupted system call

> But here's the weird thing.  As far as I can tell the only place that
> generates that sort of message is history/hisv6/hisv6.c:hisv6_closefiles()
> and as of late it has an "errno != EINTER" wrapped around it.  So I have
> no idea how I'm getting this error.
>
> I lost track of the thread where we were debugging this.

Something's definately not liking it.....

wfms