Stephen Gran wrote:
On Sat, Dec 30, 2006 at 02:39:35PM +0000, G.W. Haywood said:
Hi there,
On Sat, 30 Dec 2006 Stephen Gran wrote:
I have seen freshclam fail several times when run as a daemon, but
all in exactly the same way - stuck in a read() on a network socket.
This problem has been fixed for some time, and I have not seen
another failure since the fix went in.
That's useful information, thanks. In the changelog I see this, which
seems a little too recent to be described as "fixed for some time".
======================================================================
Mon Aug 7 20:49:48 CEST 2006
...
- freshclam/manager.c: fix possible infinite loop when read() fails
in get_database(), spotted by Everton da Silva Marques
======================================================================
Is it the fix to which you refer?
It's actually more recent than that, I think.
Sun Nov 5 17:08:22 CET 2006
-----------------------------
* Bugfixes:
- freshclam: apply timeout patch from Everton da Silva Marques
<everton*lab.ipaccess.diveo.net.br> (new options: ConnectTimeout and
ReceiveTimeout)
I apparently applied it as a patch to the Debian packages slightly
earlier:
clamav (0.88.4-3) unstable; urgency=low
* Apply upstream freshclam timeout patch (closes: #334911, #382353)
-- Stephen Gran <[EMAIL PROTECTED]> Mon, 2 Oct 2006 19:47:06 +0100
So, in my mind, it has been fixed or almost 3 months. That might not
qualify as "some time", it's true.
I know I was seeing this and mentioned it in this list in November.
dp
_______________________________________________
Help us build a comprehensive ClamAV guide: visit http://wiki.clamav.net
http://lurker.clamav.net/list/clamav-users.html
