Crash protection
David Lee
t.d.lee at durham.ac.uk
Tue Mar 3 10:36:49 GMT 2009
On Mon, 2 Mar 2009, Julian Field wrote:
> On 2/3/09 14:04, David Lee wrote:
>> [...]
>> Given that I was a chief requester, naturally I feel honour-bound to
>> help test it. Could you give an indication of your production-level
>> confidence in the code? How high up the inbound MX-priority-tree
>> should I reasonably think of installing it at present?
> Start it off at the bottom (highest number) and intentionally kill it
> during a batch of messages. Do that twice and you should see
> "MailScanner --processing" start to print something. That way it should
> probably only be getting spam anyway, which will be an ideal test
> environment for it.
OK. Installed on an MX high-value (low-priority) machine.
Testing it with "MailScanner --debug" etc. took a little more than two
interrupts (perhaps successive "--debug" invocations were picking up
different emails) but it seems to be OK. "MailScanner --processing" is
showing some output for two or more attempts. So looking good! And in
the log file I'm seeing messages of the form "Making attempt 2 at ..."
which (in this controlled-test context) looks promising.
Thanks.
> Cheers for the idea in the first place! (even though I did change it a
> bit as I progressed... :)
You're welcome. The idea was simply a sketch of a possibility. Your
change was to deliver a working product! Many thanks.
An observation: "--processing" queries with "SELECT ... count>1". So on
a busy, but well-working, server its output is always empty. Could there
be a variant "count>0"? Thus on this same server there would generally be
output, continually changing, reflecting (more or less) the active part of
the inbound MS queue. Don't worry if you would rather not.
Anyway, we have it in service, and I'll keep monitoring it.
Many thanks again.
--
: David Lee I.T. Service :
: Senior Systems Programmer Computer Centre :
: UNIX Team Leader Durham University :
: South Road :
: http://www.dur.ac.uk/t.d.lee/ Durham DH1 3LE :
: Phone: +44 191 334 2752 U.K. :
More information about the MailScanner
mailing list