RE: Timeout before data read with opendkim-2.3.2

From: Murray S. Kucherawy <msk_at_cloudmark.com>
Date: Mon, 16 May 2011 12:21:51 -0700

> -----Original Message-----
> From: opendkim-dev-bounce_at_lists.opendkim.org [mailto:opendkim-dev-bounce_at_lists.opendkim.org] On Behalf Of Gary Mills
> Sent: Monday, May 16, 2011 12:08 PM
> To: Murray S. Kucherawy
> Cc: opendkim-dev_at_lists.opendkim.org
> Subject: Re: Timeout before data read with opendkim-2.3.2
>
> Ah, I was reading the error message incorrectly. That certainly makes
> more sense now. The DKIM filter must have been unable to respond to
> the MTA within the timeout limit. I assume that opendkim was itself
> waiting for a response from the DNS server, perhaps for several
> responses.

Right, or the as-yet-unresolved apparent libar deadlock had struck.

> I wonder if opendkim could report what it was doing when the MTA
> declared a timeout? Would that help at all?

The only thing opendkim does that could incur a delay is any kind of I/O, such as querying a database or waiting for a DNS reply. If libar does indeed have a real deadlock problem, that would cause this as well.

> Is letting the MTA timeout occur the best way for opendkim to
> cooperate with the MTA, or is there a better way?

opendkim will attempt to tell the filter to keep waiting if it gets control back from the resolver periodically. Note the "if" there...

> I could check on how frequently this timeout occurs; it it's
> infrequent, it's not a problem.

The interesting thing to me is the previous report you gave about the seemingly random distribution of timeouts you see, when the configured timeout is five seconds.
Received on Mon May 16 2011 - 19:21:59 PST