> When the peer is informed by an autowhitelist instance
> both times for the data and aw argument are
> the assigned the time that should be the aw time.
The bug was introduced around the beginning of 2010:
1c1
< /* $Id: milter-greylist.c,v 1.222 2009/09/09 12:19:17 manu Exp $ */
---
> /* $Id: milter-greylist.c,v 1.227 2010/02/15 16:38:03 manu Exp $ */
37c37
< __RCSID("$Id: milter-greylist.c,v 1.222 2009/09/09 12:19:17 manu Exp $");
---
> __RCSID("$Id: milter-greylist.c,v 1.227 2010/02/15 16:38:03 manu Exp $");
319a320,321
> struct tuple_fields tuple;
>
326d327
< struct rcpt *rcpt;
328,331c329,337
< rcpt = priv->priv_rcpt.lh_first;
< pending_update(SA(&priv->priv_addr), priv->priv_addrlen,
< priv->priv_from, rcpt->r_addr,
< priv->priv_sr.sr_autowhite, TU_AUTOWHITE);
---
> tuple.sa = SA(&priv->priv_addr);
> tuple.salen = priv->priv_addrlen;
> tuple.from = priv->priv_from;
> tuple.rcpt = priv->priv_rcpt.lh_first->r_addr;
> tuple.autowhite = 0;
> tuple.updatetype = TU_AUTOWHITE;
>
> mg_tuple_update(&tuple);
>
I am attaching a patch which fixes that problem.
In addition, it contains a cosmetic patch to log the queue ID
for certain debug level log message.
Best,
AttilaMessage
RE: [milter-greylist] Why autowhite isn't distributed over all MX?
2018-06-10 by Bruncsak, Attila
Attachments
- No local attachments were found for this message.