Hi Sean,
Again, you are correct. I have been watching my server for the last
hour or so now and it is operating perfectly.
best wishes
Tony White
On 4/03/2017 16:37, Sean P. Murphy wrote:
OK, everyone.
On my CentOS 5 main and test servers, I deleted the old database files and ran
freshclam.
After the files had downloaded, I restarted clamd without re-enabling it in simcontrol to see if the compile errors
started up again. After a few runs through clam's self check, no errors present, so I re-enabled clam in simcontrol and
everything is operating as it should be. I'm going to stay up for a bit and watch it, but I think the latest daily build
has fixed things.
Come Monday, it's time to find a new server and start preparing to set up a new
toaster.
Stay safe out there.
On Mar 4, 2017, at 12:00 AM, Sean P. Murphy <[email protected]
<mailto:[email protected]>> wrote:
Tony,
This is an issue with a signature in the daily clamav database update. It affects servers running on CentOS 5 due to
the fact that they have a version of pcre that clamav can't link to properly to form the database. CentOS 6 and 7
aren't affected. Supposedly the clamav folks have built a version of daily that won't break mail delivery, but I don't
know for sure. If you turn off clamav in simcontrol and recompile qmail cab, then mail delivery will resume unaffected.
You won't have virus scanning on incoming mail, but at least delivery will work.
Check the clamav mailing list: lists.clamav.net/pipermail/clamav-users <http://lists.clamav.net/pipermail/clamav-users>
for a better explanation.
They won't support clamav 99.3 on pre-CentOS 6 installs, so looks like it's
finally time to
upgrade those old servers, eh?
On Mar 3, 2017, at 10:26 PM, Tony White <[email protected]
<mailto:[email protected]>> wrote:
Hi folks,
This is Tony sorry to have to use a Yahoo acount but my mail server
broke this morning at 8am.
Submission and SMTP servers are now responding to connections
with this error...
qmail-smtpd: qq soft reject (mail server temporarily rejected message (#4.3.0))
I have made no changes so I am at a loss to the issue. Clamd is at 100% on
a single core might that be the issue?
TIA.
Tony
