Hi out there,
I completely reinstalled chipermail about a month ago. Now I use Version: 3.1.1-1. Built: 2016-06-29-19:42. All configurations have been done manually, so there is no old config or something like that in place. Before I had Version: 2.5.0-4. Built: 2013-04-17-19:13 in use for quite a while. The config is very basic, both installations came as ovf-files for VMware from the offical homepage.
Problem:
Since the upgrade I am experiancing timeouts while sending email from my gateway to the next hop, an email server behind an vpn and under controll of a service provider.
Example:
qmgr[9037]: 9643ACB: from=<A(a)mydomain.de>, size=14519708, nrcpt=1 (queue active)
smtp[22929]: 9643ACB: to=<B(a)b.de>, relay=x.x.x.103 [x.x.x.103]:25, delay=144634, delays=144451/0.01/0.08/183, dsn=4.4.2, status=deferred (conversation with x.x.x.103 [x.x.x.103] timed out while sending message body)
The transferrate over my internet / vpn connection goes up to 900kb/sec in upload, other data passes the same way as my emails. The connection is established and the gateway is sending data much slower then possible, I would say most times less then 10kb/sec and it doesn't even vary during the transfer. After 600sec the service provider closes the connection, resulting in a timeout. A few trys later the email gets send, I'm not sure at what speed, at some point it just goes through. This happens during all daytimes, even in the evening when servers are not bussy.
Things I tryed:
- restarts everywhere
- changing ISP modem / testing bandwidth
- checking DNS
- comparing logs, it really all starts with change of gateway version
- I upgraded RAM of the ciphermail VM from 1 to 2gb. In result it seems to happen less often, but it is reproducable with emails size 20mb+
- I did not try yet to change the type of VM networkadapter from E1000 to flexible
Any solutions out there would be helpfull, I'm in the dark right now