Hello,

Got a serious issue with RTP traffic and the new Vigor2927.

Setup is as follows:

Vigor2927 at site 1, linked to another Vigor2927 at site 2 via a LAN to LAN IPsec VPN. VPN is working as it should and both sites have strong internet connections.

Need to transfer streaming audio from site 1 to site 2. This is done with a hardware IP audio codec that sends the data to another IP audio codec (decoder). Audio data transfer is via the RTP protocol at around 1.5mbps - well within the capabilities of the internet and VPN connections.

Within a few seconds of the IP audio codecs turning on and transferring data, the Vigor 2927's lockup and ALL lan and WAN traffic stops. Basically, the 2927 is killed by this RTP data transfer, as if it is getting "swamped" just like a DoS attack.

Testing both the encoder and decoder at the same site on the same LAN, all is ok. So basically it appears that the VPN is the issue? What is most concerning however, is the fact that the Vigor2927 can be brought to its knees like this - no good for a critical network environment!!

I have checked and run through settings, including the firewall, but to no avail.

Any ideas? Don't want to have to rip these new 2927's out if I can help it!

Thanks!

Alex

LA2 wrote:
Hello,

Got a serious issue with RTP traffic and the new Vigor2927.

Setup is as follows:

Vigor2927 at site 1, linked to another Vigor2927 at site 2 via a LAN to LAN IPsec VPN. VPN is working as it should and both sites have strong internet connections.

Need to transfer streaming audio from site 1 to site 2. This is done with a hardware IP audio codec that sends the data to another IP audio codec (decoder). Audio data transfer is via the RTP protocol at around 1.5mbps - well within the capabilities of the internet and VPN connections.

Within a few seconds of the IP audio codecs turning on and transferring data, the Vigor 2927's lockup and ALL lan and WAN traffic stops. Basically, the 2927 is killed by this RTP data transfer, as if it is getting "swamped" just like a DoS attack.

Testing both the encoder and decoder at the same site on the same LAN, all is ok. So basically it appears that the VPN is the issue? What is most concerning however, is the fact that the Vigor2927 can be brought to its knees like this - no good for a critical network environment!!

I have checked and run through settings, including the firewall, but to no avail.

Any ideas? Don't want to have to rip these new 2927's out if I can help it!

Thanks!

Alex

Did you resolve this issue?

Seeing something similar, although not with the same use case. Essentially around the same time every day, give or take 20 minutes, the site loses WAN connectivity. The 2927 is semi-responsive, and a reboot fixes. There are two 2927s on site that were originally setup in HA Active mode, but we've since broken that and tried each router separately. Both have the same issue.

Hi,

Did you sort this, having the same issue.

Regards,

I can't find a ticket regarding this issue covering its resolution, either with the UK support or with the firmware team. Can anyone experiencing these issues please make a support ticket with DrayTek UK so that this issue can be resolved?

I think this sounds like an issue with the hardware acceleration, try disabling that to see if it resolves the problem.

fw 4.2.4 now available on www.draytek.com - doesn't appears to be available on .co.uk yet

Doesn't list this specific issue in the release notes but worth a shot if still having problems -

https://fw.draytek.com.tw/Vigor2927/Firmware/v4.2.4/DrayTek_Vigor2927_V4.2.4_release-note.pdf