Many timeout-failures for DNS over TCP
Hi folks, I setup measurements for DNS queries over both TCP and UDP. I found the timeout and failures of TCP is far more than UDP. For example : DNS over TCP: https://atlas.ripe.net/measurements/8552847/#!probes DNS over UDP: https://atlas.ripe.net/measurements/8552846/#!probes Is it normal ? I mean is there any special timeout routine/parameters of atlas probes from normal resolver? Best regards, Davey
Hi Davey, There used to be a message on the atlas.ripe.net frontpage, I retrieved it from google cache: Due to an unfortunate bug in a recent probe firmware (version 4760), TCP based DNS measurements are not providing correct results. This affects all ongoing and one-off measurements. The issue also affected DNSMON measurements between 14-23 March. On 23 March we deployed a corrected firmware on RIPE Atlas anchors (version 4770), and in the near future we'll deploy it on regular probes as well. Our excuses for the inconvenience caused!? So any probe that is still on version 4760 will provide timeouts for TCP DNS lookups. Cheers, Wouter ________________________________ From: ripe-atlas <ripe-atlas-bounces@ripe.net> on behalf of Davey Song(宋林健) <ljsong@biigroup.cn> Sent: Tuesday, May 9, 2017 10:03 AM To: ripe-atlas@ripe.net Subject: [atlas] Many timeout-failures for DNS over TCP Hi folks, I setup measurements for DNS queries over both TCP and UDP. I found the timeout and failures of TCP is far more than UDP. For example : DNS over TCP: https://atlas.ripe.net/measurements/8552847/#!probes DNS over UDP: https://atlas.ripe.net/measurements/8552846/#!probes Is it normal ? I mean is there any special timeout routine/parameters of atlas probes from normal resolver? Best regards, Davey
For reference, it still appears on https://atlas.ripe.net/ (https://atlas.ripe.net/) for me - Tom May 9, 2017 9:14 AM, w.b.devries@utwente.nl (mailto:w.b.devries@utwente.nl) wrote: Hi Davey, There used to be a message on the atlas.ripe.net frontpage, I retrieved it from google cache: Due to an unfortunate bug in a recent probe firmware (version 4760), TCP based DNS measurements are not providing correct results. This affects all ongoing and one-off measurements. The issue also affected DNSMON measurements between 14-23 March. On 23 March we deployed a corrected firmware on RIPE Atlas anchors (version 4770), and in the near future we'll deploy it on regular probes as well. Our excuses for the inconvenience caused! So any probe that is still on version 4760 will provide timeouts for TCP DNS lookups. Cheers, Wouter ------------------------------------ From: ripe-atlas on behalf of Davey Song($BAWNS7r(B) Sent: Tuesday, May 9, 2017 10:03 AM To: ripe-atlas@ripe.net (mailto:ripe-atlas@ripe.net) Subject: [atlas] Many timeout-failures for DNS over TCP Hi folks, I setup measurements for DNS queries over both TCP and UDP. I found the timeout and failures of TCP is far more than UDP. For example : DNS over TCP: https://atlas.ripe.net/measurements/8552847/#!probes (https://atlas.ripe.net/measurements/8552847/#!probes) DNS over UDP: https://atlas.ripe.net/measurements/8552846/#!probes (https://atlas.ripe.net/measurements/8552846/#!probes) Is it normal ? I mean is there any special timeout routine/parameters of atlas probes from normal resolver? Best regards, Davey
Thanks for the information. It seems to me that the comparison on failures between TCP and UDP make no sense at this moment. Davey 发件人: ripe-atlas [mailto:ripe-atlas-bounces@ripe.net] 代表 Tom 发送时间: 2017年5月9日 16:33 收件人: ripe-atlas@ripe.net 主题: Re: [atlas] Many timeout-failures for DNS over TCP For reference, it still appears on https://atlas.ripe.net/ for me - Tom May 9, 2017 9:14 AM, w.b.devries@utwente.nl wrote: Hi Davey, There used to be a message on the atlas.ripe.net frontpage, I retrieved it from google cache: Due to an unfortunate bug in a recent probe firmware (version 4760), TCP based DNS measurements are not providing correct results. This affects all ongoing and one-off measurements. The issue also affected DNSMON measurements between 14-23 March. On 23 March we deployed a corrected firmware on RIPE Atlas anchors (version 4770), and in the near future we'll deploy it on regular probes as well. Our excuses for the inconvenience caused! So any probe that is still on version 4760 will provide timeouts for TCP DNS lookups. Cheers, Wouter _____ From: ripe-atlas <ripe-atlas-bounces@ripe.net> on behalf of Davey Song($BAWNS7r(B) <ljsong@biigroup.cn> Sent: Tuesday, May 9, 2017 10:03 AM To: ripe-atlas@ripe.net Subject: [atlas] Many timeout-failures for DNS over TCP Hi folks, I setup measurements for DNS queries over both TCP and UDP. I found the timeout and failures of TCP is far more than UDP. For example : DNS over TCP: https://atlas.ripe.net/measurements/8552847/#!probes DNS over UDP: https://atlas.ripe.net/measurements/8552846/#!probes Is it normal ? I mean is there any special timeout routine/parameters of atlas probes from normal resolver? Best regards, Davey
On 2017/05/09 10:03 , Davey Song(宋林健) wrote:
I setup measurements for DNS queries over both TCP and UDP. I found the timeout and failures of TCP is far more than UDP.
DNS over TCP: https://atlas.ripe.net/measurements/8552847/#!probes
Hi, Unfortunately, due to a bug, DNS over TCP measurements do not work in firmware 4760. I went over the results of this measurement. And the in most cases the probe was indeed still running 4760. Over time, probes upgrade 4770 and this issue will be fixed. There are a few probes that have a timeout but are running 4770. I created a TCP traceroute measurement using the same probes of measurement 8552847. https://atlas.ripe.net/measurements/8552847/#!probes In those remaining cases, the traceroute fails to complete. So for those probes there a network related issue. Philip
participants (4)
-
Davey Song(宋林健) -
Philip Homburg -
Tom -
w.b.devries＠utwente.nl