[Yeti DNS Discuss] Five additional servers added to Yeti testbed & an bug finding on priming response

Davey Song(宋林健) ljsong at biigroup.cn
Wed May 4 06:22:12 UTC 2016


Hi folks, 

 

Last week, we added 5 servers into Yeti testbed, beneath the top domain
yeti-dns.net. You can find them in the map and list
(http://yeti-dns.org/operators.html ).  Please update your hint file if you
run a yeti resolver.

 

Right now, we have 9 additional servers have common suffix with other name,
like yeti.eu.org, dns-lab.net, yeti-dns.net. We intentionally pick five
random labels (first 30 characters of SHA256([a-e])) to offset the effect of
name compression. So the names for the five servers are :

 

ca978112ca1bbdcafac231b39a23dc.yeti-dns.net

3e23e8160039594a33894f6564e1b1.yeti-dns.net

2e7d2c03a9507ae265ecf5b5356885.yeti-dns.net

18ac3e7343f016890c510e93f93526.yeti-dns.net

3f79bb7b435b05321651daefd374cd.yeti-dns.net

 

Now there are 23 severs in Yeti and the size of priming response is up to
1400. It is observed that more servers have wired behavior in priming
response after this adding(from 18 to 23). 

 


The number of root server : 18 server 

23 server (after this round of adding)


yeti-ns.ix.ru.
yeti-ns.lab.nic.cl.
yeti-ns.switch.ch.,
yeti.bofh.priv.at.

dose not return all glue information

dahu1.yeti.eu.org.

yeti-ns.as59715.net.

yeti-ns.conit.co.

yeti-ns.ix.ru.

yeti-ns.lab.nic.cl.

yeti-ns.switch.ch.

yeti-ns.wide.ad.jp.

yeti.aquaray.com.

yeti.bofh.priv.at.

yeti.ipv6.ernet.in.

dose not return all glue information

 

For example, dahu1.yeti.eu.org. returned all 18 server in additional section
with glue RR when the number of server is 18, but does not return the new
server glue information after 5 servers added into the system.

 

The case for yeti-ns.conit.co. is wired. after this adding it returns
nothing in answer and additional section no matter in UDP or TCP.

 

Best regard,

Davey

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.yeti-dns.org/pipermail/discuss/attachments/20160504/7a92364a/attachment.html>


More information about the discuss mailing list