Re: notice: send failed: Permission denied

2021-04-05 Thread Caroptions Caroptions via Unbound-users
Important note, the firewall block rule is reject, not block ...

Thanks,
John


From: Unbound-users  on behalf of 
Caroptions Caroptions via Unbound-users 
Sent: Monday, April 5, 2021 2:20 PM
To: unbound-users@lists.nlnetlabs.nl 
Subject: notice: send failed: Permission denied

Hi,

Probably it is discussed already, then sorry for reiterating the same problem, 
but I couldn't find solution.

unbound 1.13.1

I block certain ASNs/IPs on firewall. unbound starts normally, then after some 
time flood log with messages:

unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53
unbound[90575]: [90575:2] notice: send failed: Permission denied
unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53
unbound[90575]: [90575:2] notice: send failed: Permission denied
unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53
unbound[90575]: [90575:2] notice: send failed: Permission denied
unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53
unbound[90575]: [90575:2] notice: send failed: Permission denied
unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53

the SAME ip for hours. My firewall process CPU load jumps and stays on high 
level. unbound process CPU load high as well.

My temporary workaround is adding:

do-not-query-address: xx.xx.xx.xx

When I add new ip to this list it stays normal for some time till unbound find 
new NS server IP which is blocked on firewall and all loads jumps and flood log 
with "notice" messages.

In my understanding unbound should stop attempting to contact specific NS if it 
is not reachable/down?

Thanks,
John





notice: send failed: Permission denied

2021-04-05 Thread Caroptions Caroptions via Unbound-users
Hi,

Probably it is discussed already, then sorry for reiterating the same problem, 
but I couldn't find solution.

unbound 1.13.1

I block certain ASNs/IPs on firewall. unbound starts normally, then after some 
time flood log with messages:

unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53
unbound[90575]: [90575:2] notice: send failed: Permission denied
unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53
unbound[90575]: [90575:2] notice: send failed: Permission denied
unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53
unbound[90575]: [90575:2] notice: send failed: Permission denied
unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53
unbound[90575]: [90575:2] notice: send failed: Permission denied
unbound[90575]: [90575:2] notice: remote address is xx.xx.xx.xx port 53

the SAME ip for hours. My firewall process CPU load jumps and stays on high 
level. unbound process CPU load high as well.

My temporary workaround is adding:

do-not-query-address: xx.xx.xx.xx

When I add new ip to this list it stays normal for some time till unbound find 
new NS server IP which is blocked on firewall and all loads jumps and flood log 
with "notice" messages.

In my understanding unbound should stop attempting to contact specific NS if it 
is not reachable/down?

Thanks,
John





default reply when not resolved

2021-04-05 Thread Luiz Fernando Softov via Unbound-users
Hi,

I have a server and want to reply a default IN A to some queries, if the
recursive was not found in Forward, local-zone, etc.

Example:
The user tries to resolve A for www.example.com.
If I was able to resolve, send back the reply 192.168.1.1.
If not, reply with another info, like 172.16.1.1

So, when users put something wrong, like w.example.com, www.gooogle.com,
or netflixx.com, etc.

I want this working for all domains. not only those in example.
This way I can send private server IP, and in HTTP requests show some info
about the wrong domain.

There is a way to configure something like that?


unbound becomes stale after transport interface flap

2021-04-05 Thread Zeus Panchenko via Unbound-users
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

hi

I experience pretty same behavior of unbound: often it becomes stale after
transport interface flap

In my VPN I have DNS server which serves local zones.

For clients I configure unbound to forward requests for those zones to
that DNS server, and when VPN interface flaps of re-keying occures,
unbound misbehaves and nothing valuable appears in log files


so, please advise, where to look at?


- ---[ unbound.conf quotation start ]---
domain-insecure: "abc."
...

private-domain: "abc."
...

local-zone: "abc." transparent
...

include: /var/unbound/conf.d/*.conf
- ---[ unbound.conf quotation end   ]---

- ---[ conf.d/stub-zones.core.conf quotation start ]
stub-zone:
name: "abc."
stub-addr: 1.2.3.4
stub-prime: yes

stub-zone:
...
- ---[ conf.d/stub-zones.core.conf quotation end   ]

- -- 
Zeus V. Panchenko   jid:z...@im.ibs.dn.ua
IT Dpt., I.B.S. LLC   GMT+2 (EET)
-BEGIN PGP SIGNATURE-

iF0EARECAB0WIQQYIXL6FUmD7SUfqoOveOk+D/ejKgUCYGr+gAAKCRCveOk+D/ej
KumZAKCT7+uaGIG09Lj0i6NvAEJApWNJZACg7vBh+uve4tCWu6sbfCq4tGAP3PI=
=Lb31
-END PGP SIGNATURE-