[Unbound-users] Resource temporarily unavailable and query problem with unbound

[Sebastian Nickel - Hetzner Online AG]

Hello,
I tried to increase so-rcvbuf to 8m and to 16m. I changed
net.core.rmem_max values in system, too.


I still see many of those lines in the log:

"""
Nov 09 16:34:26 unbound[3333:3] notice: sendmsg failed: Resource
temporarily unavailable
Nov 09 16:34:26 unbound[3333:3] notice: remote address is
xxx.xxx.xxx.xxx port 38777
Nov 09 16:34:26 unbound[3333:5] notice: remote address is
xxx.xxx.xxx.xxx port 45012
Nov 09 16:34:26 unbound[3333:4] notice: sendmsg failed: Resource
temporarily unavailable
Nov 09 16:34:26 unbound[3333:7] notice: sendmsg failed: Resource
temporarily unavailable
"""

Here is my unbound config (defaults not printed):

"""
server:

        so-rcvbuf: 16m
	verbosity: 1
	num-threads: 8
	interface: 0.0.0.0
	interface-automatic: yes

	outgoing-range: 8192
	msg-cache-size: 2150m
	msg-cache-slabs: 8
	num-queries-per-thread: 4096
	rrset-cache-size: 4300m
	rrset-cache-slabs: 8
	infra-cache-slabs: 8

	access-control: 127.0.0.0/8 allow_snoop
       
        <many access-control lines here>

	chroot: ""
	logfile: "/var/log/unbound.log"
        log-time-ascii: yes
	use-syslog: no
	pidfile: "/var/run/unbound.pid"
        do-not-query-address: 192.0.2.0/24
	key-cache-slabs: 8


remote-control:
	control-enable: yes 

"""

Here again the linked libs for our unbound 1.4.6:
"""
linked libs: libev 3.9 (it uses epoll), ldns 1.4.0, OpenSSL 0.9.8o
"""

The server has a quadcore CPU (with hyperthreading) and 12GB of RAM.


Are there any ideas what goes wrong?


Best regards,
Sebastian


Am Montag, den 08.11.2010, 19:58 +0100 schrieb Sebastian Nickel:
> Hello Wouter,
> > Are you out of socket-buffers?
> thank you for your answer. I will try to increase the so-rcvbuf on the
> system. The system is really heavy in use, so this could be a cause for
> those lines.
> 
> > The domain runs software that returns malformed packets with EDNS.
> It's a pity that there is still software which returns malformed
> packets. But now I know that it is not unbound's fault.
> 
> Thank you...
> 
> Best regards,
> Sebastian
> 
> Am Montag, den 08.11.2010, 19:35 +0100 schrieb W.C.A. Wijngaards:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> > 
> > Hi Sebiastian,
> > 
> > On 11/08/2010 05:05 PM, Sebastian Nickel - Hetzner Online AG wrote:
> > > [1289226893] unbound[5117:1] notice: sendmsg failed: Resource
> > > temporarily unavailable
> > 
> > Are you out of socket-buffers?
> > 
> > 
> > > Another problem I have is a "SERVFAIL" with the domain "imgw.pl".
> > 
> > The domain runs software that returns malformed packets with EDNS.
> > Bind 'accepts' them, but unbound does not.
> > 
> > dig @212.160.142.68 imgw.pl  +bufsize=1024
> > ;; Warning: Message parser reports malformed message packet.
> > ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 63400
> > ;; flags: qr aa; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 1
> > 
> > ;; QUESTION SECTION:
> > ;imgw.pl.			IN	A
> > 
> > ;; ANSWER SECTION:
> > imgw.pl.		3600	IN	A	195.187.86.13
> > imgw.pl.		3600	IN	A	195.187.86.14
> > 
> > ;; Query time: 55 msec
> > ;; SERVER: 212.160.142.68#53(212.160.142.68)
> > ;; MSG SIZE  rcvd: 71
> > 
> > So, we have to be lenient too, and a fix is in svn r2339 (not in 1.4.7
> > released today; but in the svn trunk).
> > 
> > Best regards,
> >    Wouter
> > -----BEGIN PGP SIGNATURE-----
> > Version: GnuPG v1.4.11 (GNU/Linux)
> > Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org/
> > 
> > iEYEARECAAYFAkzYQt0ACgkQkDLqNwOhpPgB0ACfZBPaTC1dLo+jFyk5QhNuwmFI
> > eosAoLMpTyIp6YJ7yt5HQ9rKWKo4IBUq
> > =lzNO
> > -----END PGP SIGNATURE-----
> > _______________________________________________
> > Unbound-users mailing list
> > Unbound-users at unbound.net
> > http://unbound.nlnetlabs.nl/mailman/listinfo/unbound-users
> 
> 
> _______________________________________________
> Unbound-users mailing list
> Unbound-users at unbound.net
> http://unbound.nlnetlabs.nl/mailman/listinfo/unbound-users