9:26 a.m.
Since few days now i have a name resolution problem. For example when i
entrer a new address in firefox it returns that the name can't be
resolved. Reloading the page and firefox display fine the page. I have
the same problem from thunderbird or cli using yum.
The network is up using network manager with ethernet and dhcp. The name
server is the box ip. This configuration is the same as other pc on the
same network and these pc don't have any problem so i think the problem
is specific to this f15 machine but i don't how to investigate.
Any help ?
Thanks
Eric
1:31 p.m.
>> yeah... I just can't be bothered to set up BIND.
That's what things like
>> Google Public DNS is for. :D
>
> No, the purpose of Google Public DNS is to give Google insight into
> every network query you make. Your filterbubble is heavily influenced by
> your history record in Google's DNS system if you have dodged the other
> ways of tracking. This sort of profiling goes further than syndicate
> cookies and trackers ever could -- and is a brilliant, if somewhat
> seductively evil, idea; so long as it is being used on someone other
> than me.
>
> http://dontbubble.us/
>
> Avoiding Google entirely has brought a great deal of standardization and
> rationality back to my organization -- that we didn't realize was
> beginning to get shaky until just recently. Such an insidious thing,
> filtered and tracked search.
Do you have any proof that Google's using queries to its Public DNS
service to profile anyone (in spite of its FAQ clarifying that it
isn't)?
Requests for proof on the internet from a person who does not work for
Google -- an internet argument deconstructionist in his opening play.
Nice. (Reminds me demands issued in Usenet of the Apollo landings...)
The filter bubble issue is very real. If you and I do a search on Google
for any given string, logged in to a Google account of any sort or not,
we will receive different results. This is a fact.
DNS query history would be the single most potent addition to Google's
profiling tags (as in naked profiling, on subjects who are not logged in
to a Google service or accepting tracking cookies or other devices). To
think that Google would *not* use their DNS query data is absurd. Let us
remember that this company has already found no problem in blowing off
German court subpoenas to account for and explain their violations of
European privacy laws while collecting Wifi and associated personal
network data while collecting local map data. Fibbing about their
privacy standards, which is nothing new for them anyway, is a minor
social foul by comparison.
From a technical perspective whether Google is actually using the data
actively right now to profile you or not is unprovable. That does not
mean it is not a major security risk now, and increasingly in the
future. Conversely, do you have a way of proving that they are not using
such data in such a way other than what they say? The evidence of result
filtering is highly evident. Are you going to simply trust their story
on it?
Security and power is all about capability, not intent. Intent can
change with the wind (and always eventually does). Consciously seeking
the creation of a position of leveragable power is never done for the
"common good" there being no such thing anyway. It does not matter if
the intent is to use DNS query histories right now or to use them in the
future -- the position is being formed for a very deliberate reason. And
even if not for a deliberate reason, if Google realizes later on down
the road the position they are in the game will suddenly change because
capacbility will force a shift in intent. This is strictly in line with
everything history has ever demonstrated.
There is simply no such thing as a free lunch and Google stands to gain
enormously over the long term by marking DNS query records, and
ultimately (if possible) by exerting de facto control over the DNS root.
Handing Google your DNS root is even less democratic a way of running
things than continuing along with the IANA circus -- at least there is
more than one entity with a stake to argue there (hence at least some
argument), with Google controlling your DNS queries you have eliminated
the chance that someone could be at odds with an abusive interest
present in the system.
If this needs further explaining to anyone on this list then no degree
of explaination will ever suffice.
-Iwao
1:58 p.m.
On 07/05/2011 07:40 AM, JB wrote:
Not only that !
She claimed to have smoked but not inhaled too ... :-)
Obviously you've never smoked either a pipe or a cigar. The only form
of tobacco you inhale is a cigarette.
Oh, wait, you probably weren't talking about tobacco, were you? Never mind!
6:24 a.m.
New subject: Name resolution - take this offline please
On Tue, 2011-07-05 at 23:32 -0400, Genes MailLists wrote:
On 07/05/2011 11:23 PM, 夜神 岩男 wrote:
>
> The footprint of a user by Google's way of doing things is quite a bit
> larger than cookies or IP tracking. They do not rely on any one set of
>
This conspiracy opinion stuff has nothing to do with fedora - please
take this discussion out of the mail lists.
I have no view on the topic other than it does not belong here.
Thank you ..
gene
I completely agree.
poc
6:32 a.m.
New subject: Name resolution - take this offline please
On 6 July 2011 15:24, Patrick O'Callaghan <pocallaghan(a)gmail.com> wrote:
On Tue, 2011-07-05 at 23:32 -0400, Genes MailLists wrote:
> On 07/05/2011 11:23 PM, 夜神 岩男 wrote:
>
> >
> > The footprint of a user by Google's way of doing things is quite a bit
> > larger than cookies or IP tracking. They do not rely on any one set of
>
> >
>
> This conspiracy opinion stuff has nothing to do with fedora - please
> take this discussion out of the mail lists.
>
> I have no view on the topic other than it does not belong here.
>
> Thank you ..
>
> gene
I completely agree.
poc
+1
--
Hiisi.
Registered Linux User #487982. Be counted at: http://counter.li.org/
--
Spandex is a privilege, not a right.
11:04 a.m.
On Tue, 05 Jul 2011 20:45:13 +0900
夜神 岩男 <supergiantpotato(a)yahoo.co.jp> wrote:
...
Your filterbubble is heavily influenced
by your history record in Google's DNS system if you have dodged the
other ways of tracking.
http://dontbubble.us/
...
ixquick ( https://www.ixquick.com/ ) is another privacy search option.
It even allows search results to be viewed via proxy if you choose to
do so.
1:52 a.m.
On Tue, 2011-07-05 at 22:55 -0400, Tom H wrote:
So, until there's an official complaint of some sort in this
regard,
you're just spreading FUD - unless you have a relevant URL to a valid
news report.
i.e. Do not query that something may be happening, until someone else
says so...
--
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686
Don't send private replies to my address, the mailbox is ignored. I
read messages from the public lists.
10:32 p.m.
New subject: Name resolution - take this offline please
On 07/05/2011 11:23 PM, 夜神 岩男 wrote:
This conspiracy opinion stuff has nothing to do with fedora - please
take this discussion out of the mail lists.
I have no view on the topic other than it does not belong here.
Thank you ..
gene
The footprint of a user by Google's way of doing things is quite a bit
larger than cookies or IP tracking. They do not rely on any one set of
9:55 p.m.
2011/7/5 夜神 岩男 <supergiantpotato(a)yahoo.co.jp>:
> >> yeah... I just can't be bothered to set up BIND. That's what things
like
> >> Google Public DNS is for. :D
> >
> > No, the purpose of Google Public DNS is to give Google insight into
> > every network query you make. Your filterbubble is heavily influenced by
> > your history record in Google's DNS system if you have dodged the other
> > ways of tracking. This sort of profiling goes further than syndicate
> > cookies and trackers ever could -- and is a brilliant, if somewhat
> > seductively evil, idea; so long as it is being used on someone other
> > than me.
> >
> > http://dontbubble.us/
> >
> > Avoiding Google entirely has brought a great deal of standardization and
> > rationality back to my organization -- that we didn't realize was
> > beginning to get shaky until just recently. Such an insidious thing,
> > filtered and tracked search.
>
> Do you have any proof that Google's using queries to its Public DNS
> service to profile anyone (in spite of its FAQ clarifying that it
> isn't)?
Requests for proof on the internet from a person who does not work for
Google -- an internet argument deconstructionist in his opening play.
Nice. (Reminds me demands issued in Usenet of the Apollo landings...)
Feel free to qualify my request whichever way you like...
Google's one of the most heavily scrutinized companies in the world.
Should it start to use its DNS service to collect information on its
users, its competitors and regulators will raise the issue very
quickly.
So, until there's an official complaint of some sort in this regard,
you're just spreading FUD - unless you have a relevant URL to a valid
news report.
3:28 p.m.
On 07/05/2011 11:31 AM, 夜神 岩男 wrote:
DNS query history would be the single most potent addition to
Google's
profiling tags (as in naked profiling, on subjects who are not logged in
to a Google service or accepting tracking cookies or other devices).
How do they keep track of people like me who have dynamic IP addresses?
And, for that matter, I'm house sitting right now over fifty miles
from home using a friend's connection. How can they tell that my DNS
queries are mine, not his? (Not that I use Google's DNS, but the
question's still valid in the general case.)
2:49 p.m.
On Wed, 2011-07-06 at 03:31 +0900, 夜神 岩男 wrote:
The filter bubble issue is very real. If you and I do a search on
Google for any given string, logged in to a Google account of any sort
or not, we will receive different results. This is a fact.
Something they can do perfectly easily with cookies. No need for DNS
munging.
poc
2:12 p.m.
Joe Zeff <joe <at> zeff.us> writes:
On 07/05/2011 07:40 AM, JB wrote:
> Not only that !
> She claimed to have smoked but not inhaled too ...
Obviously you've never smoked either a pipe or a cigar. The only form
of tobacco you inhale is a cigarette.
Oh, wait, you probably weren't talking about tobacco, were you? Never mind!
Yes, it was a play on both, "her" and "smoking" - things in life are
sometimes
so multithreaded, one could not imagine them, ever.
Peace.
JB
10:21 a.m.
Am 05.07.2011 17:17, schrieb John Aldrich:
On Tue July 5 2011, 夜神 岩男 wrote:
>
> No, the purpose of Google Public DNS is to give Google insight into
> every network query you make. Your filterbubble is heavily influenced by
> your history record in Google's DNS system if you have dodged the other
> ways of tracking. This sort of profiling goes further than syndicate
> cookies and trackers ever could -- and is a brilliant, if somewhat
> seductively evil, idea; so long as it is being used on someone other
> than me.
>
Ok... so maybe OpenDNS if you don't like Google. There's also a list here:
http://forums.techguy.org/networking/460800-free-public-dns-servers.html
why not simply "yum install bind-chroot"?
BIND works like a charme and needs no configuration for recursion
on the loopback-device
10:46 a.m.
On Tue July 5 2011, 夜神 岩男 wrote:
No, the purpose of Google Public DNS is to give Google insight into
every network query you make. Your filterbubble is heavily influenced by
your history record in Google's DNS system if you have dodged the other
ways of tracking. This sort of profiling goes further than syndicate
cookies and trackers ever could -- and is a brilliant, if somewhat
seductively evil, idea; so long as it is being used on someone other
than me.
http://dontbubble.us/
Avoiding Google entirely has brought a great deal of standardization and
rationality back to my organization -- that we didn't realize was
beginning to get shaky until just recently. Such an insidious thing,
filtered and tracked search.
Another list of free, public DNS servers is listed here:
http://www.joomlaspan.com/technology/blazing-fast-and-free-public-dns-
servers.php
9:40 a.m.
Tim <ignored_mailbox <at> yahoo.com.au> writes:
...
Want an example? There's the president who "did not have sex with that
woman." Well, he apparently did have some sexually intimate relations,
just not conjoined genitals. So the denial is correct, but incorrect.
...
Not only that !
She claimed to have smoked but not inhaled too ... :-)
JB
8:53 a.m.
On Tue, 2011-07-05 at 08:16 -0400, Tom H wrote:
Do you have any proof that Google's using queries to its Public
DNS
service to profile anyone (in spite of its FAQ clarifying that it
isn't)?
I'd certainly have my doubts. I tend to have little faith in the public
declarations of what corporations or governments intend, or are doing.
Often they just play with words, so their precise declaration is
technically correct, but what they've actually done is something else.
Want an example? There's the president who "did not have sex with that
woman." Well, he apparently did have some sexually intimate relations,
just not conjoined genitals. So the denial is correct, but incorrect.
Given that Google has had a stated aim of databasing everything, I'd be
surprised if at some time they hadn't databased people's use of their
DNS servers, or if they hadn't intended to. I wouldn't be surprised if
they weren't databasing it, somehow.
And it's not hard for Google to say we don't database /your/ queries to
the DNS servers, if what they database is just the queries, and not who.
But, since their search engine also databases queries of another kind,
it would be possible to merge the data.
Of course, there's people who just love stuffing databases with bogus
data, to poison their data with nonsense. I used to like doing that
with a friend's website, asking for pages with suggestive names, or
simply typing messages in the URL request, to see if he looked at his
site's 404 logs. I can't be bothered to do that sort of thing all the
time, though.
--
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686
Don't send private replies to my address, the mailbox is ignored. I
read messages from the public lists.
8:33 a.m.
Tim:
> I run my own DNS server, for a similar reason: Every ISP
I've tried
> has a crappy DNS server. Before I did that, I had to put some
> domain's IP into my hosts file, because their DNS server usually gave
> no answer.
John Aldrich:
yeah... I just can't be bothered to set up BIND. That's what
things
like Google Public DNS is for. :D
Well, mine does something /it/ can't do: Resolve the device addresses
inside my LAN.
--
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686
Don't send private replies to my address, the mailbox is ignored. I
read messages from the public lists.
7:16 a.m.
2011/7/5 夜神 岩男 <supergiantpotato(a)yahoo.co.jp>:
On Tue, 2011-07-05 at 06:34 -0400, John Aldrich wrote:
> On Tue July 5 2011, Tim wrote:
> > On Mon, 2011-07-04 at 12:52 -0400, John Aldrich wrote:
> > > might I suggest trying Google Public DNS servers? 8.8.8.8 and 8.8.8.4
> > > are the IP addresses. My ISP apparently runs some sort of filtering
> > > and occasionally I have problems with their DNS, so I switched to
> > > Google and that pretty much resolves the issue for me. :D
> >
> > I run my own DNS server, for a similar reason: Every ISP I've tried has
> > a crappy DNS server. Before I did that, I had to put some domain's IP
> > into my hosts file, because their DNS server usually gave no answer.
> >
> yeah... I just can't be bothered to set up BIND. That's what things like
> Google Public DNS is for. :D
No, the purpose of Google Public DNS is to give Google insight into
every network query you make. Your filterbubble is heavily influenced by
your history record in Google's DNS system if you have dodged the other
ways of tracking. This sort of profiling goes further than syndicate
cookies and trackers ever could -- and is a brilliant, if somewhat
seductively evil, idea; so long as it is being used on someone other
than me.
http://dontbubble.us/
Avoiding Google entirely has brought a great deal of standardization and
rationality back to my organization -- that we didn't realize was
beginning to get shaky until just recently. Such an insidious thing,
filtered and tracked search.
Do you have any proof that Google's using queries to its Public DNS
service to profile anyone (in spite of its FAQ clarifying that it
isn't)?
6:45 a.m.
On Tue, 2011-07-05 at 06:34 -0400, John Aldrich wrote:
On Tue July 5 2011, Tim wrote:
> On Mon, 2011-07-04 at 12:52 -0400, John Aldrich wrote:
> > might I suggest trying Google Public DNS servers? 8.8.8.8 and 8.8.8.4
> > are the IP addresses. My ISP apparently runs some sort of filtering
> > and occasionally I have problems with their DNS, so I switched to
> > Google and that pretty much resolves the issue for me. :D
>
> I run my own DNS server, for a similar reason: Every ISP I've tried has
> a crappy DNS server. Before I did that, I had to put some domain's IP
> into my hosts file, because their DNS server usually gave no answer.
>
yeah... I just can't be bothered to set up BIND. That's what things like
Google Public DNS is for. :D
No, the purpose of Google Public DNS is to give Google insight into
every network query you make. Your filterbubble is heavily influenced by
your history record in Google's DNS system if you have dodged the other
ways of tracking. This sort of profiling goes further than syndicate
cookies and trackers ever could -- and is a brilliant, if somewhat
seductively evil, idea; so long as it is being used on someone other
than me.
http://dontbubble.us/
Avoiding Google entirely has brought a great deal of standardization and
rationality back to my organization -- that we didn't realize was
beginning to get shaky until just recently. Such an insidious thing,
filtered and tracked search.
-Iwao
6:05 a.m.
I finish my mail : i just try 3 times to send the mail because
thunderbid failed to send it due to configuration problem on the server
smtp.googlemail.com I open a CLI and run ping smtp.googlemail.com the
server answer fine and i achieve to send my email.
??
Eric
Perhaps a bogus DNS cache near you recently expired? I've had a bunch of
bad DNS data sit latent for a bit and trip me up for a bit (particularly
on Windows systems where flushing it was an unclear process to me) --
but it has been a long time since anything like that was common (or am I
just lucky?).
-Iwao
5:34 a.m.
On Tue July 5 2011, Tim wrote:
On Mon, 2011-07-04 at 12:52 -0400, John Aldrich wrote:
> might I suggest trying Google Public DNS servers? 8.8.8.8 and 8.8.8.4
> are the IP addresses. My ISP apparently runs some sort of filtering
> and occasionally I have problems with their DNS, so I switched to
> Google and that pretty much resolves the issue for me. :D
I run my own DNS server, for a similar reason: Every ISP I've tried has
a crappy DNS server. Before I did that, I had to put some domain's IP
into my hosts file, because their DNS server usually gave no answer.
yeah... I just can't be bothered to set up BIND. That's what things like
Google Public DNS is for. :D
7:19 p.m.
On Mon, 2011-07-04 at 18:56 +0200, Eric Tanguy wrote:
> nameserver 1.2 3.4
>
> then try:
>
> $ dig @1.2.3.4 mit.edu
>
> Do the same on your other machines. Are they all using the same
> nameserver?
>
> poc
>
Yes i mean my router's ip.
$ cat /etc/resolv.conf
# Generated by NetworkManager
domain home
search home
nameserver 192.168.1.1
$ dig @1.2.3.4 mit.edu
Seriously? You're supposed to use the value for nameserver in
the /etc/resolv.conf file, not literally 1.2.3.4. Wasn't that obvious
from the example?
poc
12:02 p.m.
On Mon, Jul 04, 2011 at 06:59:34PM +0200, Eric Tanguy wrote:
Le 04/07/2011 18:56, Eric Tanguy a écrit :
> Le 04/07/2011 16:29, Patrick O'Callaghan a écrit :
>> On Mon, 2011-07-04 at 16:26 +0200, Eric Tanguy wrote:
>>> Since few days now i have a name resolution problem. For example when i
>>> entrer a new address in firefox it returns that the name can't be
>>> resolved. Reloading the page and firefox display fine the page. I have
>>> the same problem from thunderbird or cli using yum.
>>> The network is up using network manager with ethernet and dhcp. The
>>> name
>>> server is the box ip. This configuration is the same as other pc on the
>>> same network and these pc don't have any problem so i think the problem
>>> is specific to this f15 machine but i don't how to investigate.
>>> Any help ?
>> What do you mean by "the name server is the box ip"? Do you mean your
>> router's IP or the computer's own IP? If the latter, then unless
you're
>> running bind on your local machine that's your problem.
>>
>> Check the real situation thus:
>>
>> $ cat /etc/resolv.conf
>>
>> That will tell you what your name servers are. Make sure they exist and
>> are actually resolving names, e.g. if /etc/resolv.com contains:
>>
>> nameserver 1.2 3.4
>>
>> then try:
>>
>> $ dig @1.2.3.4 mit.edu
>>
>> Do the same on your other machines. Are they all using the same
>> nameserver?
>>
>> poc
>>
> Yes i mean my router's ip.
>
> $ cat /etc/resolv.conf
> # Generated by NetworkManager
> domain home
> search home
> nameserver 192.168.1.1
>
> $ dig @1.2.3.4 mit.edu
>
> ; <<>> DiG 9.8.0-P2-RedHat-9.8.0-5.P2.fc15 <<>> @1.2.3.4
mit.edu
> ; (1 server found)
> ;; global options: +cmd
> ;; connection timed out; no servers could be reached
>
> Difficult to compare with other computers because they are under winxp
> ...
>
> it seems that the problem is mainly after return from hibernate but
> i'm not sure ...
>
> An example : i
>
> Eric
>
I finish my mail : i just try 3 times to send the mail because
thunderbid failed to send it due to configuration problem on the server
smtp.googlemail.com I open a CLI and run ping smtp.googlemail.com the
server answer fine and i achieve to send my email.
??
Eric
sounds like what used to happen at my home, back when we had dialup.
We used ppp with autodial set, so that when an outgoing packet appeared
at the linux box acting as firewall/router, it would start up ppp. since
that takes a short while, we often had to retry the first access to the
internet.
is there something about your net access that would be expected to
act like that?
--
---- Fred Smith -- fredex(a)fcshome.stoneham.ma.us -----------------------------
The Lord is like a strong tower.
Those who do what is right can run to him for safety.
--------------------------- Proverbs 18:10 (niv) -----------------------------
11:59 a.m.
Le 04/07/2011 18:56, Eric Tanguy a écrit :
Le 04/07/2011 16:29, Patrick O'Callaghan a écrit :
> On Mon, 2011-07-04 at 16:26 +0200, Eric Tanguy wrote:
>> Since few days now i have a name resolution problem. For example when i
>> entrer a new address in firefox it returns that the name can't be
>> resolved. Reloading the page and firefox display fine the page. I have
>> the same problem from thunderbird or cli using yum.
>> The network is up using network manager with ethernet and dhcp. The
>> name
>> server is the box ip. This configuration is the same as other pc on the
>> same network and these pc don't have any problem so i think the problem
>> is specific to this f15 machine but i don't how to investigate.
>> Any help ?
> What do you mean by "the name server is the box ip"? Do you mean your
> router's IP or the computer's own IP? If the latter, then unless you're
> running bind on your local machine that's your problem.
>
> Check the real situation thus:
>
> $ cat /etc/resolv.conf
>
> That will tell you what your name servers are. Make sure they exist and
> are actually resolving names, e.g. if /etc/resolv.com contains:
>
> nameserver 1.2 3.4
>
> then try:
>
> $ dig @1.2.3.4 mit.edu
>
> Do the same on your other machines. Are they all using the same
> nameserver?
>
> poc
>
Yes i mean my router's ip.
$ cat /etc/resolv.conf
# Generated by NetworkManager
domain home
search home
nameserver 192.168.1.1
$ dig @1.2.3.4 mit.edu
; <<>> DiG 9.8.0-P2-RedHat-9.8.0-5.P2.fc15 <<>> @1.2.3.4 mit.edu
; (1 server found)
;; global options: +cmd
;; connection timed out; no servers could be reached
Difficult to compare with other computers because they are under winxp
...
it seems that the problem is mainly after return from hibernate but
i'm not sure ...
An example : i
Eric
I finish my mail : i just try 3 times to send the mail because
thunderbid failed to send it due to configuration problem on the server
smtp.googlemail.com I open a CLI and run ping smtp.googlemail.com the
server answer fine and i achieve to send my email.
??
Eric
11:56 a.m.
Le 04/07/2011 16:29, Patrick O'Callaghan a écrit :
On Mon, 2011-07-04 at 16:26 +0200, Eric Tanguy wrote:
> Since few days now i have a name resolution problem. For example when i
> entrer a new address in firefox it returns that the name can't be
> resolved. Reloading the page and firefox display fine the page. I have
> the same problem from thunderbird or cli using yum.
> The network is up using network manager with ethernet and dhcp. The name
> server is the box ip. This configuration is the same as other pc on the
> same network and these pc don't have any problem so i think the problem
> is specific to this f15 machine but i don't how to investigate.
> Any help ?
What do you mean by "the name server is the box ip"? Do you mean your
router's IP or the computer's own IP? If the latter, then unless you're
running bind on your local machine that's your problem.
Check the real situation thus:
$ cat /etc/resolv.conf
That will tell you what your name servers are. Make sure they exist and
are actually resolving names, e.g. if /etc/resolv.com contains:
nameserver 1.2 3.4
then try:
$ dig @1.2.3.4 mit.edu
Do the same on your other machines. Are they all using the same
nameserver?
poc
Yes i mean my router's ip.
$ cat /etc/resolv.conf
# Generated by NetworkManager
domain home
search home
nameserver 192.168.1.1
$ dig @1.2.3.4 mit.edu
; <<>> DiG 9.8.0-P2-RedHat-9.8.0-5.P2.fc15 <<>> @1.2.3.4 mit.edu
; (1 server found)
;; global options: +cmd
;; connection timed out; no servers could be reached
Difficult to compare with other computers because they are under winxp ...
it seems that the problem is mainly after return from hibernate but i'm
not sure ...
An example : i
Eric
11:52 a.m.
On Mon July 4 2011, Eric Tanguy wrote:
Since few days now i have a name resolution problem. For example when
i
entrer a new address in firefox it returns that the name can't be
resolved. Reloading the page and firefox display fine the page. I have
the same problem from thunderbird or cli using yum.
The network is up using network manager with ethernet and dhcp. The name
server is the box ip. This configuration is the same as other pc on the
same network and these pc don't have any problem so i think the problem
is specific to this f15 machine but i don't how to investigate.
Any help ?
Thanks
Eric
Eric:
In addition to the suggestions given by anyone else, might I suggest trying
Google Public DNS servers? 8.8.8.8 and 8.8.8.4 are the IP addresses. My ISP
apparently runs some sort of filtering and occasionally I have problems with
their DNS, so I switched to Google and that pretty much resolves the issue
for me. :D
10:22 a.m.
Regards, Eric
Are you using a proxy for Internet connection? Because, this kind of
issues I've seen on it.
1- I think that you have to check your /etc/resolv.conf which are the
entries that you have there.
2- The second advice that I give you is to analyze your network
connection (rates, bandwidth, etc). You can use
tcpdump
- sar(included on the systat package
http://pagesperso-orange.fr/sebastien.godard/) for all kind of statistics.
For the network stats, you can use sar -n DEV | more
- netstat: only focused on the network statistics
a example: netstat --statistics --raw
- traceroute: this is very useful for tracking down the cause
of disappearing packages
a example: traceroute host_name
3- You have to check the /etc/nsswitch.conf, which is the main
responsible of the name resolution in Solaris and Linux. You will
check too the nscd daemon
I hope that's can help you
On 07/04/2011 09:56 AM, Eric Tanguy wrote:
Since few days now i have a name resolution problem. For example when
i
entrer a new address in firefox it returns that the name can't be
resolved. Reloading the page and firefox display fine the page. I have
the same problem from thunderbird or cli using yum.
The network is up using network manager with ethernet and dhcp. The name
server is the box ip. This configuration is the same as other pc on the
same network and these pc don't have any problem so i think the problem
is specific to this f15 machine but i don't how to investigate.
Any help ?
Thanks
Eric
--
Marcos Luís Ortíz Valmaseda
Software Engineer (UCI)
Linux User # 418229
http://marcosluis2186.posterous.com
http://twitter.com/marcosluis2186
9:29 a.m.
On Mon, 2011-07-04 at 16:26 +0200, Eric Tanguy wrote:
Since few days now i have a name resolution problem. For example when
i
entrer a new address in firefox it returns that the name can't be
resolved. Reloading the page and firefox display fine the page. I have
the same problem from thunderbird or cli using yum.
The network is up using network manager with ethernet and dhcp. The name
server is the box ip. This configuration is the same as other pc on the
same network and these pc don't have any problem so i think the problem
is specific to this f15 machine but i don't how to investigate.
Any help ?
What do you mean by "the name server is the box ip"? Do you mean your
router's IP or the computer's own IP? If the latter, then unless you're
running bind on your local machine that's your problem.
Check the real situation thus:
$ cat /etc/resolv.conf
That will tell you what your name servers are. Make sure they exist and
are actually resolving names, e.g. if /etc/resolv.com contains:
nameserver 1.2 3.4
then try:
$ dig @1.2.3.4 mit.edu
Do the same on your other machines. Are they all using the same
nameserver?
poc
10:23 p.m.
On Tue, 2011-07-05 at 13:28 -0700, Joe Zeff wrote:
On 07/05/2011 11:31 AM, 夜神 岩男 wrote:
> DNS query history would be the single most potent addition to Google's
> profiling tags (as in naked profiling, on subjects who are not logged in
> to a Google service or accepting tracking cookies or other devices).
How do they keep track of people like me who have dynamic IP addresses?
And, for that matter, I'm house sitting right now over fifty miles
from home using a friend's connection. How can they tell that my DNS
queries are mine, not his? (Not that I use Google's DNS, but the
question's still valid in the general case.)
The footprint of a user by Google's way of doing things is quite a bit
larger than cookies or IP tracking. They do not rely on any one set of
criteria, but instead use everything available to build a profile. You
browser type and version, what OS you are using and its flavor, the time
of day you are searching, the type of search if you use bang syntaxes,
recent affiliate add display history to the IP you are on (this is
probably one of the most powerful tools), what type of thing you are
searching for at the moment and, critically, what you are clicking on of
the results, etc. These things alone provide an extremely close picture
of the user(s), without resorting to cookies or relying too much on any
one particular type of data. (Look through your own web server logs and
you can discern patterns of use, even if you delete the IPs completely,
even without going into a very deep analysis.)
DNS query history can tighten the picture considerably and give insight
into things search cannot -- which is my point here. Google is a
for-profit company (as it should be), not a social service funded by
dreamy eyed populists who don't understand economics. I owned a good bit
of stock in them until recently, so they *better* be all about making
money and not about being nice. The expense involved in providing fast,
free DNS service everywhere at once is a good investment only if the
information harvested by the effort is used in a revenue bearing way or
unless the DNS service is to be suddenly changed to a for-pay service.
Considering the alternatives, for-pay DNS through anybody is
preposterous, and so that leaves data collection.
Considering what Google's non-cookie profiling methods are, DNS
collection makes perfect sense as a market to corner and to get insight
into everything that is not search based. This is a great strategy for
Google -- it is just not something I am going to expose myself to as a
user. DNS data collection puts Google in the best possible place for
data collection because they can harvest URL call that occurs outside of
search. This type of information, combined with their proven business
model, is too powerful to ignore, no matter what their FAQ says.
-Iwao
9:01 p.m.
夜神 岩男:
> DNS query history would be the single most potent addition to
Google's
> profiling tags (as in naked profiling, on subjects who are not logged in
> to a Google service or accepting tracking cookies or other devices).
Joe Zeff:
How do they keep track of people like me who have dynamic IP
addresses?
And, for that matter, I'm house sitting right now over fifty miles
from home using a friend's connection. How can they tell that my DNS
queries are mine, not his? (Not that I use Google's DNS, but the
question's still valid in the general case.)
How hard is to imagine how? Obviously, there's very likely to be some
holes in the data. But once you login somewhere that's associated with
Google, you're known. From then on, you're databaseable.
Before that, using the world's largest database, that's already been
databasing you, it's a matter of correlating the trend of what you're
doing with someone who's been databased for doing the same thing, to
establish a probability that it's the same person.
While that's fraught with errors, it's the evidential methodology in
solving a crime (fit the likely pieces together until you find something
very corroborating that links to something provable, then you follow
those links back in reverse).
But I think most people will identify themselves, somewhere, by logging
into at least one service, and that service will be one of Google's, or
something they've got a finger in. I think this happening will be far
more likely than people think might happen.
And right now, a search engine probably knows your IP, since they could
have indexed these emails (with headers), and you're probably not doing
anything to hide it.
--
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686
Don't send private replies to my address, the mailbox is ignored. I
read messages from the public lists.
10:17 a.m.
On Tue July 5 2011, 夜神 岩男 wrote:
No, the purpose of Google Public DNS is to give Google insight into
every network query you make. Your filterbubble is heavily influenced by
your history record in Google's DNS system if you have dodged the other
ways of tracking. This sort of profiling goes further than syndicate
cookies and trackers ever could -- and is a brilliant, if somewhat
seductively evil, idea; so long as it is being used on someone other
than me.
Ok... so maybe OpenDNS if you don't like Google. There's also a list here:
http://forums.techguy.org/networking/460800-free-public-dns-servers.html
8:04 a.m.
On 07/05/2011 08:16 AM, Tom H wrote:
>
> http://dontbubble.us/
>
> Avoiding Google entirely has brought a great deal of standardization and
> rationality back to my organization -- that we didn't realize was
> beginning to get shaky until just recently. Such an insidious thing,
> filtered and tracked search.
Do you have any proof that Google's using queries to its Public DNS
service to profile anyone (in spite of its FAQ clarifying that it
isn't)?
For that matter - what proof is there that the bubble thing which noone
ever heard of is trustworthy at all - or of any use as a search engine.
It looks pretty amateurish to me ...
11:03 p.m.
On Mon, 2011-07-04 at 12:52 -0400, John Aldrich wrote:
might I suggest trying Google Public DNS servers? 8.8.8.8 and
8.8.8.4
are the IP addresses. My ISP apparently runs some sort of filtering
and occasionally I have problems with their DNS, so I switched to
Google and that pretty much resolves the issue for me. :D
I run my own DNS server, for a similar reason: Every ISP I've tried has
a crappy DNS server. Before I did that, I had to put some domain's IP
into my hosts file, because their DNS server usually gave no answer.
--
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686
Don't send private replies to my address, the mailbox is ignored. I
read messages from the public lists.
4701
days inactive
4703
days old
32 comments
14 participants
participants (14)
-
Fred Smith -
Genes MailLists -
Hiisi -
JB -
Joe Zeff -
John Aldrich -
Marcos Ortiz Valmaseda -
Patrick O'Callaghan -
Reindl Harald -
stan -
Tanguy Eric -
Tim -
Tom H -
夜神 岩男