No DNS lookup for MX, A, AAAA or A6?

Good day folks,
Does SJS MS 6.2 support querying DNS for MX,A,AAAA or A6 records for the domain specified in the SMTP mail from:<[email protected]>? Exim and sendmail support this measure as a means of reducing spam but I can't find the equivalent for SJS MS 6.2. mailfromdnsverify on the tcp_local channel only sets up verfication of existence of the domain itself.
Cheers

Hi Shane,
Let me take this piece by piece:
"Really? That's not a feature I've ever actually heard that either of these products have. I can't imagine how it might impact performance. Badly, I strongly suspect."
The following is an extract from the central mail servers' SMTP dialogue with my MS 6.2:
*** START MESSAGE ***
This report relates to a message you sent with the following header fields:
Message-id: <000001c73417$3ddaf380$0100007f@localhost>
Date: Tue, 09 Jan 2007 12:55:29 -0500
From: Ethan Edwards <[email protected]>
To: [email protected]
Subject: What IS 0EM Software And Why D0 You Care?
Your message cannot be delivered to the following recipients:
Recipient address: [email protected]
Original address: [email protected]
Reason: Remote SMTP server has rejected address
Diagnostic code: smtp;553 5.1.8 <[email protected]>... Domain of sender address [email protected] does not exist
Remote system: dns;centraldnsserver.mycentraldomain.com(TCP|mymailserveripaddress|62084|centralmailserveripaddress|25) (centralmailserver.um.edu.mt ESMTP CSCMAIL/External server ready)
*** END MESSAGE ***
"No, if we had to query every server that mail appears to come from, to check the validity of that mail address, we'd never be able to process mail in any quantity. I strongly doubt that any other product does as you suggest, either"
This kind of checking is done on every message by mail servers running sendmail that host about 500 users.
"How about SpamAssassin?
RBL?
Greylisting?"
I think that these techniques all get past the SMTP stage - I want better handling at the SMTP stage.
"If your central systems accept mail addressed to fake users, then that's a configuration error on their part, assuming that they can know all the real users."
What I intended is that the mail servers reject external, incoming mail from source addresses that include domains that only have SOA records.
Finally, regarding the blah123.com example: this domain has no SOA record so mailfromdnsverify works in rejecting the incoming e-mail at the SMTP stage. But try these:
pipex.co.uk
fstngt.org
lists.midterme.com
com.br
net.my
Thanks for keeping up this discussion.
Cheers,
Etiennen

Similar Messages

Constant dns lookups for non-existent addresses

Hi. I'm connected to a large network and I've noticed
that there are constant dns lookups for addresses that
do not exist.
When i run tcpdump, almost every second
I see a few requests to the dns server from my IP. And all
of them get the response NXDOMAIN.
Is there a reason this should happen or is there something
not working properly on my computer?
Thanks
Last edited by m00nblade (2010-01-25 21:42:23)

It all depends on your setup.
If you use only local mail domains, just make sure you do not have a catchall address (luser_relay) and messages to unknown users will not be accepted by Postfix in the first place.
If you use virtual mail domain, you will need to change your setup as Apple's default setup will always accept mail for unknown users and then bounce it back to sender. See here for a how to: Making Virtual Mail Users in OS X 10.4/10.5 Server
HTH,
Alex

Push Notifications: DNS Lookup for pubsub.localhost failed

Hello,
after a server crash and restore i cant manage to get push notifications back working. When adding a caldav account to ical, it doesnt offer "Push" in "Refresh Calendars" option anymore. Everything in the log files seems to look fine but this:
6/4/12 11:45:26.024 AM jabberd_notification/s2s: dns lookup for pubsub.localhost failed
Can someone help me out on this please?

Some more info. I think its just the last line which fails. Everything else looks like it works…
DNS seems fine as well.
6/4/12 5:32:37.179 PM jabberd_notification/c2s: [8] [::ffff:127.0.0.1, port=50810] connect
6/4/12 5:32:37.217 PM jabberd_notification/c2s: Authentication succeeded, mech: DIGEST-MD5 client IP: ::ffff:127.0.0.1 client port: 50810 username: com.apple.notificationuser
6/4/12 5:32:37.219 PM jabberd_notification/c2s: [8] SASL authentication succeeded: mechanism=DIGEST-MD5; authzid=com.apple.notificationuser@localhost, TLS negotiated
6/4/12 5:32:37.223 PM jabberd_notification/c2s: [8] bound: jid=com.apple.notificationuser@localhost/icalserver.3c311a83d7b04f1a977481cfea6 c8d9e
6/4/12 5:32:37.224 PM jabberd_notification/sm: session started: jid=com.apple.notificationuser@localhost/icalserver.3c311a83d7b04f1a977481cfea6 c8d9e
6/4/12 5:32:37.233 PM jabberd_notification/s2s: dns lookup for pubsub.localhost failed

GWIA doing DNS lookup for local address

Hello,
I am running GW8.0.2 on Netware 6.5sp8. I have a server that our recreation department uses to send out confirmation emails when a customer signs up for a class. The recreation server and the GWIA are on the same subnet.
Here's the problem: When the Rec server sends out the first email confirmation, it gets sent out successfully. Subsequent emails after that fail. After about twenty minutes the next email will go out OK again but subsequent emails will fail.
The verbose logs on the GWIA don't tell me much but the diagnostic logs show what looks like a reverse DNS lookup happening at the GWIA for my local IP address of 10.0.0.3 (the Rec server). This reverse DNS lookup fails (probably a timeout) and subsequent emails from this local Rec server get dropped by the GWIA without the DNS lookup.
DNS is being done by DNS proxy on Bordermanager 9.2. I've bypassed the Bordermanager DNS and the same thing happens. I've made entries for the local Rec server into a route.cfg file but the GWIA seems to want to ignore these entries and keeps doing the DNS lookup.
The wierdest part of the puzzle is that if I restart the proxy on the Bordermanager the next email will go out with, of course, subsequent emails failing. I've looked at the proxy dns cache and can't even find an entry for my Rec server.
Attached are the entries from the Diagnostic logs of the GWIA. Novell tech support has assured me that the GWIA and the BM are working fine. I am also having this problem with a scanner that scans then emails but all other email and Bordermanager are functioning fine. This server and scanner were not having this problem before upgrading to GW8.0.2.
I don't understand why GWIA is doing DNS lookups for a local address and I don't know what I can do to stop it. Any help would be greatly appreciated.
This is a successful transfer right after restarting the proxy: 10.0.0.3 is the Rec server, 10.0.0.130 is the GWIA and 10.0.0.1 is the Bordermanager.
16:04:13 D15 NgwResQuery(3.0.0.10.in-addr.arpa, 1, 12)
16:04:13 D15 Querying server (# 1) address = 10.0.0.1
16:04:13 D15 HEADER:
16:04:13 D15 opcode = QUERY, id = 17615, rcode = SERVFAIL, flags: qr aa rd
16:04:13 D15 query = 1, answer = 0, authority = 0, additional = 0
16:04:13 D15
16:04:13 D15 QUESTIONS:
16:04:13 D15 3.0.0.10.in-addr.arpa, type = PTR, class = IN
16:04:13 D15
16:04:13 D15 rcode = 2, ancount=0
16:04:13 D15 NgwResQuery failed
16:04:13 D15 DMN: MSG 2000909 Accepted connection: [10.0.0.3] ()
16:04:13 D15 Successful login with client/server access: 10.0.0.130:1677
16:04:13 D15 DMN: MSG 2000909 Receiving file: ECMAIL/SYS:\PROGRAMS\GRPWISE\WPGATE\GWIA\3RD\receive\df30 fad4.221
16:04:13 D15 DMN: MSG 2000909 SMTP session ended: [10.0.0.3] ()
This is an unsuccessful transfer:
16:06:08 D04 timeout
16:06:08 D04 NgwResQuery: send error
16:06:08 D04 NgwResQuery failed
16:06:08 D04 DMN: MSG 2000933 Accepted connection: [10.0.0.3] ()
16:06:08 D04 DMN: MSG 2000933 SMTP session ended: [10.0.0.3] ()
Then the successful email comes back into the system:
16:06:26 AA8 MSG 2000909 Processing inbound message: ECMAIL/SYS:\PROGRAMS\GRPWISE\WPGATE\GWIA\receive\DF30FAD4 .221
16:06:26 AA8 MSG 2000909 Sender: [email protected]
16:06:26 AA8 MSG 2000909 Recipient: [email protected]
16:06:26 AA8 MSG 2000909 Queuing to MTA
16:06:26 AA8 MSG 2000909 File: ECMAIL/SYS:\PROGRAMS\GRPWISE\WPGATE\GWIA\wpcsin\4\4daf048 2.8m1 Message Id: (4DAF66F2.B67:244:35687) Size: 163.3 Kb

Thanks Massimo. I could have swore I already did that but when I did it again just to make sure it solved the problem. Appreciate the help. Have a good one.
Originally Posted by mrosen
On 02.05.2011 21:06, avanrav wrote:
>
> Hello,
>
> I am running GW8.0.2 on Netware 6.5sp8. I have a server that our
> recreation department uses to send out confirmation emails when a
> customer signs up for a class. The recreation server and the GWIA are on
> the same subnet.
>
> Here's the problem: When the Rec server sends out the first email
> confirmation, it gets sent out successfully. Subsequent emails after
> that fail. After about twenty minutes the next email will go out OK
> again but subsequent emails will fail.
>
> The verbose logs on the GWIA don't tell me much but the diagnostic logs
> show what looks like a reverse DNS lookup happening at the GWIA for my
> local IP address of 10.0.0.3 (the Rec server). This reverse DNS lookup
> fails (probably a timeout) and subsequent emails from this local Rec
> server get dropped by the GWIA without the DNS lookup.
>
> DNS is being done by DNS proxy on Bordermanager 9.2. I've bypassed the
> Bordermanager DNS and the same thing happens. I've made entries for the
> local Rec server into a route.cfg file but the GWIA seems to want to
> ignore these entries and keeps doing the DNS lookup.
>
> The wierdest part of the puzzle is that if I restart the proxy on the
> Bordermanager the next email will go out with, of course, subsequent
> emails failing. I've looked at the proxy dns cache and can't even find
> an entry for my Rec server.
The reverse DNS done by GWIA is normal, and can't be stopped or tricked.
That it fails in such odd ways must be a bug with the reverse DNS proxy
of Bordermanager though. Apparently on the second lookups, it doesn't
answer in a timely manner (the type of answer is irrelevant, just it
*has* to answer). Use a different, "real" DNS server for your GWIA.
CU,
Massimo Rosen
Novell Product Support Forum Sysop
No emails please!
Untitled Document

How do I setup 'local' dns lookups for hostnames using DHCP server on WRT610N?

Said router with latest firmware has DHCP reservation setup with reserved clients names mapped to specific MAC addresses/IP addresses. I'd like my clients (a mixture of Mac and W7) to access mapped devices via hostnames and not IP addresses. I have static IP address provided by my ISP configured on the router with it's corresponding ISP DNS server addresses. FYI, the DHCP server DNS settings on the router are masked out (can't be changed) on the router and looks to be inherited from the Internet setup (in which I've entered the DNS IP addresses of my ISP's DNS).
On my clients, nslookup (or equivalent) clearly specify that the DHCP is pumping out the DNS servers from the ISP. Why would it not look for the reserved client hostnames first?
I know I can setup local /etc/hosts (or equivalents) or override the DNS and point to the router's IP address on the clients but I'd rather not - that's a lot of maintenance. I can also setup an internal DNS server, but again, overkill.
Is this feature available on this router or am I doing something wrong? An aging integrated Westell 327W DSL model/router used to be able to provide internal network DNS - but this more modern capable WRT610N router does not? Thoughts? Thanks.

Well its not possible to access the storage driver using the host name.. You can access the storage driver using only the IP address. As the Linksys Router doesn't work on a Host Name.

DNS Lookup for DHCP-assigned devices

Hi!
I expected that my router would be able to act as a DNS server to allow machines on my network to look up the IP addresses of other machines on my network, given that every machine on my network gets its IP address from the router via DHCP - but the router doesn't appear to be capable of that.
The router in question is a Linksys E2000 device, and it has the latest firmware revision installed (1.0.03 build 2Sep 7, 2010).
So, the questions I have are:
1. Can this router act as a DNS server?
2. If the answer to (1) is "yes", then how should it be configured?
Mike

Name resolution in LANs in Windows happens usually based on broadcasts in a workgroup or homegroup and that's generally the best way to do it. Some routers add DHCP entries to the DNS proxy but you can read about a lot of issues with that either, in particular if the DNS proxy information gets incorrect somehow and the DNS resolution is configured with precedence to LAN name resolution...
I would suggest you make sure your LAN name resolution is working properly.

Externally Hosted DNS - How do I set up my 2003 DNS server for sub domain to point to internal IP address??

I have a domain name(domain.com) DNS hosted at my ISP. I also have 3 sub domains DNS hosted at the same ISP pointing to various external ip addresses (mail.domain.com, vpn.domain.com and ts.domain.com). We want to set up sales.domain.com to point to an
internal 10. IP address. We have AD integrated DNS servers for our 2003 AD domain. The AD domain name is totally different than the hosted domain name in question. I currently edit the host file for a couple of PC's but this isnt practical company wide so
I want to add entries on our internal AD DNS servers to resolve the locally hosted site. If i recall, someone once told me that you cannot just put an A record for one sub domain, I would have to have entries on my 2003 DNS server to resolve anything related
to the domain.com name. Is this accurate? If so, what is the proper way to configure my 2003 AD DNS server to resolve anything domain.com related for my internal users while still allowing my ISP to do the DNS lookup for the internet.

On my 2003 AD integrated DNS server...i rightclick forward lookup zone and choose...new zone..primary zone (store zone in AD checkbox checked)..i chose to all DNS servers in the AD domain for replication...zone name sales.domain.com....allow secure updates
option....then i added an A record in that zone...sales.domain.com..pointed that towards my internal 10. IP address...is this correct? It seems to be working correctly for the sales.domain.com DNS record...and i tested the other sub domains...and those look
like they are going to my ISP for DNS resolution...
Is this the correct procedure? I did this on a test AD domain and not my production...i want to make sure i dont break everything under the domain.com by incorrectly adding 1 sub domain..

Finding LDAP server names by DNS lookup.

Hi,
I'm very new with JNDI and DNS
We are hardcoding the ldap server name in our configuration to connect to the Active directory, but the requirement is to know the ldap server name dynmaically by querying the DNS server.
The input given to us are below.
Dns domain : indbank.is.
SRV RRecord : ldap.tcp.
Query dns : ldap.tcp.indbank.is.
The domain controller should be found by a DNS lookup for the domain, then a DNS for Domain controllers that advertise the service, then try to see if the domain controllers areanswering, and if so choose the one with the fastest answer time (to avoid choosing a domain controller over WAN).
Kindly help me.I am beginner and some code sample and tip will be welcome. :)
Thanks in advance.
Hiubert

Thanks a lot to All.
My code is as follows...
import javax.naming.*;
import javax.naming.directory.*;
import java.util.*;
public class dns1
     public static void main(String[] args)     {
          try {
               Hashtable env = new Hashtable();
               env.put("java.naming.factory.initial","com.sun.jndi.dns.DnsContextFactory");
               env.put("java.naming.provider.url", "dns://indbank.is");
               DirContext ctx = new InitialDirContext(env);
               System.out.println("Intial context created...");
               Attributes attrs = ctx.getAttributes("_ldap._tcp.indbank.is",new String[] {"SRV"});
               System.out.println("Attributes are been retrieved...");
               for (NamingEnumeration ae = attrs.getAll();ae.hasMoreElements();)
                    Attribute attr = (Attribute)ae.next();
                    String attrId = attr.getID();
                    System.out.println("Attribute ID retrieved is" + attrId);
                    for (Enumeration vals = attr.getAll();vals.hasMoreElements(); System.out.println(attrId + ": " + vals.nextElement()));
               ctx.close();
          catch(Exception e)
               System.err.println("Problem querying DNS: " + e);
               e.printStackTrace();
The code runs fine and I get the output as follows
0 100 389 ib500ad1.indbank.in
0 100 389 ib500ad2.indbank.in
I have following questions
1) Do these servers which are returned are the domain controllers or the ldap servers.?(or both domaincontrollers and ldap server are same)
2) how to extract the server name alone from this string.
3) If these two server names are domain controllers then how can I query for a DNS for Domain controllers that advertise the service, then try to see if the domain controllers are answering, and if so choose the one with the fastest answer time (to avoid choosing a domain controller over WAN).
Thanks in advance.
-Hiubert

DNS lookup behaviour of cisco devices

Hi All,
Does anyone know how routers and ASA's deal with dns lookups for hostnames in their configuration?
For instance, an NTP server (0.pool.ntp.org), or a hostname in a VPN tunnel configuration.
I get the impression that there is only a 1 time resolve happening.
Can anyone confirm this and if this is the case, does anyone know how to modify this behaviour....
Thanks in advance.

Hi,
Cisco routers can be set up as proxy-dns but with a lot of limitations but ASA can't so on the ASA you must enter static host commands which stay indefinitely.
I don' t know if it's possible to tune the cache timeout and what is the default value.
Here is the link descibing the caveats for IOS:
http://nil.si/ipcorner/RouterDNS/
Regards.
Alain.

HELP! Disabling reverse DNS lookups on client

Is there a property that can be set to disable the reverse DNS
lookup for client requests? I read that if reverse lookups are
no working then client requests can take an extra 15-30 seconds.
In our environment reverse lookups are not something we can
count on so we would like to disable them completely. Please let
me know which property can be set it if any to accomplish this.
Regards,
Robert

Don't we all ;)
WL 5.1 sp3
Sol 2.6
J2 1.2.1_04
Rich Nill wrote in message <[email protected]>...
Paul,
What version of Weblogic are you running? I want to make sure we don'tsuffer
from the same problem.
Thanks,
Rich
Paul Iter wrote:
Would this patch have any impact on the problem I described in
"performance
degradation PROBLEM"?
Thanks,
Paul
Mark Griffith wrote:
There is another issue here though, when we print out server ID's we
call
java.net.InetAddress.toString() which ends up in a DNS call.
Contact support they have a one-off patch.
cheers
mbg
In article <[email protected]>, [email protected]
says...
Is there a property that can be set to disable the reverse DNS
lookup for client requests? I read that if reverse lookups are
no working then client requests can take an extra 15-30 seconds.
In our environment reverse lookups are not something we can
count on so we would like to disable them completely. Please let
me know which property can be set it if any to accomplish this.
Regards,
Robert
==================================================
NewsGroup Rant
==================================================
Rant 1.
The less info you provide about your problem means
the less we can help you. Try to look at the
problem from an external perspective and provide
all the data necessary to put your problem in
perspective.

9i app 9.0.2.01?Does the reverse DNS lookup have to be set up for a FQDN

HEy guys:
I'M ALWAYS GETTING STUCK IN THE SAME PLACE WHEN I AM TRYING TO INSTALL 9I APPSERVER 9.0.2.0.1 REL 2. ITS ALWATYS HAPPENING AT THE oRACLE db CONFIG assistant i have set up my host file and when i ping -a servername i get the full reply back ex. servername.domain.com but now when i ping -a 111.111.111.111 i do not get the host name back this is b/c i do not have the PTR record set up. Do i have to have a reverse dnslookup working for what oracle is stating is "FQDN" and not just the dns lookup working...how is oracle installer looking at this piece.
that is the only i see that i don't have when i look at my computer name (by the way this is a winnt environment)in properties it has the FQDN. i also have set up the host file correctly resembling 111.111.111.111 servername.domainname.com servername oracleinstall. What else am i missing here guys? thanks for the help in advance
regards,
robert

Actually, these issues were/are documented - see the addendum. Also, the install guide details which files need to be updated with the FQDN/IP.
Though it does not have to be setup in your DNS server (say if you are just doing it on a single tier to test), those machines which are looking to connect to it would need to have the proper additions to the hosts file as well.
As for why the 'non-default' http ports, this was a result of Unix security. Non-root users cannot start processes using ports below a specific range. As a result, oracle defaults them to a higher number allowing your oracle account whom lacks root access to start the http service.
As for non-oracle responses, this isn't really an official forumn. I believe those oracle peeps who do respond here are doing so on their own. If you need official/immediate responses, then i would recommend using metalink for an itar or the metalink forums.
Now on to Robert's second question. See metalink Note:209114.1: How to Change the Port used for Oracle 9iAS Portal 9.0.x. If you don't have access to metalink, let me know and I can forward the note or post it here.
Have fun!

For anyone who is experiencing slow DNS lookups...

I finally worked out what was wrong with my network config last night and thought I'd share it with everyone in a simgle post in the hope it'll help someone else.
I tried the BIND work around, but it wasn't all that much faster.
I tried disabling IPv6, but that didn't do much...
The solution?
In 'System Preferences' -> 'Network'
Go to configure the adaptor (Airport / Ethernet / etc)
In 'DNS Servers' where you'd normally specify the DNS servers given to you by your ISP... don't do this! As crazy as it sounds don't
Of course, if you're using newer routers you'd not be having this slow DNS lookup problem and specifying the ISPs DNS Servers would be appropriate... still
What you want to specify here is your ROUTER's IP:
eg. 192.168.0.1
With this simple modifcation you'll be fine. Why? You ask?
In Linux / OSX (I imagine in Unix as well) the way the lookups are carried out are different from Windows. I have other Windows computers on our network and they never had DNS lookup problems and they've been given the ISPs DNS IPs... anyway I think I'm talking out of my depth now heh.
This works!
Remember: Specify your router as the DNS Server!

I've had this problem on a G4 PowerMac running Panther, and it still had it after a Tiger upgrade. I just replaced it with a Core Duo MacMini, 10.4.7, same problem of slow DNS lookups (i.e., slow initial start to loading a web page, then it goes quickly). Windows machines on the same subnet have no such problem. I've tried the various suggestions on various forums, none of which worked. I tried:
- turn off IPv6 (no help)
- directly enter my ISPs DNS servers (no help)
- manually configure both IP and DNS (no help, went back to DHCP)
- swear at the computer (a little help, mentally)
After some more reading, I tried resolving some addresses using the host command from the Terminal:
host -v www.apple.com 24.34.240.9
where the IP address is one of the DNS servers for my ISP (Comcast). I got a no server found message! I then tried the second DNS server in the Comcast list (found from my router), also no server found. Tried the third one in Comcast's list of DNS servers, and it worked. Entered it in System Preferences -> Network as a DNS server, and now web browsing is zippy! I verified that the two DNS servers that MacOS couldn't see are also down as far as Windows was concerned (using the nslookup command in windows).
What this tells me is that the OS X algorithm for handling unreachable or slow DNS servers is different from that in Windows. Maybe Windows remembers a bad experience with a DNS server and uses ones that it has success with, while OS X just keeps trying them in order, slowing timing them out until it finds one that works?
This could also explain many of the puzzling symptoms people have been seeing (things work some times, other times not; some people have luck specifying the DNS server manually, others don't). It all depends on what DNS servers got distributed to the Mac via DHCP, and how far down the list you have to go to find one that is responsive.
Anyone reading this forum with technical knowledge of both UNIX and Windows DNS lookup implementations? Is there some way to tweak in MacOS to make it perform more like Windows in this situation (like, maybe shortening the DNS server failure timeout)?

Different DNS lookup stratergi for recognized and not recognized root domain

Firefox DNS lookup stratergy is different if the root domain in the URL is recognized (like .se) or not recognized (like .local). How can I add my selfintroduced root domain in the list of recognized root domains?

Firefox DNS lookup stratergy is different if the root domain in the URL is recognized (like .se) or not recognized (like .local). How can I add my selfintroduced root domain in the list of recognized root domains?

Wireless Intermittent Super Slow DNS lookup bug in 10.6.4

I don't normally post things on forums these days, as usually I can find just about any solution by searching long enough, but this issue has perplexed me to the point I actually had to come on here.
Believe me, that's a big deal, I don't give up easily.
I have spent -countless- hours searching, on here, on google, on any "solutions" or "technical" sites I could find, and the closest I can find to a solution are countless people complaining about the EXACT SAME PROBLEM that I have observed and, repeatedly, reproduced again and again, which in every single case boil down to this:
You had 10.6.x (x being 3 or less) with a wireless connection on your home network and all is well.
You upgraded to 10.6.4 and all seemed fine for maybe 24 hours or so... then it happens. You go to load a website, and it's "looking for site" or "waiting for site" in your status bar... hmm, maybe it's just this site you say, so you try another, or a few others in other tabs, but they all have the same problem.
You try to ping the sites, but the network utility can't resolve the domain to even ping them.
Your roommate, all the while, is surfing and gaming just fine on the exact same router you are on, so no, it's not the network hardware, it's not your ISP, hmm, what could it be?
All of a sudden, ALL of the sites you had in like 20 tabs load up at screaming speeds, "WOW" you say, "guess there must have just been some gunk in the wires or something" (notice the irony of the situation: no wires)... anyway, all seems fine again suddenly, surfing is fine for a few minutes, you're back to normal... and it happens again, suddenly NO site will resolve, NO dns will resolve, you can't check email or ping any domain... and so the cycle begins. Of course, you can just plug an ethernet cable straight into the router, but doesn't that kind of defeat the purpose of having wireless networking in the first place?
It continues like this, indefinitely, and it all starts roughly 24 hours after 10.6.4 has been installed.
I have read reports of people on macbooks, people on imacs, people on all sorts of different wireless hardware, but the symptoms are the same.
I know the problem is with the OS update, it's purely software. I know that it has nothing to do with hardware because simply reverting to 10.6.3 solves the problem -every single time- and then "upgrading" to 10.6.4 causes the problem to come back within 24 hours -every single time- (have been reverting using Time Machine to simplify this testing process), so no, where the problem is isn't what perplexes me; what perplexes me is that there are posts that started almost a few days after 10.6.4 came out, and so far there's STILL no fix? Are you freaking serious? Does the Apple programming team not have access to anything other than Apple-Branded Airport Extreme Base Stations to perform wireless network QA testing on?
Get a Linksys guys, grab a D-Link, go get some of the hardware people actually USE and test it on that and see what happens, it doesn't take long to see what's happening.
I blame the programmers because I am one myself and know how easy it is to screw up a rock-solid system with one little typo. Heck, which patch was it, 10.5.7 or 10.5.8 I think? Can't remember exactly, but it was supposed to be such a great "bug fix" patch... and it came with the config file for Apache set to DENY ALL INCOMING EXTERNAL CONNECTIONS by default (in a hidden file that can only be modified by the root user mind you... so much for the average user running a personal web server on THAT version), so yeah, one tiny mistake and it has huge consequences, my question is: what's taking so long to track down what's going on in 10.6.4 and fix it? Can we at least get a patch or something?
I find it really lame and really such a cop-out to see so many irrelevant "solutions" offered, "try specifying different DNS servers" (doesn't matter, whatever causes this bug doesn't care which servers you have specified, it simply sits there and does NOTHING for 2-3 minutes, and THEN when it actually DOES do a dns lookup, it gets the results in the time expected: instantly), to more extreme matters, like resetting hardware, which again has absolutely nothing to do with this bug.
Here is why anyone can see this is an obvious bug that the programming team needs to admit, investigate and correct:
A. happens immediately after the software update
B. happens to EVERYONE who uses traditional wireless routers for internet use
C. is 100% repeatedly reproducible
D. occurs on all different models of computers and all different ISP's and with all different DNS servers specified.
E. has the same symptoms on every system (lightning fast internet for 2-3 minutes, then "waiting for site" for 1-3 minutes)
F. affects EVERY network-using program on the computer (email, network utility, firefox, safari) SIMULTANEOUSLY
G. does not affect surfing to or interacting with IP addresses directly, only with trying to perform DNS lookups from ANY program with ANY dns server (or no dns server) set in network preferences.
Come on guys, just read it through, think about it for a few minutes, for anyone that has worked with and knows the underlying source code, and what changes went in between 10.6.3 and 10.6.4 specifically to networking, should have a light bulb pop up over their head and say "oh YEAH, we never uncommented that one line..." or something to that effect.

I see a very similar issue, but it's been occurring on my laptop for 4 or 5 months, which must be way before 10.6.4. My roommate and friend's laptops all work fine on my network. And my laptop works fine on anyone else's network. But MY laptop on MY network always gives the abysmal DNS performance as described in the original post: 40% of requests time out. Wireless or wired, it doesn't matter. Exact same behavior.
It also doesn't matter whether I use my Netgear router as DNS server, or my ISP, or OpenDNS, or Google. Exact same behavior.
When I do a network trace, it looks like most DNS requests my computer sends out simply never get responded to. (Could they be malformed when they hit the wire? I don't even see an error reply) A few make it through. And when there's a IPv6 (AAAA) record sent, my computer returns a "port unreachable" ICMP message. A screenshot of all of this dialogue is here:
http://img545.imageshack.us/i/screenshot20100913at114.png/
I recently had opportunity to cancel my cable service, and reinstate it for a lower price. They came out, tested the line (strong signal), gave me a new cable box. Yet the issue persists. Exact same behavior.
Firewall is disabled. I've deleted the network interfaces and added them back. Nothing helps.
(As I recall, this issue may even have been present before I reinstalled 10.6 over 10.5, so I'm not too confident a total reinstall would help.)
Any help? I'm about ready to buy a new laptop to fix this damned problem. Web browsing is nearly impossible, as is.

[SOLVED] Slow DNS lookup, I think

Hi
I have a really annoying problem. My DNS lookup in Arch is painfully slow. I know it's not a network problem, as I don't have any problems in my Ubuntu installation. I have tried to run two simple tests to show you what I mean. The first is a simple ping google.
########### Ubuntu ###########
carsten@carsten-laptop:~$ time ping -c 3 www.google.com
PING www.l.google.com (216.239.61.104) 56(84) bytes of data.
64 bytes from sn-in-f104.google.com (216.239.61.104): icmp_seq=1 ttl=245 time=17.4 ms
64 bytes from sn-in-f104.google.com (216.239.61.104): icmp_seq=2 ttl=245 time=20.6 ms
64 bytes from sn-in-f104.google.com (216.239.61.104): icmp_seq=3 ttl=245 time=11.4 ms
--- www.l.google.com ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2002ms
rtt min/avg/max/mdev = 11.465/16.529/20.641/3.809 ms
real 0m2.290s
user 0m0.000s
sys 0m0.004s
########### Arch ###########
carsten ~/Desktop $ time ping -c 3 www.google.com
PING www.l.google.com (216.239.61.104) 56(84) bytes of data.
64 bytes from sn-in-f104.google.com (216.239.61.104): icmp_seq=1 ttl=245 time=12.3 ms
64 bytes from sn-in-f104.google.com (216.239.61.104): icmp_seq=2 ttl=245 time=10.7 ms
64 bytes from sn-in-f104.google.com (216.239.61.104): icmp_seq=3 ttl=245 time=12.4 ms
--- www.l.google.com ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2007ms
rtt min/avg/max/mdev = 10.776/11.867/12.476/0.778 ms
real 0m15.305s
user 0m0.013s
sys 0m0.007s
Ubuntu: 0m2.290s vs. Arch: 0m15.305s.
In the second test I tried to fake a pacman update by downloading the .db files from my primary server. On both Ubuntu and Arch I used this simple script
repos=( core extra community )
time for repo in ${repos[@]}
do
wget http://archlinux.unixheads.org/$repo/os/i686/$repo.db.tar.gz
done
When I run it in, I get this result
########### Ubuntu ###########
carsten@carsten-laptop:~/Desktop$ ./updatetest
--2008-11-10 07:58:23-- http://archlinux.unixheads.org/core/os/i686/core.db.tar.gz
Resolving archlinux.unixheads.org... 204.152.186.174
Connecting to archlinux.unixheads.org|204.152.186.174|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 32515 (32K) [application/x-gzip]
Saving to: `core.db.tar.gz'
100%[=============================================================>] 32.515 --.-K/s in 0,1s
2008-11-10 07:58:23 (331 KB/s) - `core.db.tar.gz' saved [32515/32515]
--2008-11-10 07:58:23-- http://archlinux.unixheads.org/extra/os/i686/extra.db.tar.gz
Resolving archlinux.unixheads.org... 204.152.186.174
Connecting to archlinux.unixheads.org|204.152.186.174|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 422622 (413K) [application/x-gzip]
Saving to: `extra.db.tar.gz'
100%[=============================================================>] 422.622 242K/s in 1,7s
2008-11-10 07:58:25 (242 KB/s) - `extra.db.tar.gz' saved [422622/422622]
--2008-11-10 07:58:25-- http://archlinux.unixheads.org/community/os/i686/community.db.tar.gz
Resolving archlinux.unixheads.org... 204.152.186.174
Connecting to archlinux.unixheads.org|204.152.186.174|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 369845 (361K) [application/x-gzip]
Saving to: `community.db.tar.gz'
100%[=============================================================>] 369.845 206K/s in 1,8s
2008-11-10 07:58:27 (206 KB/s) - `community.db.tar.gz' saved [369845/369845]
real 0m3.837s
user 0m0.016s
sys 0m0.036s
########### Arch ###########
carsten ~/Desktop $ ./updatetest
--2008-11-10 08:01:33-- http://archlinux.unixheads.org/core/os/i686/core.db.tar.gz
Resolving archlinux.unixheads.org... 204.152.186.174
Connecting to archlinux.unixheads.org|204.152.186.174|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 32515 (32K) [application/x-gzip]
Saving to: `core.db.tar.gz'
100%[==============================================================================>] 32,515 --.-K/s in 0.1s
2008-11-10 08:01:47 (303 KB/s) - `core.db.tar.gz' saved [32515/32515]
--2008-11-10 08:01:47-- http://archlinux.unixheads.org/extra/os/i686/extra.db.tar.gz
Resolving archlinux.unixheads.org... 204.152.186.174
Connecting to archlinux.unixheads.org|204.152.186.174|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 422622 (413K) [application/x-gzip]
Saving to: `extra.db.tar.gz'
100%[==============================================================================>] 422,622 253K/s in 1.6s
2008-11-10 08:02:02 (253 KB/s) - `extra.db.tar.gz' saved [422622/422622]
--2008-11-10 08:02:02-- http://archlinux.unixheads.org/community/os/i686/community.db.tar.gz
Resolving archlinux.unixheads.org... 204.152.186.174
Connecting to archlinux.unixheads.org|204.152.186.174|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 369845 (361K) [application/x-gzip]
Saving to: `community.db.tar.gz'
100%[==============================================================================>] 369,845 262K/s in 1.4s
2008-11-10 08:02:17 (262 KB/s) - `community.db.tar.gz' saved [369845/369845]
real 0m44.153s
user 0m0.047s
sys 0m0.017s
Ubuntu: 0m3.837s vs. Arch: 0m44.153s
I get the same update time whenever I update pacman normally.
I have googled a lot to figure out an answer, but nothing helps, so I was hoping somebody could help me figure this out, as it's very annoying. My hosts file looks like this
hosts:
# /etc/hosts: static lookup table for host names
#<ip-address> <hostname.domain.org> <hostname>
127.0.0.1 localhost.localdomain localhost arch
# End of file
rc.conf:
# /etc/rc.conf - Main Configuration for Arch Linux
# LOCALIZATION
# LOCALE: available languages can be listed with the 'locale -a' command
# HARDWARECLOCK: set to "UTC" or "localtime"
# USEDIRECTISA: use direct I/O requests instead of /dev/rtc for hwclock
# TIMEZONE: timezones are found in /usr/share/zoneinfo
# KEYMAP: keymaps are found in /usr/share/kbd/keymaps
# CONSOLEFONT: found in /usr/share/kbd/consolefonts (only needed for non-US)
# CONSOLEMAP: found in /usr/share/kbd/consoletrans
# USECOLOR: use ANSI color sequences in startup messages
LOCALE="en_US.utf8"
HARDWARECLOCK="UTC"
USEDIRECTISA="no"
TIMEZONE="Asia/Singapore"
KEYMAP="dk"
CONSOLEFONT=
CONSOLEMAP=
USECOLOR="yes"
# HARDWARE
# MOD_AUTOLOAD: Allow autoloading of modules at boot and when needed
# MOD_BLACKLIST: Prevent udev from loading these modules
# MODULES: Modules to load at boot-up. Prefix with a ! to blacklist.
# NOTE: Use of 'MOD_BLACKLIST' is deprecated. Please use ! in the MODULES array.
MOD_AUTOLOAD="yes"
#MOD_BLACKLIST=() #deprecated
MODULES=(e100 mii iwl3945 fuse acpi-cpufreq cpufreq_ondemand cpufreq_conservative cpufreq_powersave loop !pcspkr !snd_pcsp)
# Scan for LVM volume groups at startup, required if you use LVM
USELVM="no"
# NETWORKING
# HOSTNAME: Hostname of machine. Should also be put in /etc/hosts
HOSTNAME="arch"
# Use 'ifconfig -a' or 'ls /sys/class/net/' to see all available interfaces.
# Interfaces to start at boot-up (in this order)
# Declare each interface then list in INTERFACES
# - prefix an entry in INTERFACES with a ! to disable it
# - no hyphens in your interface names - Bash doesn't like it
# DHCP: Set your interface to "dhcp" (eth0="dhcp")
# Wireless: See network profiles below
#eth0="eth0 192.168.0.2 netmask 255.255.255.0 broadcast 192.168.0.255"
eth0="dhcp"
INTERFACES=(!eth0 !wlan0)
# Routes to start at boot-up (in this order)
# Declare each route then list in ROUTES
# - prefix an entry in ROUTES with a ! to disable it
gateway="default gw 192.168.0.1"
ROUTES=(!gateway)
# Enable these network profiles at boot-up. These are only useful
# if you happen to need multiple network configurations (ie, laptop users)
# - set to 'menu' to present a menu during boot-up (dialog package required)
# - prefix an entry with a ! to disable it
# Network profiles are found in /etc/network.d
# This now requires the netcfg package
#NETWORKS=(main)
# DAEMONS
# Daemons to start at boot-up (in this order)
# - prefix a daemon with a ! to disable it
# - prefix a daemon with a @ to start it up in the background
DAEMONS=(syslog-ng !network hal !netfs crond fam wicd cups laptop-mode oss gdm)
SPLASH="splashy"
Thanks in advance!
Last edited by Sharpeee (2008-11-15 10:39:42)

Just tried to remove the "search..." line from my /etc/resolv.conf file, but nothing! It's okay if I remove the line after it connects right? Wicd overwrites the file anyways if I reconnect.
I don't really think changing to a different network-manager will help me. It works perfectly fine in Ubuntu with both network-manager and wicd, do don't think that's the problem. It must be a configuration file somewhere.
#### EDIT ####
I just tried to disable wicd and enable the wired network in /etc/rc.conf. After a reboot and it's still the same, even on the wired, so it's got be some other settings somewhere that's messing things up!
Also, for some reason my theme, in Gnome, isn't loaded after I disabled wicd? I have to manually run "gnome-appearance-manager"??
Last edited by Sharpeee (2008-11-11 05:01:46)

No DNS lookup for MX, A, AAAA or A6?

Similar Messages

Maybe you are looking for