Discussion on:
View:
Show:
Check if there is a proxy configured. This can be a proxy configuration in the browser or something as winsock. Depending on the program, they use the local dns or a dns server from the proxy location. Results can differ between those.
One may also want to check the Hosts file and see if a name/IP combination is set.
On UNIX servers it is also a good idea to check the resolving sequence. In /etc/nsswitch.conf or /etc/netsvc.conf (AIX) you can determine if it should look first in the host file before going to bind or first bind and then the host file. The correct sequence can eliminate resolving timeouts.
A great tool for making sure that you are preforming and an acceptable.
http://www.grc.com/dns/benchmark.htm
http://www.grc.com/dns/benchmark.htm
Is it safe to clear the DNS records in the server? Our DNS server (MSWS2k3 std sp2) has multiple entries for several IP addresses and several machine names, so if I try a "ping -a", I'm never sure if I'm getting the correct resolution. Can I clear the records from the server without disrupting the LAN users?
Please correct me if I am wrong - I am but a newbie - but NSLookup just looks up the DNS entries from the Primary DNS server and displays that, even if it's wrong.
We have entries which, if I do NSlookup on them, return the entry from the DNS table, yet the computer was decommissioned months ago and is no longer connected to teh network.
In these cases, and in the cases where I find multiple entries for the same host, I just delete their incorrect Forward entry. Likewise I first verify, then I delete incorrect Reverse lookup entries.
We have entries which, if I do NSlookup on them, return the entry from the DNS table, yet the computer was decommissioned months ago and is no longer connected to teh network.
In these cases, and in the cases where I find multiple entries for the same host, I just delete their incorrect Forward entry. Likewise I first verify, then I delete incorrect Reverse lookup entries.
They should be clearing themselves. You can always clear these.
If your authoritative records are wrong, however, you must correct them.
If your authoritative records are wrong, however, you must correct them.
Using NSLOOKUP debug options; Provides a list of DNS servers used to search for your query. Found is helpful to determine the last DNS server not responding to the query.
Ref link: http://support.microsoft.com/kb/200525
Example:
C:\nslookup
Default Server: XXX.XXXX.XXX.com
Address: xxx.xxx.xxx.xxx
>Set debug
>www.google.com
Ref link: http://support.microsoft.com/kb/200525
Example:
C:\nslookup
Default Server: XXX.XXXX.XXX.com
Address: xxx.xxx.xxx.xxx
>Set debug
>www.google.com
Don't forget to flush your systems local cache. For Windows:
ipconfig /flushdns
ipconfig /flushdns
I was waiting for someone to mention IPCONFIG. It's my go to guy when funny connection issues strike. I usually use this process when end users are having connectivity problems:
1. check router, check all cables.
2. power cycle router, power cycle end user machine
3. ipconfig/release ipconfig/renew
fixes 90% of my problems. I may have to add this lookup command mentioned in the article.. sounds useful.
1. check router, check all cables.
2. power cycle router, power cycle end user machine
3. ipconfig/release ipconfig/renew
fixes 90% of my problems. I may have to add this lookup command mentioned in the article.. sounds useful.
like the virus in the stack, sometimes users may decide they know better than IT, and change their DNS manually to an outside DNS like OpenDNS. That works fine for internet, but of course OpenDNS cannot resolve intranet names. If WINS is available in the domain, some names may be resolved through WINS, making troubleshooting even more difficult.
Easy solution: reset the workstation's DNS to "Obtain Automatically" (DHCP) and set a system policy so the user can't do it again.
Easy solution: reset the workstation's DNS to "Obtain Automatically" (DHCP) and set a system policy so the user can't do it again.
Well, I would say at clear the DNS Cache should always be on the top, and then verify the settings of TTL on records, as there are known bugs with Short TTL. Like if the A record have shorter TTL than the CNAME Record pointing to it, or if the Nameserver Record (NS) is shorter than the A Record belonging to that Nameserver, many DNS server will respond with server failure (specially those build on Bind8), this bug is also in Windows servers (and yes in W2K8 R2 as well). Clear the cache fix it temporary, but correcting the TTL or overvride them in the DNS Server (not possible in Windows DNS) will solve it as well.
That's nice, but you aren't telling us HOW to DO any on those things.
I have only done some of them before and don't remember how.
Where are the links to HOW to do those things?
I have only done some of them before and don't remember how.
Where are the links to HOW to do those things?
- Keyboard Shortcuts:
- Prev
- Next
- Toggle
