Is Easyfind down?

[Ubi]

there's still a problem with secret-key translation though. A fair few machines do not get proper updates because the secret is mangled. This I think has always been the case though (i never changed that code).

The problem really seems to be in the original code, on both ends of the interaction. The client sends updates that is likes a new IP address. If server then cannot find client MAC, it inserts a new record, but doesnt fill in the preferred hostname. It cannot do this because client does not supply this information. However the reality is that because of this , server creates an incomplete record in the hwlist table which breaks the code later on.

Now I changed the code that it doesn't insert a record into hwlist if it doesn't have the hostname. I tested deleting, creating, renaming a host and it seems to work fine.

[Puma]

Dear Ubi,

First of all thank you very much for your time and effort!

I tried the easyfind with:

Code: Select all

 /usr/lib/web-admin/easyfind.pl
{"opcode":1,"record":{"domain_id":"1","content":"##.##.54.207","name":"###.myownb3.com","type":"A","id":"84"},"error":false,"operation":"IPupdate"}

First time I did this my old IP address was shown.
So I did an update in the b3 and b2 webinterface which was succesfull

After that I got the right IP address when doing /usr/lib/web-admin/easyfind.pl

Tested:

Code: Select all

host -s ####.myownb3.com ns1.mybubba.org
Using domain server:
Name: ns1.mybubba.org
Address: 95.170.90.16#53
Aliases: ###.myownb3.com has address ##.##.54.207

This looks OK to me as well.

But when I try to connect with the browser to http://###.myownb3.com it fails, should I use a different link?

Puma

[Ubi]

you need to restart your browser. These days browswers have an internal DNS cache that changes slowly. If you want to knwo what your OS thinks the IP is, just "ping yourname.myownb3.com"

[Puma]

Ubi,

Thanks that did the trick It works perfectly now!

Puma

[kruemelprinz]

Dear all

thanks for your efforts trying to help us out with the easyfind problem.
As gordon recommended, I tried to force an update of my easyfind record by disabling the function in the webadmin interface and then registering again. One positive thing here is that, when enabling easyfind again, I get a message saying "update successful". Previously, I got a timeout connecting to the service.

Testing easyfind with /usr/lib/web-admin/easyfind.pl yields the following result:

Code: Select all

/usr/lib/web-admin/easyfind.pl
{"opcode":1,"record":{"domain_id":"1","content":"92.***.***.***","name":"****.myownb3.com","type":"A","id":"2135"},"error":false,"operation":"IPupdate"}

It shows the right IP adress, but when I try to ping my adress, it does not find the host

Code: Select all

ping ****.myownb3.com
ping: cannot resolve ****.myownb3.com: Unknown host

Connecting to ****.myownb3.com via browser fails (after clearing cache and doing browser restart).

A fair few machines do not get proper updates because the secret is mangled.

As already earlier mentioned, I did not have any problems with easyfind previously, so I assume that I should not be one of these?

Is there anything more to test to provide you with more info?

[Ubi]

the problem is on your end. My guess is that you first pinged your machine, and then updated easyfind. This first ping loaded the cache of your upstream DNS provider, and thus it takes a while for the change to propagate.

Code: Select all

C:\Users\ubi>ping su****drobert.myownb3.com
Pinging su****drobert.myownb3.com [92.*.*.117] with 32 bytes of data:
Request timed out.

[kruemelprinz]

the problem is on your end. My guess is that you first pinged your machine, and then updated easyfind. This first ping loaded the cache of your upstream DNS provider, and thus it takes a while for the change to propagate.

OK, thanks for checking that. Well actually I turned off/ restarted easyfind in the webadmin interface first. I then pinged *****.myownb3.com both from my mobile internet device and from home (which of course is the same IP as the B3). Both ping attempts failed.

Now I even tried to use another unique easyfind name and updated via webadmin interface. It said "update successful". When I ping the new *****.myownb3.com, I get the same error:

Code: Select all

ping test***.myownb3.com
ping: cannot resolve test***.myownb3.com: Unknown host

although

Code: Select all

/usr/lib/web-admin/easyfind.pl
{"opcode":1,"record":{"domain_id":"1","content":"92.*.*.*","name":"test***.myownb3.com","type":"A","id":"2139"},"error":false,"operation":"IPupdate"}

The new name should not be in any DNS provider caches, as it was never used before. Is there anything else I can do to search for the error causing the problem?

[Gordon]

If you would change your DNS settings to ns1.mybubba.org then the change would be instant. The problem with this is that you won't be able to get an address for hosts that are not in myownb?.com. The change has to propagate throughout the whole of the internet and how fast that goes depends on the Time To Live (TTL) that has been set on the records, the amount of 'hops' that are in between mybubba.org and the DNS server that your are using. And chance. Because if TTL is set to e.g. 1 minute, every hop can take up to 1:59.

[Ubi]

Ns1.mybubba.org will not act as a recursive ns, so setting ns1.mybubba.org in resolve.conf would be foolish.

However he's not getting the wrong ip, he's getting a failure to resolve. Its almost like he's still asking the old nameservers. In that case, try

Code: Select all

dig -t ns myownb3.com

and see what that gives. In any way, the problem is on the client side. The database received your update and i showed you i can resolve your b3 without issues.

[kruemelprinz]

Code: Select all

dig -t ns myownb3.com

; <<>> DiG 9.7.3 <<>> -t ns myownb3.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: SERVFAIL, id: 7423
;; flags: qr rd ra; QUERY: 1, ANSWER: 0, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;myownb3.com.			IN	NS

;; Query time: 36 msec
;; SERVER: 192.168.1.254#53(192.168.1.254)
;; WHEN: Wed Dec  3 20:49:40 2014
;; MSG SIZE  rcvd: 29

192.168.1.254 is my internal Thompson modem/router IP.

Unfortunately, I have no idea at all what this is telling me.

[Ubi]

You did not get an answer, This means your local DNS server (192.168.1.254) is buggered. You should get this:

Code: Select all

[root@php-sites ~]# dig -t ns myownb3.com

; <<>> DiG 9.8.2rc1-RedHat-9.8.2-0.30.rc1.el6 <<>> -t ns myownb3.com
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 43818
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 0

;; QUESTION SECTION:
;myownb3.com.			IN	NS

;; ANSWER SECTION:
myownb3.com.		86400	IN	NS	ns2.mybubba.org.
myownb3.com.		86400	IN	NS	ns1.mybubba.org.

;; Query time: 53 msec
;; SERVER: 80.69.67.66#53(80.69.67.66)
;; WHEN: Wed Dec  3 20:57:12 2014
;; MSG SIZE  rcvd: 76

if 192.168.1.254 is your router, then you need to go to another forum, or start another thread. This is not an easyfind problem so for clarity sake lets not continue here.

[Gordon]

Ns1.mybubba.org will not act as a recursive ns, so setting ns1.mybubba.org in resolve.conf would be foolish.

That was what I said. I just used more words for `not recursive`. The word of interest - although unspoken - being patience. It will most likely work by now.

PS The router in question may in fact not support the operation requested. This is hinted by the fact that if it was showing expired info, it would have reported ns?.excito.org.

[wm.bubba]

Thanks for all the hard work on getting Easyfind up and running again.
But all is not well.

Running /usr/lib/web-admin/easyfind.pl now gives:

Code: Select all

wies@b2:~$ sudo /usr/lib/web-admin/easyfind.pl
Updating IP on file.
Wrote config to file
{"opcode":1,"record":{"domain_id":"2","content":"###.###.241.41","name":"###.myownb2.com","type":"A","id":"1544"},"error":false,"operation":"IPupdate"}

and checking the namesavers:

Code: Select all

wies@b2:~$ host -s ###.myownb2.com ns1.mybubba.org
Using domain server:
Name: ns1.mybubba.org
Address: 95.170.90.16#53
Aliases: 
###.myownb2.com has address ###.###.7.45

Code: Select all

wies@b2:~$ host -s ###.myownb2.com ns2.mybubba.org
Using domain server:
Name: ns2.mybubba.org
Address: 95.170.90.19#53
Aliases: 
###.myownb2.com has address ###.###.241.41

ns1.mybubba.org has been correctly updated, but ns2.mybubba.org still has the IP address it had yesterday.
Testing this with 'ping' sometime gives me the IP address from ns1 and sometine the one from ns2.

Hope there's an easy solution to this!

[Ubi]

Thanks. Good report. I'll try to see why the slave isnt updating asap

[Ubi]

replication issue fixed and tested for domain-record 1544