Topic: Migrated to SME 7.0 problems

[groutley]

Hello,
  I require some help.
I have Installed a clean v7.0  on a new server, and after removing custome templates and add ins to the best of my ability, used the copyfromdisk method to copy the OLD sme 6.01 data to the new server.

Now I have 2 problems.
I am not sure if one causes the other or not.
1/  dyndns  does not get updated with the new server IP address.
     It is setup exactly as it was previously on the old 6.01 server but does not work.  (I have updated existing bugtracker offering log collection if someone can tell me what logs are required.

2/  While all users can send email not a problem,  the server is not recieving any incoming email.  The clients are setup fine,  i.e.  they can recieve email from the server,  (setup a new Thunderbird and it reads in all the existing Inbox fine.)
  But a send from my work email to this home sme  never arrives.
 putting OLD sme 6.01 back in place recieves no problem.

Now perhaps this is related to 1/ ?  
I did try installing ddclient  and have had no success either.

Appreciate any suggestions.  (Wife and Kids will not be happy when they discover what I have done !!)

[groutley]

OK.. seems even worse..
 while the new 7.0   is allowing us to surf the web without a problem.
None of my public internet pages, or FTP  or ability to SSH to my server from the internet works.

I am guessing all this is related.
but I have no idea what the problem is, and I cannot find any setting anywhere that would cause this.
please help.

[frond]

groutley

> dyndns  does not get updated with the new server IP address.

All your other problems are likely to be related to this. The external world cannot find your server as the dyndns record shows the wrong IP for your domain.

You could test this by manually updating dyndns records to your current IP, and see if external access is then possible eg by browsing to your domain from an external location.

How to fix it ???
Log in as admin, run Configure this server, and disable the dyndns setup, save & reboot. Then run Reconfigure this server again, this time setting your dyndns account name and password where required.
Test to see if you have external access.

[groutley]

Thanks for the reply..
  I tried your suggestion, disable the dyndns setting in the admin setup panel, reboot, then re-enable the setting and reboot.  Interestingly the re-enable had remembered my user and password, so it is obviously not a full reset of configuration.
 Alas it did not resolve the problem of updating dyndns.org with my IP address.
I have sisnce got the ddclient working, and is successfully updating my domain name with the correct IP address,
Although I still am not recieving any email,  or putting any service on the internet.
All services http / https / ftp / ssh  etc work on my intranet,  but not on the internet, yet I have made the settings to allow this.

It is not the dyndns, as I should be able to point to my IP address and get the www  page up,  but it times out.

Seems the server is not putting itself on the internet at all ??

appreciate any other thoughs..  or I am going to have to back out to the old trusty 6.01 tonight.

[frond]

groutley

> Seems the server is not putting itself on the internet at all ??

It may take a few hours and even as long as 72 hours for all dns record servers to be updated. Until that time, access requests may still be using the old wrong IP.

Get someone external to access
http://yourcurrentIP
and see if that works

I assume you are set to gateway server mode and do in fact have a dynamic IP with your ISP ?

[frond]

groutley

> ....Interestingly the re-enable had remembered my user and password, > so it is obviously not a full reset of configuration.
 
It is a full reset in the sense that the service was disabled and re-enabled, but it would have been better if you had deleted those entries completely, then did the save & reboot, then re-entered the entries, just to force the new changes into the system, which was the whole point of my advice.

I would also advise you to install any updates using the server manager panel, as there have been many updates released recently, perhaps one of them resolves your problem.

I did here some complaints about a similar thing but did not follow the topic or bugs.

[groutley]

Thanks Frond,

I do not understand ..
"It may take a few hours and even as long as 72 hours for all dns record servers to be updated. Until that time, access requests may still be using the old wrong IP. "

If I put the old 6.01 system back in place,  the emails come rolling in straight away !  This would not seem to suggest that there is a latency in the internet propergating the IP address.

I am guessing it is something todo with 7.01 and it not putting itself out on the net at all..

http://www.routley.homeip.net  does not come up !  see for yourself.
it does resolve correctly now  so http://ipaddr  'should' also work.

perhaps I should try 'unsetting' the server / gateway mode and resetting ?

[frond]

groutley

It depends which root dns servers are accessed whether the changes are seen quickly or not.

> ... 7.01 and it not putting itself out on the net at all..

http://www.routley.homeip.net
http://http://155.143.151.124

Neither resolve although ping is OK.

ping www.routley.homeip.net

PING routley.homeip.net (155.143.151.124) 56(84) bytes of data.
64 bytes from CPE-155-143-151-124.vic.bigpond.net.au (155.143.151.124): icmp_seq=0 ttl=53 time=56.7 ms

ping 155.143.151.124

PING 155.143.151.124 (155.143.151.124) 56(84) bytes of data.
64 bytes from 155.143.151.124: icmp_seq=0 ttl=53 time=53.1 ms

It does sound like something is wrong with the sme7 box, probably a carry over from the sme6 upgrade.
You could try doing a clean install of sme7 & manually configure that with basic details to get you on the web, and then try accessing it externally.
That will prove/disprove where the problem is.

[groutley]

updates installed manually.
thanks for that..
not that it fixed the problem.

I will work thru kruhns method  to rid v6.01 of any extranious rpms.
but not tonight...
I guess time to call an end to fathers day, and bring good ol' sme6 back up,
even though it is half crippled now.

Appreciate all the help and suggestions..  I'll let you know how I go with this.
thanks

[groutley]

I am still struggling with this migrate from sme6 to sme7.
I have now killed sme6 enough that email no longer works there either !

I have removed a lot more rpms from sme 6  and I assume that is the cause of email dying.

What I have noticed on the sme7 after copyfromdisk is performed, is the
/home/e-smith/db  contains 2 files  for some of the databases.. i.e
-rw-r-----  1 root admin  8270 Sep  5 21:30 accounts
-rw-r-----  1 root admin  1122 Sep  5 18:30 accounts.1157450898
-rw-r-----  1 root admin 10976 Sep  5 21:30 configuration
-rw-r-----  1 root admin  7929 Sep  5 18:48 configuration.1157450898
-rw-r-----  1 root admin   361 Sep  5 21:30 domains
-rw-r-----  1 root admin   363 Sep  5 18:30 domains.1157450898
-rw-r-----  1 root admin   844 Apr 29 21:51 hosts
-rw-r-----  1 root admin   739 Sep  5 17:54 hosts.1157450898
-rw-r-----  1 root admin   362 Sep  5 21:30 networks
-rw-r-----  1 root admin   299 Sep  5 18:30 networks.1157450898

I am assuming the name without the .1157450898 is what is used..
investigating the configuration database shows much incorrect information
e.g EXTERNALIP=  wrong IP address
it also contains dansguardian info which I removed from sme6.

So, would I be correct in thinking that the configuration database is not being updated correctly ?
Why would this be ?

I would assume this gets updated by the signal-event post-upgrade ??
I definitely ran this..
  would it be an mismatching password on the db that does not allow the updates ?

Or am I totally off track here ?

[groutley]

OK,  I took a punt here..
I'm sure things couldnt get worse.. so what the heck !

I renamed the /home/e-smith/db/configuration  to configuration.old
then copied the configuration.1157450898 to be configuration.

then /sbin/e-smith/signal-event reboot

and viola !!!  I can now get to my web page, ftp and ssh from external !!
Not sure if email is coming in yet..  time will tell..  but I have my fingers crossed.

Now I know playing with the configuration database is dangerous if you don't know what you are doing,  and that includes me !
But so far it appears to have fixed my problem, and I don't see any adverse affects.
Some updates required in the server-manager  setup to correct some information that was lost from the db swap, but I am more happy than I have been for the last week !

Have I done the right thing ?  should I have addressed this differently than swapping the config db ?

[groutley]

Just to let you know..  my email has also started working on sme7.
That's a first for me !
So the db configuration swap  appears to have resolved all my problems.

So just waiting feedback to see if there was a better way to accomplish this.

To me the configuration was copied to the configuration.1157450898 during the copyfromdisk restore and post-upgrade  process.
So this database is the sme7 clean build db.
I had defined most things in this DB to get the clean install server online and yum updated.
The entire sme6 removal of add ins was a nightmare..
I had cleaned up / removed many rpms,  yet the kruhl procedure still indicated I had 66 different rpms to original !
And I really was not comfortable removing some of them as I think it was going to render sme6 a vegetable !

So  if this Database swap method meets proper procedure, I far prefer it.
I have accomplished moving all the users email, and files, all the ibays have moved, and all is working as I expected, and is a far easier method than butchering the existing sme6.

Open to comments here..  I am no expert in sme or Linux,  but this worked for me !