Koozali.org: home of the SME Server
Contribs.org Forums => Koozali SME Server 10.x => Topic started by: ilconte on October 26, 2021, 08:27:51 PM
-
hello everyone I wanted to know which correct procedure should I follow to pass an EMS 9.2 to an EMS 10
Thank you
:-P
-
search the forum there are plenty available.
all are based on backup and restore.
1/ backup
- console backup
- workstation backup
- migratehelper contrib (my favorite)
- backuppc, create a tar.gz
- lazyadmin tools + rsync of data
- affa (not my favorite in term of migration)
2/ install
3/ restore
- console restore
- workstation restore
- rsync
- mix of above solutions (my favorite)
4/ reinstall contribs
-
search the forum there are plenty available.
all are based on backup and restore.
1/ backup
- console backup
- workstation backup
- migratehelper contrib (my favorite)
- backuppc, create a tar.gz
- lazyadmin tools + rsync of data
- affa (not my favorite in term of migration)
2/ install
3/ restore
- console restore
- workstation restore
- rsync
- mix of above solutions (my favorite)
4/ reinstall contribs
A quick question,
Once the restore is done, is it best then to run the updates in server-manager before going on to install contribs ? IIRC it took a little while for my SME10 test new system to recognise that there were updates to run.
As an aside, do you happen to know if Fetchmail is part of the default install (ie to enable the function of the "Configure External Mailbox Retrieval" in server manager).
-
A quick question,
Once the restore is done, is it best then to run the updates in server-manager before going on to install contribs ? IIRC it took a little while for my SME10 test new system to recognise that there were updates to run.
As an aside, do you happen to know if Fetchmail is part of the default install (ie to enable the function of the "Configure External Mailbox Retrieval" in server manager).
Bit of a toss up, contribs first? then a full # yum update not sure, I think I have tended to do the yum update first.
Builtin basic fetchmail is different from the contrib fetchmail..see https://wiki.koozali.org/Fetchmail contrib has more
There is currently a dev version in /smetest I think, some ongoing work on the contrib, I think basic functionality is OK
-
I think this will probably still work if your setup isn't too fancy:
https://wiki.koozali.org/UpgradeDisk#Moving_from_SME_8.x_to_SME_9.x
On the old installation
signal-event pre-backup
Fire up v10 and mount the v9 drive - easier if it isn't LVM.
mount /dev/yourolddrive /mnt/tmp
db configuration setprop backup Program CopyFromDisk status enabled
signal-event pre-restore
signal-event restore-tape /mnt/tmp (note: this step may take a long time depending on how much data is being restored
signal-event post-upgrade
signal-event reboot
You will need a reconfigure for networking on first boot.
Terry might want to give this a test drive!!
I have added a couple of other pages into the Backup Category to try and group all the methods:
https://wiki.koozali.org/Category:Backup
-
There is currently a dev version in /smetest I think, some ongoing work on the contrib, I think basic functionality is OK
Looking at my SME10 test box I seem to have this installed:
Version : 1.6
Release : 19.el7.sme
I think I installed using:
yum --enablerepo=smecontribs install smeserver-fetchmail
signal-event console-save
But looking at the wiki the current SME10 version is 1.6-19.el7
Do I need to point it to a different repo ?
-
A quick question,
Once the restore is done, is it best then to run the updates in server-manager before going on to install contribs ? IIRC it took a little while for my SME10 test new system to recognise that there were updates to run.
As an aside, do you happen to know if Fetchmail is part of the default install (ie to enable the function of the "Configure External Mailbox Retrieval" in server manager).
For completeness of answer :
YES there is a small implementation of fetchmail in the core of SME, however if you have more than one account to retrieve or need advanced configuration as Terry pointed out you need to install the contrib.
Then you need to remove the configuration from the core SME as the contrib acts as a separate service ignoring the core configuration and they could start competiting to fetch messages.
-
Then you need to remove the configuration from the core SME as the contrib acts as a separate service ignoring the core configuration and they could start competiting to fetch messages.
ah, I didn't know that. At the moment I only use it to collect from a single additional pop box but would not know how to remove\amend the config from the core.
It's generally worked ok, but occasionally do get fetch messages if the server reboots. The 'extra' Fetchmail seems to collect all mail in the second pop box irrespective of if it's already been collected prior to the reboot.
-
Looking at my SME10 test box I seem to have this installed:
Version : 1.6
Release : 19.el7.sme
I think I installed using:
yum --enablerepo=smecontribs install smeserver-fetchmail
signal-event console-save
But looking at the wiki the current SME10 version is 1.6-19.el7
Do I need to point it to a different repo ?
the following in a term window, will inform what is installed.
# rpm -q semserver-fetchmail
smeserver-fetchmail.noarch 0:1.6-19.el7.sme
Looks like you have latest stable version, there is a dev version in /smetest smeserver-fetchmail.noarch 0:1.6-21.el7.sme see Bug 10289 and 11058 and 10538 if you are confident please test and add any comments to the bugs
-
I think this will probably still work if your setup isn't too fancy:
https://wiki.koozali.org/UpgradeDisk#Moving_from_SME_8.x_to_SME_9.x (https://wiki.koozali.org/UpgradeDisk#Moving_from_SME_8.x_to_SME_9.x)
another approach is to clone the server using this (https://wiki.koozali.org/Backup_server_config#Cloning_a_server_for_Test_purposes.2C_using_inclusion_.26_exclusion_files) procedure, with no files and emails, start the new server, rename the old one and change its IP address and then rsync all ibays and users' homedirs/mailboxes.
with this approach you don't have to wait for the restore to complete to be online again regarding emails.
HTH
-
You've had a URL breakdown mate.
$this > /dev/null
:rofl:
We're all getting too old.
-
You've had a URL breakdown mate.
you're right.. corrected in my post, I paste here too:
https://wiki.koozali.org/Backup_server_config#Cloning_a_server_for_Test_purposes.2C_using_inclusion_.26_exclusion_files
We're all getting too old.
indeed, but we'd stay humble :-D
-
We're all getting too old.
indeed, but we'd stay humble :-D
I disagree ... you are (we are) not old, but seasoned, like cheese ... we improve with time... :-D
-
indeed, but we'd stay humble :-D
I disagree ... you are (we are) not old, but seasoned, like cheese ... we improve with time... :-D
yes this is the deal, problem is my wife is ever complaining about the ever growing mature chees smell LOL :-)
-
yes this is the deal, problem is my wife is ever complaining about the ever growing mature chees smell LOL :-)
ROTFL
-
yes this is the deal, problem is my wife is ever complaining about the ever growing mature chees smell LOL :-)
:P
that is what we get from cheesy jokes