Changes

From SME Server
Jump to navigationJump to search
2,556 bytes added ,  12:30, 12 April 2016
Line 579: Line 579:  
install a minimal Centos7
 
install a minimal Centos7
 
configure network access during installation with option to restore.
 
configure network access during installation with option to restore.
  yum update -y
+
yum update -y
 
  yum install nano net-tools mc man htop top iftop iptraf screen openssh-clients mlocate wget vim
 
  yum install nano net-tools mc man htop top iftop iptraf screen openssh-clients mlocate wget vim
   Line 614: Line 614:     
  [smedev]
 
  [smedev]
  enabled=0
+
  enabled=1
  baseurl=http://mirror.canada.pialasse.com/releases/testing/10/smedev/$basearch
+
  MirrorList=http://mirrorlist.contribs.org/mirrorlist/smedev-10
 
  name=SME Server - dev
 
  name=SME Server - dev
 
  gpgcheck=1
 
  gpgcheck=1
 
  enablegroups=1
 
  enablegroups=1
 
  [smetest]
 
  [smetest]
  enabled=0
+
  enabled=1
  baseurl=http://mirror.canada.pialasse.com/releases/testing/10/smetest/$basearch
+
  MirrorList=http://mirrorlist.contribs.org/mirrorlist/smetest-10
 
  name=SME Server - dev
 
  name=SME Server - dev
 
  gpgcheck=1
 
  gpgcheck=1
Line 627: Line 627:  
  [smeos]
 
  [smeos]
 
  enabled=1
 
  enabled=1
  baseurl=http://mirror.canada.pialasse.com/releases/testing/10/smeos/$basearch
+
  MirrorList=http://mirrorlist.contribs.org/mirrorlist/smeos-10
 +
name=SME Server - dev
 +
gpgcheck=1
 +
enablegroups=1
 +
  [smeupdates]
 +
enabled=1
 +
MirrorList=http://mirrorlist.contribs.org/mirrorlist/smeupdates-10
 
  name=SME Server - dev
 
  name=SME Server - dev
 
  gpgcheck=1
 
  gpgcheck=1
 
  enablegroups=1
 
  enablegroups=1
        Line 641: Line 646:  
  yum install epel-release
 
  yum install epel-release
   −
install Repoforge 7 '''repo is no more maintained. However, for test purpose'''
  −
rpm -ivh http://pkgs.repoforge.org/rpmforge-release/rpmforge-release-0.5.3-1.el7.rf.x86_64.rpm
  −
  −
install ELRepo
  −
rpm --import https://www.elrepo.org/RPM-GPG-KEY-elrepo.org
  −
rpm -Uvh http://www.elrepo.org/elrepo-release-7.0-2.el7.elrepo.noarch.rpm
   
then
 
then
  yum clean all
+
  yum --enablerepo=* clean all
 
   
 
   
 
try installing SME10 packages, you might need to enable some repo using --enablerepo=smedev,epel,elrepo ... e.g.
 
try installing SME10 packages, you might need to enable some repo using --enablerepo=smedev,epel,elrepo ... e.g.
  yum install e-smith-base --enablerepo=smedev,smetest,epel,base,updates
+
  yum install e-smith-* smeserver-* --enablerepo=smedev,smetest,epel,base,updates
 
or
 
or
  yum install e-smith-base
+
  yum install e-smith-* smeserver-*
 +
 
 +
you will have to disable postfix :
 +
systemctl stop postfix
 +
systemctl disable postfix
 +
 
 +
latter it will be good to also deactivate networkManager, to leave network service doing the job, but you also need to configure e-smith db for the server first.
 +
#systemctl stop NetworkManager
 +
#systemctl disable NetworkManager
 +
 
 +
 
 +
AT this point ( 2016/03/26 [[User:Unnilennium|Unnilennium]] ([[User talk:Unnilennium|talk]]) 13:44, 27 March 2016 (CEST)) you could issue a signal-event post-upgrade; signal-event-reboot; signal-event bootstrap-console-save;signal-event-reboot . But you would have no more network running, as default entry are empty for network. So you are better to expand necessary template using specific event to test each service.
    
report here
 
report here
Line 1,145: Line 1,155:  
- grub -> grub2
 
- grub -> grub2
 
-
 
-
 +
 +
== Horde Testing from John Bennett III ==
 +
 +
This is taken from an email sent by John to the dev discussion list, which may well contain useful information about configuring the system for testing:
 +
 +
I personally didn't install the e-smith-horde,ingo,imp,turba packages, as those were built for horde 3, and I hadn't tested whether horde would even load using those files.  I saw that some of updated horde files were going to be installed, because someone changed the dependency to Requires: horde >= 3.2, which tries to pull them from the centos repo.
 +
 +
I tested what I have built and been using.  They loaded fine, and brought in all of the other dependencies.
 +
 +
Webserver wasn't working, enabled it using:
 +
service httpd start or /bin/systemctl start  httpd.service
 +
 +
Permanently enable using:
 +
systemctl enable httpd
 +
 +
Couldn't access webserver.
 +
 +
Also need to stop firewalld for now as well using:
 +
service firewalld stop or /bin/systemctl stop  firewalld.service
 +
 +
For testing, I just disabled it permanently using:
 +
/bin/systemctl disable firewalld.service
 +
 +
to renable:
 +
/bin/systemctl enable firewalld.service
 +
 +
I am now able to access the webserver.  Note, I put a sample index.html in the doc root to verify.
 +
 +
I saved a copy of a good httpd.conf, then I did an expand-template /etc/httpd/conf/httpd.conf so that all of the horde information would be there.  Re-starting httpd threw an error out, as expected.
 +
 +
I manually changed /etc/httpd/conf.d/ssl.conf and added the alias info.  Then manually changed httpd.conf and added all of the horde <Directory parameters, and remarked out the <directory stanzas that deny access to the whole server filesystem.  Lines 102-105 for me.  Didn't want to waste any additional time figuring out what needs to be really  done at this time.
 +
 +
Restarted httpd, then went to webmail, and the page appeared.
 +
 +
Now on to mariabd.
 +
 +
Enabled using:
 +
systemctl enable mariadb
 +
 +
cd /etc/e-smith/templates/etc/e-smith/sql/init/ and started to manually run all of the horde-related scripts.
    
== Notes ==
 
== Notes ==

Navigation menu