Release of NethServer 7.4.1708 Final

[mrmarkuz (Markus Neuberger)]

Do you have a DHCP red interface?

No, static green, maybe I should change to red? From internet only necessary ports are allowed. It’s a provided managed host with minimal centos and I installed NethServer on it.

OK restarting shorewall and httpd today in the morning just made my httpd virtualhost mirror work again. During the day I noticed that SoGO has white screen after login, roundcube login not working(storage error), Webtop login not working(loading animation forever) and EAS also dead.
So I restarted dovecot in the evening and now all is working again. Maybe you can see something in the following output(I faked public IP to 1.2.3.100), I’ll try a reboot of the server later. Could not find a dovecot error in the logs…

[root@nethserver ~]# db networks show
eth0=ethernet
    bootproto=none
    gateway=1.2.3.1
    ipaddr=1.2.3.100
    netmask=255.255.255.0
    role=green
ppp0=xdsl-disabled
    AuthType=auto
    FwInBandwidth=
    FwOutBandwidth=
    Password=
    name=PPPoE
    provider=xDSL provider
    role=red
    user=

[root@nethserver ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether da:b2:2d:94:32:04 brd ff:ff:ff:ff:ff:ff
    inet 1.2.3.100/24 brd 1.2.3.255 scope global eth0
       valid_lft forever preferred_lft forever
    inet6 0000::0000:0000:0000:3604/64 scope link
       valid_lft forever preferred_lft forever
3: ip_vti0@NONE: <NOARP> mtu 1480 qdisc noop state DOWN qlen 1
    link/ipip 0.0.0.0 brd 0.0.0.0
4: virbr0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN qlen 1000
    link/ether 52:54:00:1f:3b:48 brd ff:ff:ff:ff:ff:ff
    inet 192.168.122.1/24 brd 192.168.122.255 scope global virbr0
       valid_lft forever preferred_lft forever
5: virbr0-nic: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast master virbr0 state DOWN qlen 1000
    link/ether 52:54:00:1f:3b:48 brd ff:ff:ff:ff:ff:ff

EDIT:

I removed webvirtmgr package, rebooted and everything works now. No firewall error in dashboard anymore. But I still have this virtual bridge. Maybe providers internal network for backup?
Thanks for your help!

[davidep (Davide Principi)]

virbr0

Is it from libvirt or provisioned by the host provider?

should change to red?

No, green is the correct choice in this case.

[mrmarkuz (Markus Neuberger)]

Is it from libvirt or provisioned by the host provider?

I don’t know anymore. Is there a way to check it? May I remove libvirt if I don’t use webvirtmgr?
If the provider needs it he will communicate I think…

[davidep (Davide Principi)]

I removed webvirtmgr package, rebooted and everything works now. No firewall error in dashboard anymore. But I still have this virtual bridge. Maybe providers internal network for backup?

I bet you can remove libvirt package too! If you want to be sure, look at the package history:

yum history pkg-info libvirt

[flatspin (Ralf Jeckel)]

Update worked without issue. But there were no warning about updating to new version.
Machine is a VM ported form NS6 to NS7.3 via rsync and than updated via softwarecenter to 7.4.
Accountprovider is local AD
Installed Modules:

I think it’s a realistic scenario.
Well done and a big praise to @dev_team

[m.traeumner (Michael Träumner)]

But there were no warning about updating to new version.

That’s right I haven’t it to and what’s about the sogo problem like at the following thread:

[SOLVED]Sogo+AD+Chinese Typo on sogo.conf - I do not understand

Using LDAP instead of AD is not a solution for everybody.

[davidep (Davide Principi)]

But there were no warning about updating to new version.

The warning appears when upstream (CentOS) releases a new minor version, which is a situation that created some issues in the past. Here we are releasing the NethServer 7.4, which is tested by our @quality_team and 100% safe (hopefully), so no banner is required.

Using LDAP instead of AD is not a solution for everybody.

Do you think there is a bug in SOGo configuration? Sorry I didn’t read that topic, I’m quite busy now. I’d really appreciate if someone files a bug here or on GitHub with reproducible steps!

[flatspin (Ralf Jeckel)]

The warning appears when upstream (CentOS)

Thanks for clearifying this.

[flatspin (Ralf Jeckel)]

what’s about the sogo problem like at the following thread

No problem here. SOGo works. I can login wth my users and the service starts.
Binary password is in sogo.conf.

All seems fine so far.

EDIT: I was also able to restore events and preferences on the new machine.

[m.traeumner (Michael Träumner)]

Hi Ralf,
thanks for testing.

[mrmarkuz (Markus Neuberger)]

I can confirm that. Binary password in sogo.conf works:

[SOLVED]Sogo+AD+Chinese Typo on sogo.conf - I do not understand

I got the AD naming bug: If you enter no DNS Server IP on joining NS, the domain is AD instead of DOMAIN. I freshly installed the VM with NS 7.4b1, updated, joined AD, changed ldap uri to ldaps so account-provider-test works again. Then I installed sogo and it just worked. If you change the ldap uri it is correctly written to sogo.conf hostname: /* 45 AD authentication */ SOGoUserSources =( { id = AD_Users; type = ldap; CNFieldName = cn; IDFieldName = …

[mrmarkuz (Markus Neuberger)]

I bet you can remove libvirt package too!

You are right, I removed libvirt and on next reboot the virbr0 was away. I searched the net and 192.168.122.1 seems to be a default libvirt bridge. Thanks again!

[m.traeumner (Michael Träumner)]

I can confirm that. Binary password in sogo.conf works:

I’ve updated to 7.4 final yesterday, but without a reboot. The password is still cleartext, not binary.
Did you have done a reboot after that?
I can’t do at the moment.

[mrmarkuz (Markus Neuberger)]

Did you have done a reboot after that?

On an updated LDAP Nethserver 7.4 I have no binary password too and it works after reboot.
On a fresh installed AD NS 7.4 I have binary password and it works after reboot.
Sorry, I don’t have an updated AD NS 7.4. Maybe it’s like already created passwords stay in cleartext.

[m.traeumner (Michael Träumner)]

On an updated LDAP Nethserver 7.4 I have no binary password too

@davidep Do you have an idea for this? Is it a bug?

[davidep (Davide Principi)]

The upstream samba package was updated from 4.4 to 4.6 in CentOS 7.4. I suppose they changed the random password generation algorithm…

On our side, about one year ago we disabled the automatic SSSD password renewal function, introduced in 7.3 that breaks non-sssd aware modules (like SOGo, WebTop, Roundcube…).

I think we should stop to use the machine password to browse Samba AD LDAP and always require a dedicated account, like we do with MS AD. Local AD provider can generate it automatically.

Is it a bug?

No, as long as individual configuration templates and applications support binary data.

[111 (Alexey Diachenko)]

Hello. I decided to write about the problem after the update: I get the following
systemd-219-42.el7_4.4.x86_64 conflicts with dracut <033-499
Do you have any tips on how to solve the problem?

[mrmarkuz (Markus Neuberger)]

Hi @111,

actual dracut version is 033-502, so I think dracut will be updated and the conflict is gone…
In which situation do you get this error? On Update? Just found in /var/log/messages?

[filippo_carletti]

rpm -q dracut
If you find multiple packages at different versions use yum remove on the older ones.

[davidep (Davide Principi)]

A new ISO containing a small fix is available. Same file name - checksums updated.

• MD5 bef85e768c5a0fb2b8f35d91e88f3ee7
• SHA1 b38b74608e7f0462f4369f7b1effbe92b0a0cede
• Torrent Info 7C62855279337B6ADC17920FC424288784B08D84

Please update the torrent seeders from

http://packages.nethserver.org/nethserver/iso/nethserver-7.4.1708-x86_64.iso.torrent