Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found

[Claudio Soprano]

You are welcome.

I'm happy you can replicate the problem.

Infact the problem is the name "default" used for the bridge any other 
name works (i can't test all the names :) )


Claudio

On 10/09/17 08:39, Edward Haas wrote:

This is a limitation of the brctl tool.

Attempting to define a bridge named 'default' fails:
# brctl addbr default
add bridge failed: Invalid argument

Thanks,
Edy.


On Thu, Sep 7, 2017 at 8:26 PM, Claudio Soprano 
> wrote:


Ok

After installed the host, i need to attach it the networks

so I got the Unexpected Exception while I'm trying to add a vlan 1
(called default) to a Host, in the Setup_networks gui.

I attached:

- a screenshoot of all the network we have

- two screenshoots for the "default" network proprierties (one for
vnic profile, other for the proprierties)

- supervdsm.log from host ovc2n06 (i attached all the file, but
you will find all at the beginning)

Like already explained, it is simple to replicate the problem with
a CentOS v7.3 host, you need to create a Bridge called default
(ifcfg-default) and a Vlan 1 attached to it (ifcfg-enp12s0.1), do
a systemctl restart network and you will get the error.
To fix the problem rename the bridge to "def" for example and
change the Vlan 1 file configuration to link the new name and then
all works.

This problem is found only on the CentOS v7.3 on old CentOS v7.2
hosts the "default" bridge works, to override this problem we
changed the name to "def" for now.

All the hosts now are ovirt v4.1.5 and CentOS v7.3 we have also
some old hosts on CentOS v7.2

Claudio


On 07/09/17 13:46, Dan Kenigsberg wrote:

I do not understand which steps you are doing until you get the
"Unexpected Exception" dialog.
Could you detail them?

Also, can you share supervdsm.log from the relevant host, so
we can
understand the nature of that unexpected exception?

On Wed, Aug 30, 2017 at 12:31 PM, Claudio Soprano
> wrote:

The problem is not the host-deploy, infact the deploy
finished without
errors, if i remember good in the deploy process the
network are not set, i
need to attach each network i need manually, under hosts,
network interface,
setup host networks

and the problem is there, when i attach all the VLANs
(including the
default, that is VLAN 1) it gave error, default is not
present in the
system, if i add all the VLANS (excluding the default,
that is VLAN 1) it is
OK, but the host will not activate because missing the
default vlan.

If i manually add the 2 configuration files (ifcfg-default
and ifcfg-intX.1)
to the host and make a systemctl network restart, it gave
error then

ip addr reports

the vlan 1 is added, but not the Bridge to default (the
default is missing
error).

I don't know if the name "default" (for the network, VLAN
1) could be the
problem, but i can't rename now, because i need to detach
from the hosts,
but actually i can't.

I added 3 screenshoots to show you the situation, before
adding VLAN 1
("default" network), the error when adding it, and the
info on the error.


-- 


       /        |    /   _/           /    |    /  _/   
|    /
      /       / |   /   /                /   / |   /  /        / |   /
     /       /  |  /   ___/   _/    /   /  |  /  ___/     /  |  /
    /       /   | /   /                /   /   | /   /       /   | /
  __/ _/   __/  _/               _/  _/   __/  _/    _/   __/

Claudio Soprano                phone: (+39)-06-9403.2349
/2355
Computing Service              fax: (+39)-06-9403.2649

LNF-INFN                       e-mail: claudio.sopr...@lnf.infn.it

Via Enrico Fermi, 40           www: http://www.lnf.infn.it/
I-00044 Frascati, Italy




--

   /|/   _/   /|/   _/|/
  /   / |   /   //   / |   /   // |   /
 /   /  |  /   ___/   _//   /  |  /   ___/ /  |  /
/   /   | /   //   /   | /   //   | /
  __/ _/   __/  _/   _/  _/   __/  _/   _/   __/

Claudio Sopranophone:  (+39)-06-9403.2349/2355
Computing Service  fax:(+39)-06-9403.2649
LNF-INFN   e-mail: 

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found

[Edward Haas]

This is a limitation of the brctl tool.

Attempting to define a bridge named 'default' fails:
# brctl addbr default
add bridge failed: Invalid argument

Thanks,
Edy.


On Thu, Sep 7, 2017 at 8:26 PM, Claudio Soprano  wrote:

> Ok
>
> After installed the host, i need to attach it the networks
>
> so I got the Unexpected Exception while I'm trying to add a vlan 1 (called
> default) to a Host, in the Setup_networks gui.
>
> I attached:
>
> - a screenshoot of all the network we have
>
> - two screenshoots for the "default" network proprierties (one for vnic
> profile, other for the proprierties)
>
> - supervdsm.log from host ovc2n06 (i attached all the file, but you will
> find all at the beginning)
>
> Like already explained, it is simple to replicate the problem with a
> CentOS v7.3 host, you need to create a Bridge called default
> (ifcfg-default) and a Vlan 1 attached to it (ifcfg-enp12s0.1), do a
> systemctl restart network and you will get the error.
> To fix the problem rename the bridge to "def" for example and change the
> Vlan 1 file configuration to link the new name and then all works.
>
> This problem is found only on the CentOS v7.3 on old CentOS v7.2 hosts the
> "default" bridge works, to override this problem we changed the name to
> "def" for now.
>
> All the hosts now are ovirt v4.1.5 and CentOS v7.3 we have also some old
> hosts on CentOS v7.2
>
> Claudio
>
>
> On 07/09/17 13:46, Dan Kenigsberg wrote:
>
>> I do not understand which steps you are doing until you get the
>> "Unexpected Exception" dialog.
>> Could you detail them?
>>
>> Also, can you share supervdsm.log from the relevant host, so we can
>> understand the nature of that unexpected exception?
>>
>> On Wed, Aug 30, 2017 at 12:31 PM, Claudio Soprano
>>  wrote:
>>
>>> The problem is not the host-deploy, infact the deploy finished without
>>> errors, if i remember good in the deploy process the network are not
>>> set, i
>>> need to attach each network i need manually, under hosts, network
>>> interface,
>>> setup host networks
>>>
>>> and the problem is there, when i attach all the VLANs (including the
>>> default, that is VLAN 1) it gave error, default is not present in the
>>> system, if i add all the VLANS (excluding the default, that is VLAN 1)
>>> it is
>>> OK, but the host will not activate because missing the default vlan.
>>>
>>> If i manually add the 2 configuration files (ifcfg-default and
>>> ifcfg-intX.1)
>>> to the host and make a systemctl network restart, it gave error then
>>>
>>> ip addr reports
>>>
>>> the vlan 1 is added, but not the Bridge to default (the default is
>>> missing
>>> error).
>>>
>>> I don't know if the name "default" (for the network, VLAN 1) could be the
>>> problem, but i can't rename now, because i need to detach from the hosts,
>>> but actually i can't.
>>>
>>> I added 3 screenshoots to show you the situation, before adding VLAN 1
>>> ("default" network), the error when adding it, and the info on the error.
>>>
>>
> --
>
>/|/   _/   /|/   _/|/
>   /   / |   /   //   / |   /   // |   /
>  /   /  |  /   ___/   _//   /  |  /   ___/ /  |  /
> /   /   | /   //   /   | /   //   | /
>   __/ _/   __/  _/   _/  _/   __/  _/   _/   __/
>
> Claudio Sopranophone:  (+39)-06-9403.2349/2355
> Computing Service  fax:(+39)-06-9403.2649
> LNF-INFN   e-mail: claudio.sopr...@lnf.infn.it
> Via Enrico Fermi, 40   www:http://www.lnf.infn.it/
> I-00044 Frascati, Italy
>
>
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found - SOLVED

[Dan Kenigsberg]

On Wed, Aug 30, 2017 at 1:25 PM, Claudio Soprano
 wrote:
> More investigation on this for you :)
>
> the bug is only on CentOS v7.3 infact on previous hosts with CentOS < v7.3 i
> have no problem about the name of the network.
>
> I discovered also that i don't need only to rename the network name, but
> also the vnic profile associated, infact with network name def, and vnic
> profile default, it still didn't work, renamed network name and vnic profile
> to def both, now works on CentOS v7.3 with no errors also from GUI.
>
> when i create a network both network name and vnic profile are called the
> same (for my it was default).
>
> Hope that my info can help you.
>
> Another boring issues is that i need to detach the network to all the hosts
> to rename only, but also to change the network on all VMs that use it also
> if they are stopped, i was lucky only 2 VMs and only 4 hosts, but i can't
> imagine people have 40-50 hosts and maybe 100-200 VMs on a network.

Unfortunately, we are currently unable to modify a network name while
it is attached to any host, and in particular, while it is used by a
running VM. We even have an old feature request to allow that

   https://bugzilla.redhat.com/show_bug.cgi?id=1121101

I hope we manage to implement it in time for oVirt-4.2, but I cannot
promise it right now.

Regards,
Dan.
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found

[Dan Kenigsberg]

I do not understand which steps you are doing until you get the
"Unexpected Exception" dialog.
Could you detail them?

Also, can you share supervdsm.log from the relevant host, so we can
understand the nature of that unexpected exception?

On Wed, Aug 30, 2017 at 12:31 PM, Claudio Soprano
 wrote:
> The problem is not the host-deploy, infact the deploy finished without
> errors, if i remember good in the deploy process the network are not set, i
> need to attach each network i need manually, under hosts, network interface,
> setup host networks
>
> and the problem is there, when i attach all the VLANs (including the
> default, that is VLAN 1) it gave error, default is not present in the
> system, if i add all the VLANS (excluding the default, that is VLAN 1) it is
> OK, but the host will not activate because missing the default vlan.
>
> If i manually add the 2 configuration files (ifcfg-default and ifcfg-intX.1)
> to the host and make a systemctl network restart, it gave error then
>
> ip addr reports
>
> the vlan 1 is added, but not the Bridge to default (the default is missing
> error).
>
> I don't know if the name "default" (for the network, VLAN 1) could be the
> problem, but i can't rename now, because i need to detach from the hosts,
> but actually i can't.
>
> I added 3 screenshoots to show you the situation, before adding VLAN 1
> ("default" network), the error when adding it, and the info on the error.
___
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found - SOLVED - BUG

[Claudio Soprano]

No one answered to my messages, so i don't know if someone replicated 
the bug or not in CentOS 7.3.


Let me know if you need more info.

Claudio

On 30/08/17 12:25, Claudio Soprano wrote:


More investigation on this for you :)

the bug is only on CentOS v7.3 infact on previous hosts with CentOS < 
v7.3 i have no problem about the name of the network.


I discovered also that i don't need only to rename the network name, 
but also the vnic profile associated, infact with network name def, 
and vnic profile default, it still didn't work, renamed network name 
and vnic profile to def both, now works on CentOS v7.3 with no errors 
also from GUI.


when i create a network both network name and vnic profile are called 
the same (for my it was default).


Hope that my info can help you.

Another boring issues is that i need to detach the network to all the 
hosts to rename only, but also to change the network on all VMs that 
use it also if they are stopped, i was lucky only 2 VMs and only 4 
hosts, but i can't imagine people have 40-50 hosts and maybe 100-200 
VMs on a network.


Anyway now i have Centos V7.3 and V7.2 hosts both working.

Claudio

On 30/08/17 11:52, Claudio Soprano wrote:


Ok i SOLVED the problem, but i don't know if it is a BUG, the problem 
is the name default, but i had it working with ovirt v3 and v4.0pre 
for sure.


I don't know if the check has been added in the setup_host_network or 
not.


This is that i did

i manually added a ifcfg-def and ifcfg-enp12s0.1 file (first for the 
bridge and the second for the VLAN) to an host that don't have VLAN 1


[root@X network-scripts]# more ifcfg-enp12s0.1
# Generated by VDSM version 4.19.28-1.el7.centos
DEVICE=enp12s0.1
VLAN=yes
BRIDGE=def
ONBOOT=yes
MTU=1500
DEFROUTE=no
NM_CONTROLLED=no
IPV6INIT=no

[root@X network-scripts]# more ifcfg-def
# Generated by VDSM version 4.19.28-1.el7.centos
DEVICE=def
TYPE=Bridge
DELAY=0
STP=off
ONBOOT=yes
MTU=1500
DEFROUTE=no
NM_CONTROLLED=no
IPV6INIT=no

then run

systemctl restart network

it didn't give errors, and ip addr shows

[root@X network-scripts]# ip addr

94: enp12s0.1@enp12s0:  mtu 1500 
qdisc noqueue master def state UP qlen 1000

link/ether 00:25:b5:00:10:5f brd ff:ff:ff:ff:ff:ff
inet6 fe80::225:b5ff:fe00:105f/64 scope link
   valid_lft forever preferred_lft forever
95: def:  mtu 1500 qdisc noqueue 
state UP qlen 1000

link/ether 00:25:b5:00:10:5f brd ff:ff:ff:ff:ff:ff
inet6 fe80::225:b5ff:fe00:105f/64 scope link
   valid_lft forever preferred_lft forever

I will try to change the name of that network to something different 
from default


Claudio

On 30/08/17 11:31, Claudio Soprano wrote:


The problem is not the host-deploy, infact the deploy finished 
without errors, if i remember good in the deploy process the network 
are not set, i need to attach each network i need manually, under 
hosts, network interface, setup host networks


and the problem is there, when i attach all the VLANs (including the 
default, that is VLAN 1) it gave error, default is not present in 
the system, if i add all the VLANS (excluding the default, that is 
VLAN 1) it is OK, but the host will not activate because missing the 
default vlan.


If i manually add the 2 configuration files (ifcfg-default and 
ifcfg-intX.1) to the host and make a systemctl network restart, it 
gave error then


ip addr reports

the vlan 1 is added, but not the Bridge to default (the default is 
missing error).


I don't know if the name "default" (for the network, VLAN 1) could 
be the problem, but i can't rename now, because i need to detach 
from the hosts, but actually i can't.


I added 3 screenshoots to show you the situation, before adding VLAN 
1 ("default" network), the error when adding it, and the info on the 
error.


Claudio

On 30/08/17 08:49, Sandro Bonazzola wrote:



2017-08-29 14:45 GMT+02:00 Claudio Soprano 
>:


Ok in this way i could install the hosts, but we got another
error on them when setup networks

we have on each hosts 2 interfaces

interface 1 is ovirtmgm

interface 2 is a TRUNK with VLANs inside.

All my old hosts are all the VLANs on interface2 included VLAN
1 (tagged).

When i setup networks for the new hosts 4.0.6 i can't include
VLAN 1, ovirt answer with

VDSM hostname.domainname command HostSetupNetworksVDS failed:
[Errno 19] default (that is VLAN 1) is not present in the system.

So thinking about upgrading to 4.1.5 we updated all the new
hosts (the old are still v4.0.6) and reinstall from beginning
but still we get the same error.

What means that error ?


Adding Marcin and Dan about this.

Claudio Soprano

On 25/08/17 15:27, Sandro Bonazzola wrote:



2017-08-25 10:43 GMT+02:00 Claudio Soprano

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found - SOLVED

[Claudio Soprano]

More investigation on this for you :)

the bug is only on CentOS v7.3 infact on previous hosts with CentOS < 
v7.3 i have no problem about the name of the network.


I discovered also that i don't need only to rename the network name, but 
also the vnic profile associated, infact with network name def, and vnic 
profile default, it still didn't work, renamed network name and vnic 
profile to def both, now works on CentOS v7.3 with no errors also from GUI.


when i create a network both network name and vnic profile are called 
the same (for my it was default).


Hope that my info can help you.

Another boring issues is that i need to detach the network to all the 
hosts to rename only, but also to change the network on all VMs that use 
it also if they are stopped, i was lucky only 2 VMs and only 4 hosts, 
but i can't imagine people have 40-50 hosts and maybe 100-200 VMs on a 
network.


Anyway now i have Centos V7.3 and V7.2 hosts both working.

Claudio

On 30/08/17 11:52, Claudio Soprano wrote:


Ok i SOLVED the problem, but i don't know if it is a BUG, the problem 
is the name default, but i had it working with ovirt v3 and v4.0pre 
for sure.


I don't know if the check has been added in the setup_host_network or not.

This is that i did

i manually added a ifcfg-def and ifcfg-enp12s0.1 file (first for the 
bridge and the second for the VLAN) to an host that don't have VLAN 1


[root@X network-scripts]# more ifcfg-enp12s0.1
# Generated by VDSM version 4.19.28-1.el7.centos
DEVICE=enp12s0.1
VLAN=yes
BRIDGE=def
ONBOOT=yes
MTU=1500
DEFROUTE=no
NM_CONTROLLED=no
IPV6INIT=no

[root@X network-scripts]# more ifcfg-def
# Generated by VDSM version 4.19.28-1.el7.centos
DEVICE=def
TYPE=Bridge
DELAY=0
STP=off
ONBOOT=yes
MTU=1500
DEFROUTE=no
NM_CONTROLLED=no
IPV6INIT=no

then run

systemctl restart network

it didn't give errors, and ip addr shows

[root@X network-scripts]# ip addr

94: enp12s0.1@enp12s0:  mtu 1500 
qdisc noqueue master def state UP qlen 1000

link/ether 00:25:b5:00:10:5f brd ff:ff:ff:ff:ff:ff
inet6 fe80::225:b5ff:fe00:105f/64 scope link
   valid_lft forever preferred_lft forever
95: def:  mtu 1500 qdisc noqueue 
state UP qlen 1000

link/ether 00:25:b5:00:10:5f brd ff:ff:ff:ff:ff:ff
inet6 fe80::225:b5ff:fe00:105f/64 scope link
   valid_lft forever preferred_lft forever

I will try to change the name of that network to something different 
from default


Claudio

On 30/08/17 11:31, Claudio Soprano wrote:


The problem is not the host-deploy, infact the deploy finished 
without errors, if i remember good in the deploy process the network 
are not set, i need to attach each network i need manually, under 
hosts, network interface, setup host networks


and the problem is there, when i attach all the VLANs (including the 
default, that is VLAN 1) it gave error, default is not present in the 
system, if i add all the VLANS (excluding the default, that is VLAN 
1) it is OK, but the host will not activate because missing the 
default vlan.


If i manually add the 2 configuration files (ifcfg-default and 
ifcfg-intX.1) to the host and make a systemctl network restart, it 
gave error then


ip addr reports

the vlan 1 is added, but not the Bridge to default (the default is 
missing error).


I don't know if the name "default" (for the network, VLAN 1) could be 
the problem, but i can't rename now, because i need to detach from 
the hosts, but actually i can't.


I added 3 screenshoots to show you the situation, before adding VLAN 
1 ("default" network), the error when adding it, and the info on the 
error.


Claudio

On 30/08/17 08:49, Sandro Bonazzola wrote:



2017-08-29 14:45 GMT+02:00 Claudio Soprano 
>:


Ok in this way i could install the hosts, but we got another
error on them when setup networks

we have on each hosts 2 interfaces

interface 1 is ovirtmgm

interface 2 is a TRUNK with VLANs inside.

All my old hosts are all the VLANs on interface2 included VLAN 1
(tagged).

When i setup networks for the new hosts 4.0.6 i can't include
VLAN 1, ovirt answer with

VDSM hostname.domainname command HostSetupNetworksVDS failed:
[Errno 19] default (that is VLAN 1) is not present in the system.

So thinking about upgrading to 4.1.5 we updated all the new
hosts (the old are still v4.0.6) and reinstall from beginning
but still we get the same error.

What means that error ?


Adding Marcin and Dan about this.

Claudio Soprano

On 25/08/17 15:27, Sandro Bonazzola wrote:



2017-08-25 10:43 GMT+02:00 Claudio Soprano
>:

Hi all,

we are installing new nodes on a Ovirt 4.0.6 Cluster, the
new nodes were installed from a Minimal 1707 iso image,
Centos v7.3.1611


Any reason for 

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found - SOLVED

[Claudio Soprano]

Ok i SOLVED the problem, but i don't know if it is a BUG, the problem is 
the name default, but i had it working with ovirt v3 and v4.0pre for sure.


I don't know if the check has been added in the setup_host_network or not.

This is that i did

i manually added a ifcfg-def and ifcfg-enp12s0.1 file (first for the 
bridge and the second for the VLAN) to an host that don't have VLAN 1


[root@X network-scripts]# more ifcfg-enp12s0.1
# Generated by VDSM version 4.19.28-1.el7.centos
DEVICE=enp12s0.1
VLAN=yes
BRIDGE=def
ONBOOT=yes
MTU=1500
DEFROUTE=no
NM_CONTROLLED=no
IPV6INIT=no

[root@X network-scripts]# more ifcfg-def
# Generated by VDSM version 4.19.28-1.el7.centos
DEVICE=def
TYPE=Bridge
DELAY=0
STP=off
ONBOOT=yes
MTU=1500
DEFROUTE=no
NM_CONTROLLED=no
IPV6INIT=no

then run

systemctl restart network

it didn't give errors, and ip addr shows

[root@X network-scripts]# ip addr

94: enp12s0.1@enp12s0:  mtu 1500 qdisc 
noqueue master def state UP qlen 1000

link/ether 00:25:b5:00:10:5f brd ff:ff:ff:ff:ff:ff
inet6 fe80::225:b5ff:fe00:105f/64 scope link
   valid_lft forever preferred_lft forever
95: def:  mtu 1500 qdisc noqueue state 
UP qlen 1000

link/ether 00:25:b5:00:10:5f brd ff:ff:ff:ff:ff:ff
inet6 fe80::225:b5ff:fe00:105f/64 scope link
   valid_lft forever preferred_lft forever

I will try to change the name of that network to something different 
from default


Claudio

On 30/08/17 11:31, Claudio Soprano wrote:


The problem is not the host-deploy, infact the deploy finished without 
errors, if i remember good in the deploy process the network are not 
set, i need to attach each network i need manually, under hosts, 
network interface, setup host networks


and the problem is there, when i attach all the VLANs (including the 
default, that is VLAN 1) it gave error, default is not present in the 
system, if i add all the VLANS (excluding the default, that is VLAN 1) 
it is OK, but the host will not activate because missing the default vlan.


If i manually add the 2 configuration files (ifcfg-default and 
ifcfg-intX.1) to the host and make a systemctl network restart, it 
gave error then


ip addr reports

the vlan 1 is added, but not the Bridge to default (the default is 
missing error).


I don't know if the name "default" (for the network, VLAN 1) could be 
the problem, but i can't rename now, because i need to detach from the 
hosts, but actually i can't.


I added 3 screenshoots to show you the situation, before adding VLAN 1 
("default" network), the error when adding it, and the info on the error.


Claudio

On 30/08/17 08:49, Sandro Bonazzola wrote:



2017-08-29 14:45 GMT+02:00 Claudio Soprano 
>:


Ok in this way i could install the hosts, but we got another
error on them when setup networks

we have on each hosts 2 interfaces

interface 1 is ovirtmgm

interface 2 is a TRUNK with VLANs inside.

All my old hosts are all the VLANs on interface2 included VLAN 1
(tagged).

When i setup networks for the new hosts 4.0.6 i can't include
VLAN 1, ovirt answer with

VDSM hostname.domainname command HostSetupNetworksVDS failed:
[Errno 19] default (that is VLAN 1) is not present in the system.

So thinking about upgrading to 4.1.5 we updated all the new hosts
(the old are still v4.0.6) and reinstall from beginning but still
we get the same error.

What means that error ?


Adding Marcin and Dan about this.

Claudio Soprano

On 25/08/17 15:27, Sandro Bonazzola wrote:



2017-08-25 10:43 GMT+02:00 Claudio Soprano
>:

Hi all,

we are installing new nodes on a Ovirt 4.0.6 Cluster, the
new nodes were installed from a Minimal 1707 iso image,
Centos v7.3.1611


Any reason for keeping 4.0.6 hosts? oVirt 4.0 has reached End Of
Life back in January[1]
And you can install 4.1 hosts in 4.0 cluster compatibility mode.

That said, you may want to "yum install centos-release-opstools"
which should provide you the missing collectd packages by
providing the same repos already included in ovirt-release41 rpm.

[1]
http://lists.ovirt.org/pipermail/announce/2017-January/000308.html



in the log of host-deploy (on the manager) it indicates that
collectd packages was not found.

I found this bug in the ovirt 4.1.0 version where it tells
to disable the EPEL repository, in another tread on redhat
someone told to add

excludepkgs=collectd*

always in the epel repository

I checked all my repository and i found epel reference only in

[root@ yum.repos.d]# grep epel *
ovirt-4.0-dependencies.repo:[ovirt-4.0-epel]
  

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found

[Sandro Bonazzola]

2017-08-29 14:45 GMT+02:00 Claudio Soprano :

> Ok in this way i could install the hosts, but we got another error on them
> when setup networks
>
> we have on each hosts 2 interfaces
>
> interface 1 is ovirtmgm
>
> interface 2 is a TRUNK with VLANs inside.
>
> All my old hosts are all the VLANs on interface2 included VLAN 1 (tagged).
>
> When i setup networks for the new hosts 4.0.6 i can't include VLAN 1,
> ovirt answer with
>
> VDSM hostname.domainname command HostSetupNetworksVDS failed: [Errno 19]
> default (that is VLAN 1) is not present in the system.
>
> So thinking about upgrading to 4.1.5 we updated all the new hosts (the old
> are still v4.0.6) and reinstall from beginning but still we get the same
> error.
>
> What means that error ?
>

Adding Marcin and Dan about this.



> Claudio Soprano
> On 25/08/17 15:27, Sandro Bonazzola wrote:
>
>
>
> 2017-08-25 10:43 GMT+02:00 Claudio Soprano :
>
>> Hi all,
>>
>> we are installing new nodes on a Ovirt 4.0.6 Cluster, the new nodes were
>> installed from a Minimal 1707 iso image, Centos v7.3.1611
>>
>
> Any reason for keeping 4.0.6 hosts? oVirt 4.0 has reached End Of Life back
> in January[1]
> And you can install 4.1 hosts in 4.0 cluster compatibility mode.
>
> That said, you may want to "yum install centos-release-opstools" which
> should provide you the missing collectd packages by providing the same
> repos already included in ovirt-release41 rpm.
>
> [1] http://lists.ovirt.org/pipermail/announce/2017-January/000308.html
>
>
>
>
>> in the log of host-deploy (on the manager) it indicates that collectd
>> packages was not found.
>>
>> I found this bug in the ovirt 4.1.0 version where it tells to disable the
>> EPEL repository, in another tread on redhat someone told to add
>>
>> excludepkgs=collectd*
>>
>> always in the epel repository
>>
>> I checked all my repository and i found epel reference only in
>>
>> [root@ yum.repos.d]# grep epel *
>> ovirt-4.0-dependencies.repo:[ovirt-4.0-epel]
>> ovirt-4.0-dependencies.repo:#baseurl=http://download.fedorap
>> roject.org/pub/epel/7/$basearch
>> ovirt-4.0-dependencies.repo:mirrorlist=https://mirrors.fedor
>> aproject.org/metalink?repo=epel-7=$basearch
>> ovirt-4.0-dependencies.repo:includepkgs=epel-release,python-
>> uinput,puppet,python-lockfile,python-cpopen,python-ordereddi
>> ct,python-pthreading,python-inotify,python-argparse,novnc,
>> python-ply,python-kitchen,python-daemon,python-
>> websockify,livecd-tools,spice-html5,mom,python-IPy,python-
>> ioprocess,ioprocess,safelease,python-paramiko,python2-
>> paramiko,python2-crypto,libtomcrypt,libtommath,python-
>> cheetah,python-ecdsa,python2-ecdsa,python-markdown,rubygem-
>> rgen,ovirt-guest-agent*,userspace-rcu,protobuf-java,objenesis,python34*
>> ovirt-4.0-dependencies.repo:gpgkey=https://dl.fedoraproject.
>> org/pub/epel/RPM-GPG-KEY-EPEL-7
>> ovirt-4.0-dependencies.repo:[ovirt-4.0-patternfly1-noarch-epel]
>> ovirt-4.0-dependencies.repo:baseurl=http://copr-be.cloud.fed
>> oraproject.org/results/patternfly/patternfly1/epel-7-$basearch/
>>
>> so before includespkgs i added the excludepkgs line
>>
>> removed the host, reinserted and still again collectd not found
>>
>> so i added the excludepkgs line in all the ovirt sections
>>
>> removed the host, reinserted and still again collectd not found
>>
>> I removed on the manager also the file 
>> /var/cache/ovirt-engine/ovirt-host-deploy.tar
>> but still the same error
>>
>> these are the ovirt-repos files with the excludepkgs line added
>>
>> [ovirt-4.0]
>> name=Latest oVirt 4.0 Release
>> #baseurl=http://resources.ovirt.org/pub/ovirt-4.0/rpm/el$releasever/
>> mirrorlist=http://resources.ovirt.org/pub/yum-repo/mirrorlis
>> t-ovirt-4.0-el$releasever
>> enabled=1
>>
>> excludepkgs=collectd*
>>
>> skip_if_unavailable=1
>> gpgcheck=1
>> gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-ovirt-4.0
>>
>> [ovirt-4.0-epel]
>> name=Extra Packages for Enterprise Linux 7 - $basearch
>> #baseurl=http://download.fedoraproject.org/pub/epel/7/$basearch
>> mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=
>> epel-7=$basearch
>> failovermethod=priority
>> enabled=1
>>
>> excludepkgs=collectd*
>>
>> includepkgs=epel-release,python-uinput,puppet,python-lockfil
>> e,python-cpopen,python-ordereddict,python-pthreading,python-
>> inotify,python-argparse,novnc,python-ply,python-kitchen,
>> python-daemon,python-websockify,livec
>> d-tools,spice-html5,mom,python-IPy,python-ioprocess,ioproces
>> s,safelease,python-paramiko,python2-paramiko,python2-
>> crypto,libtomcrypt,libtommath,python-cheetah,python-ecdsa,
>> python2-ecdsa,python-markdown,rubygem-rgen
>> ,ovirt-guest-agent*,userspace-rcu,protobuf-java,objenesis,python34*
>> gpgcheck=1
>> gpgkey=https://dl.fedoraproject.org/pub/epel/RPM-GPG-KEY-EPEL-7
>>
>> [ovirt-4.0-centos-gluster37]
>> name=CentOS-7 - Gluster 3.7
>> baseurl=http://mirror.centos.org/centos/7/storage/$basearch/gluster-3.7/
>>
>> excludepkgs=collectd*
>>
>> 

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found

[Claudio Soprano]

Ok in this way i could install the hosts, but we got another error on 
them when setup networks


we have on each hosts 2 interfaces

interface 1 is ovirtmgm

interface 2 is a TRUNK with VLANs inside.

All my old hosts are all the VLANs on interface2 included VLAN 1 (tagged).

When i setup networks for the new hosts 4.0.6 i can't include VLAN 1, 
ovirt answer with


VDSM hostname.domainname command HostSetupNetworksVDS failed: [Errno 19] 
default (that is VLAN 1) is not present in the system.


So thinking about upgrading to 4.1.5 we updated all the new hosts (the 
old are still v4.0.6) and reinstall from beginning but still we get the 
same error.


What means that error ?

Claudio Soprano

On 25/08/17 15:27, Sandro Bonazzola wrote:



2017-08-25 10:43 GMT+02:00 Claudio Soprano 
>:


Hi all,

we are installing new nodes on a Ovirt 4.0.6 Cluster, the new
nodes were installed from a Minimal 1707 iso image, Centos v7.3.1611


Any reason for keeping 4.0.6 hosts? oVirt 4.0 has reached End Of Life 
back in January[1]

And you can install 4.1 hosts in 4.0 cluster compatibility mode.

That said, you may want to "yum install centos-release-opstools" which 
should provide you the missing collectd packages by providing the same 
repos already included in ovirt-release41 rpm.


[1] http://lists.ovirt.org/pipermail/announce/2017-January/000308.html


in the log of host-deploy (on the manager) it indicates that
collectd packages was not found.

I found this bug in the ovirt 4.1.0 version where it tells to
disable the EPEL repository, in another tread on redhat someone
told to add

excludepkgs=collectd*

always in the epel repository

I checked all my repository and i found epel reference only in

[root@ yum.repos.d]# grep epel *
ovirt-4.0-dependencies.repo:[ovirt-4.0-epel]

ovirt-4.0-dependencies.repo:#baseurl=http://download.fedoraproject.org/pub/epel/7/$basearch


ovirt-4.0-dependencies.repo:mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-7=$basearch


ovirt-4.0-dependencies.repo:includepkgs=epel-release,python-uinput,puppet,python-lockfile,python-cpopen,python-ordereddict,python-pthreading,python-inotify,python-argparse,novnc,python-ply,python-kitchen,python-daemon,python-websockify,livecd-tools,spice-html5,mom,python-IPy,python-ioprocess,ioprocess,safelease,python-paramiko,python2-paramiko,python2-crypto,libtomcrypt,libtommath,python-cheetah,python-ecdsa,python2-ecdsa,python-markdown,rubygem-rgen,ovirt-guest-agent*,userspace-rcu,protobuf-java,objenesis,python34*

ovirt-4.0-dependencies.repo:gpgkey=https://dl.fedoraproject.org/pub/epel/RPM-GPG-KEY-EPEL-7

ovirt-4.0-dependencies.repo:[ovirt-4.0-patternfly1-noarch-epel]

ovirt-4.0-dependencies.repo:baseurl=http://copr-be.cloud.fedoraproject.org/results/patternfly/patternfly1/epel-7-$basearch/



so before includespkgs i added the excludepkgs line

removed the host, reinserted and still again collectd not found

so i added the excludepkgs line in all the ovirt sections

removed the host, reinserted and still again collectd not found

I removed on the manager also the file
/var/cache/ovirt-engine/ovirt-host-deploy.tar but still the same error

these are the ovirt-repos files with the excludepkgs line added

[ovirt-4.0]
name=Latest oVirt 4.0 Release
#baseurl=http://resources.ovirt.org/pub/ovirt-4.0/rpm/el$releasever/


mirrorlist=http://resources.ovirt.org/pub/yum-repo/mirrorlist-ovirt-4.0-el$releasever

enabled=1

excludepkgs=collectd*

skip_if_unavailable=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-ovirt-4.0

[ovirt-4.0-epel]
name=Extra Packages for Enterprise Linux 7 - $basearch
#baseurl=http://download.fedoraproject.org/pub/epel/7/$basearch


mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-7=$basearch

failovermethod=priority
enabled=1

excludepkgs=collectd*


includepkgs=epel-release,python-uinput,puppet,python-lockfile,python-cpopen,python-ordereddict,python-pthreading,python-inotify,python-argparse,novnc,python-ply,python-kitchen,python-daemon,python-websockify,livec

Re: [ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found

[Sandro Bonazzola]

2017-08-25 10:43 GMT+02:00 Claudio Soprano :

> Hi all,
>
> we are installing new nodes on a Ovirt 4.0.6 Cluster, the new nodes were
> installed from a Minimal 1707 iso image, Centos v7.3.1611
>

Any reason for keeping 4.0.6 hosts? oVirt 4.0 has reached End Of Life back
in January[1]
And you can install 4.1 hosts in 4.0 cluster compatibility mode.

That said, you may want to "yum install centos-release-opstools" which
should provide you the missing collectd packages by providing the same
repos already included in ovirt-release41 rpm.

[1] http://lists.ovirt.org/pipermail/announce/2017-January/000308.html




> in the log of host-deploy (on the manager) it indicates that collectd
> packages was not found.
>
> I found this bug in the ovirt 4.1.0 version where it tells to disable the
> EPEL repository, in another tread on redhat someone told to add
>
> excludepkgs=collectd*
>
> always in the epel repository
>
> I checked all my repository and i found epel reference only in
>
> [root@ yum.repos.d]# grep epel *
> ovirt-4.0-dependencies.repo:[ovirt-4.0-epel]
> ovirt-4.0-dependencies.repo:#baseurl=http://download.
> fedoraproject.org/pub/epel/7/$basearch
> ovirt-4.0-dependencies.repo:mirrorlist=https://mirrors.
> fedoraproject.org/metalink?repo=epel-7=$basearch
> ovirt-4.0-dependencies.repo:includepkgs=epel-release,
> python-uinput,puppet,python-lockfile,python-cpopen,python-
> ordereddict,python-pthreading,python-inotify,python-
> argparse,novnc,python-ply,python-kitchen,python-daemon,
> python-websockify,livecd-tools,spice-html5,mom,python-
> IPy,python-ioprocess,ioprocess,safelease,python-paramiko,python2-paramiko,
> python2-crypto,libtomcrypt,libtommath,python-cheetah,
> python-ecdsa,python2-ecdsa,python-markdown,rubygem-rgen,
> ovirt-guest-agent*,userspace-rcu,protobuf-java,objenesis,python34*
> ovirt-4.0-dependencies.repo:gpgkey=https://dl.fedoraproject.org/pub/epel/
> RPM-GPG-KEY-EPEL-7
> ovirt-4.0-dependencies.repo:[ovirt-4.0-patternfly1-noarch-epel]
> ovirt-4.0-dependencies.repo:baseurl=http://copr-be.cloud.
> fedoraproject.org/results/patternfly/patternfly1/epel-7-$basearch/
>
> so before includespkgs i added the excludepkgs line
>
> removed the host, reinserted and still again collectd not found
>
> so i added the excludepkgs line in all the ovirt sections
>
> removed the host, reinserted and still again collectd not found
>
> I removed on the manager also the file 
> /var/cache/ovirt-engine/ovirt-host-deploy.tar
> but still the same error
>
> these are the ovirt-repos files with the excludepkgs line added
>
> [ovirt-4.0]
> name=Latest oVirt 4.0 Release
> #baseurl=http://resources.ovirt.org/pub/ovirt-4.0/rpm/el$releasever/
> mirrorlist=http://resources.ovirt.org/pub/yum-repo/
> mirrorlist-ovirt-4.0-el$releasever
> enabled=1
>
> excludepkgs=collectd*
>
> skip_if_unavailable=1
> gpgcheck=1
> gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-ovirt-4.0
>
> [ovirt-4.0-epel]
> name=Extra Packages for Enterprise Linux 7 - $basearch
> #baseurl=http://download.fedoraproject.org/pub/epel/7/$basearch
> mirrorlist=https://mirrors.fedoraproject.org/metalink?
> repo=epel-7=$basearch
> failovermethod=priority
> enabled=1
>
> excludepkgs=collectd*
>
> includepkgs=epel-release,python-uinput,puppet,python-
> lockfile,python-cpopen,python-ordereddict,python-pthreading,
> python-inotify,python-argparse,novnc,python-ply,
> python-kitchen,python-daemon,python-websockify,livec
> d-tools,spice-html5,mom,python-IPy,python-ioprocess,
> ioprocess,safelease,python-paramiko,python2-paramiko,
> python2-crypto,libtomcrypt,libtommath,python-cheetah,
> python-ecdsa,python2-ecdsa,python-markdown,rubygem-rgen
> ,ovirt-guest-agent*,userspace-rcu,protobuf-java,objenesis,python34*
> gpgcheck=1
> gpgkey=https://dl.fedoraproject.org/pub/epel/RPM-GPG-KEY-EPEL-7
>
> [ovirt-4.0-centos-gluster37]
> name=CentOS-7 - Gluster 3.7
> baseurl=http://mirror.centos.org/centos/7/storage/$basearch/gluster-3.7/
>
> excludepkgs=collectd*
>
> gpgcheck=1
> enabled=1
> gpgkey=https://raw.githubusercontent.com/CentOS-
> Storage-SIG/centos-release-storage-common/master/RPM-GPG-
> KEY-CentOS-SIG-Storage
>
> [ovirt-4.0-patternfly1-noarch-epel]
> name=Copr repo for patternfly1 owned by patternfly
> baseurl=http://copr-be.cloud.fedoraproject.org/results/
> patternfly/patternfly1/epel-7-$basearch/
> enabled=1
>
> excludepkgs=collectd*
>
> skip_if_unavailable=1
> gpgcheck=0
>
> [virtio-win-stable]
> name=virtio-win builds roughly matching what was shipped in latest RHEL
> baseurl=http://fedorapeople.org/groups/virt/virtio-win/repo/stable
> enabled=1
>
> excludepkgs=collectd*
>
> skip_if_unavailable=1
> gpgcheck=0
>
> [centos-ovirt40-release]
> name=CentOS-7 - oVirt 4.0
> baseurl=http://mirror.centos.org/centos/7/virt/$basearch/ovirt-4.0/
> gpgcheck=0
> enabled=1
>
> excludepkgs=collectd*
>
> I add also the part of log on the manager relatives to host-deploy where
> the error is logged
>
> [root@ovcmgr host-deploy]# more 
> 

[ovirt-users] Centos 7.3 ovirt 4.0.6 Can't add host to cluster collectd or collectd-disk not found

[Claudio Soprano]

Hi all,

we are installing new nodes on a Ovirt 4.0.6 Cluster, the new nodes were 
installed from a Minimal 1707 iso image, Centos v7.3.1611


in the log of host-deploy (on the manager) it indicates that collectd 
packages was not found.


I found this bug in the ovirt 4.1.0 version where it tells to disable 
the EPEL repository, in another tread on redhat someone told to add


excludepkgs=collectd*

always in the epel repository

I checked all my repository and i found epel reference only in

[root@ yum.repos.d]# grep epel *
ovirt-4.0-dependencies.repo:[ovirt-4.0-epel]
ovirt-4.0-dependencies.repo:#baseurl=http://download.fedoraproject.org/pub/epel/7/$basearch
ovirt-4.0-dependencies.repo:mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-7=$basearch
ovirt-4.0-dependencies.repo:includepkgs=epel-release,python-uinput,puppet,python-lockfile,python-cpopen,python-ordereddict,python-pthreading,python-inotify,python-argparse,novnc,python-ply,python-kitchen,python-daemon,python-websockify,livecd-tools,spice-html5,mom,python-IPy,python-ioprocess,ioprocess,safelease,python-paramiko,python2-paramiko,python2-crypto,libtomcrypt,libtommath,python-cheetah,python-ecdsa,python2-ecdsa,python-markdown,rubygem-rgen,ovirt-guest-agent*,userspace-rcu,protobuf-java,objenesis,python34*
ovirt-4.0-dependencies.repo:gpgkey=https://dl.fedoraproject.org/pub/epel/RPM-GPG-KEY-EPEL-7
ovirt-4.0-dependencies.repo:[ovirt-4.0-patternfly1-noarch-epel]
ovirt-4.0-dependencies.repo:baseurl=http://copr-be.cloud.fedoraproject.org/results/patternfly/patternfly1/epel-7-$basearch/

so before includespkgs i added the excludepkgs line

removed the host, reinserted and still again collectd not found

so i added the excludepkgs line in all the ovirt sections

removed the host, reinserted and still again collectd not found

I removed on the manager also the file 
/var/cache/ovirt-engine/ovirt-host-deploy.tar but still the same error


these are the ovirt-repos files with the excludepkgs line added

[ovirt-4.0]
name=Latest oVirt 4.0 Release
#baseurl=http://resources.ovirt.org/pub/ovirt-4.0/rpm/el$releasever/
mirrorlist=http://resources.ovirt.org/pub/yum-repo/mirrorlist-ovirt-4.0-el$releasever
enabled=1

excludepkgs=collectd*

skip_if_unavailable=1
gpgcheck=1
gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-ovirt-4.0

[ovirt-4.0-epel]
name=Extra Packages for Enterprise Linux 7 - $basearch
#baseurl=http://download.fedoraproject.org/pub/epel/7/$basearch
mirrorlist=https://mirrors.fedoraproject.org/metalink?repo=epel-7=$basearch
failovermethod=priority
enabled=1

excludepkgs=collectd*

includepkgs=epel-release,python-uinput,puppet,python-lockfile,python-cpopen,python-ordereddict,python-pthreading,python-inotify,python-argparse,novnc,python-ply,python-kitchen,python-daemon,python-websockify,livec
d-tools,spice-html5,mom,python-IPy,python-ioprocess,ioprocess,safelease,python-paramiko,python2-paramiko,python2-crypto,libtomcrypt,libtommath,python-cheetah,python-ecdsa,python2-ecdsa,python-markdown,rubygem-rgen
,ovirt-guest-agent*,userspace-rcu,protobuf-java,objenesis,python34*
gpgcheck=1
gpgkey=https://dl.fedoraproject.org/pub/epel/RPM-GPG-KEY-EPEL-7

[ovirt-4.0-centos-gluster37]
name=CentOS-7 - Gluster 3.7
baseurl=http://mirror.centos.org/centos/7/storage/$basearch/gluster-3.7/

excludepkgs=collectd*

gpgcheck=1
enabled=1
gpgkey=https://raw.githubusercontent.com/CentOS-Storage-SIG/centos-release-storage-common/master/RPM-GPG-KEY-CentOS-SIG-Storage

[ovirt-4.0-patternfly1-noarch-epel]
name=Copr repo for patternfly1 owned by patternfly
baseurl=http://copr-be.cloud.fedoraproject.org/results/patternfly/patternfly1/epel-7-$basearch/
enabled=1

excludepkgs=collectd*

skip_if_unavailable=1
gpgcheck=0

[virtio-win-stable]
name=virtio-win builds roughly matching what was shipped in latest RHEL
baseurl=http://fedorapeople.org/groups/virt/virtio-win/repo/stable
enabled=1

excludepkgs=collectd*

skip_if_unavailable=1
gpgcheck=0

[centos-ovirt40-release]
name=CentOS-7 - oVirt 4.0
baseurl=http://mirror.centos.org/centos/7/virt/$basearch/ovirt-4.0/
gpgcheck=0
enabled=1

excludepkgs=collectd*

I add also the part of log on the manager relatives to host-deploy where 
the error is logged


[root@ovcmgr host-deploy]# more 
ovirt-host-deploy-20170825092719-xx-53018ceb.log  | grep collectd
2017-08-25 09:26:49 DEBUG otopi.context context.dumpSequence:744 
METHOD 
otopi.plugins.ovirt_host_deploy.collectd.packages.Plugin._packages (None)
2017-08-25 09:26:50 DEBUG otopi.context context.dumpSequence:744 
METHOD 
otopi.plugins.ovirt_host_deploy.collectd.packages.Plugin._packages (None)
2017-08-25 09:27:19 DEBUG otopi.context context._executeMethod:128 Stage 
packages METHOD 
otopi.plugins.ovirt_host_deploy.collectd.packages.Plugin._packages
2017-08-25 09:27:19 DEBUG otopi.plugins.otopi.packagers.yumpackager 
yumpackager.verbose:76 Yum queue package collectd for install/update
2017-08-25 09:27:19 ERROR otopi.plugins.otopi.packagers.yumpackager 
yumpackager.error:85 Yum