Ovirt 4.2.8.. Possible bug?
by matteo fedeli
Hi considering that the deploy with 4.2.7.8 failed I try to reinstall ovirt to version 4.2.8 and there are appened two strange things.
During the volume step if i choose jbod mode in the deploy conf remain raid6 type... Why? To solve I have only tried to editing manually the file at line about volume type and the deploy stuck on creating physical volume...
this is my conf file: (I used 3 HDDs 500GB each, node,engine + vmstore and data)
#gdeploy configuration generated by cockpit-gluster plugin
[hosts]
kansas.planet.bn
germany.planet.bn
singapore.planet.bn
[script1:kansas.planet.bn]
action=execute
ignore_script_errors=no
file=/usr/share/gdeploy/scripts/grafton-sanity-check.sh -d sdb -h kansas.planet.bn, germany.planet.bn, singapore.planet.bn
[script1:germany.planet.bn]
action=execute
ignore_script_errors=no
file=/usr/share/gdeploy/scripts/grafton-sanity-check.sh -d sdb -h kansas.planet.bn, germany.planet.bn, singapore.planet.bn
[script1:singapore.planet.bn]
action=execute
ignore_script_errors=no
file=/usr/share/gdeploy/scripts/grafton-sanity-check.sh -d sdb -h kansas.planet.bn, germany.planet.bn, singapore.planet.bn
[disktype]
jbod
[diskcount]
12
[stripesize]
256
[service1]
action=enable
service=chronyd
[service2]
action=restart
service=chronyd
[shell2]
action=execute
command=vdsm-tool configure --force
[script3]
action=execute
file=/usr/share/gdeploy/scripts/blacklist_all_disks.sh
ignore_script_errors=no
[pv1:kansas.planet.bn]
action=create
devices=sdb
ignore_pv_errors=no
[pv1:germany.planet.bn]
action=create
devices=sdb
ignore_pv_errors=no
[pv1:singapore.planet.bn]
action=create
devices=sdb
ignore_pv_errors=no
[vg1:kansas.planet.bn]
action=create
vgname=gluster_vg_sdb
pvname=sdb
ignore_vg_errors=no
[vg1:germany.planet.bn]
action=create
vgname=gluster_vg_sdb
pvname=sdb
ignore_vg_errors=no
[vg1:singapore.planet.bn]
action=create
vgname=gluster_vg_sdb
pvname=sdb
ignore_vg_errors=no
[lv1:kansas.planet.bn]
action=create
poolname=gluster_thinpool_sdb
ignore_lv_errors=no
vgname=gluster_vg_sdb
lvtype=thinpool
size=1005GB
poolmetadatasize=5GB
[lv2:germany.planet.bn]
action=create
poolname=gluster_thinpool_sdb
ignore_lv_errors=no
vgname=gluster_vg_sdb
lvtype=thinpool
size=1005GB
poolmetadatasize=5GB
[lv3:singapore.planet.bn]
action=create
poolname=gluster_thinpool_sdb
ignore_lv_errors=no
vgname=gluster_vg_sdb
lvtype=thinpool
size=1005GB
poolmetadatasize=5GB
[lv4:kansas.planet.bn]
action=create
lvname=gluster_lv_engine
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/engine
size=100GB
lvtype=thick
[lv5:kansas.planet.bn]
action=create
lvname=gluster_lv_data
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/data
lvtype=thinlv
poolname=gluster_thinpool_sdb
virtualsize=500GB
[lv6:kansas.planet.bn]
action=create
lvname=gluster_lv_vmstore
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/vmstore
lvtype=thinlv
poolname=gluster_thinpool_sdb
virtualsize=500GB
[lv7:germany.planet.bn]
action=create
lvname=gluster_lv_engine
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/engine
size=100GB
lvtype=thick
[lv8:germany.planet.bn]
action=create
lvname=gluster_lv_data
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/data
lvtype=thinlv
poolname=gluster_thinpool_sdb
virtualsize=500GB
[lv9:germany.planet.bn]
action=create
lvname=gluster_lv_vmstore
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/vmstore
lvtype=thinlv
poolname=gluster_thinpool_sdb
virtualsize=500GB
[lv10:singapore.planet.bn]
action=create
lvname=gluster_lv_engine
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/engine
size=100GB
lvtype=thick
[lv11:singapore.planet.bn]
action=create
lvname=gluster_lv_data
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/data
lvtype=thinlv
poolname=gluster_thinpool_sdb
virtualsize=500GB
[lv12:singapore.planet.bn]
action=create
lvname=gluster_lv_vmstore
ignore_lv_errors=no
vgname=gluster_vg_sdb
mount=/gluster_bricks/vmstore
lvtype=thinlv
poolname=gluster_thinpool_sdb
virtualsize=500GB
[selinux]
yes
[service3]
action=restart
service=glusterd
slice_setup=yes
[firewalld]
action=add
ports=111/tcp,2049/tcp,54321/tcp,5900/tcp,5900-6923/tcp,5666/tcp,16514/tcp,54322/tcp
services=glusterfs
[script2]
action=execute
file=/usr/share/gdeploy/scripts/disable-gluster-hooks.sh
[shell3]
action=execute
command=usermod -a -G gluster qemu
[volume1]
action=create
volname=engine
transport=tcp
replica=yes
replica_count=3
key=group,storage.owner-uid,storage.owner-gid,network.ping-timeout,performance.strict-o-direct,network.remote-dio,cluster.granular-entry-heal
value=virt,36,36,30,on,off,enable
brick_dirs=kansas.planet.bn:/gluster_bricks/engine/engine,germany.planet.bn:/gluster_bricks/engine/engine,singapore.planet.bn:/gluster_bricks/engine/engine
ignore_volume_errors=no
[volume2]
action=create
volname=data
transport=tcp
replica=yes
replica_count=3
key=group,storage.owner-uid,storage.owner-gid,network.ping-timeout,performance.strict-o-direct,network.remote-dio,cluster.granular-entry-heal
value=virt,36,36,30,on,off,enable
brick_dirs=kansas.planet.bn:/gluster_bricks/data/data,germany.planet.bn:/gluster_bricks/data/data,singapore.planet.bn:/gluster_bricks/data/data
ignore_volume_errors=no
[volume3]
action=create
volname=vmstore
transport=tcp
replica=yes
replica_count=3
key=group,storage.owner-uid,storage.owner-gid,network.ping-timeout,performance.strict-o-direct,network.remote-dio,cluster.granular-entry-heal
value=virt,36,36,30,on,off,enable
brick_dirs=kansas.planet.bn:/gluster_bricks/vmstore/vmstore,germany.planet.bn:/gluster_bricks/vmstore/vmstore,singapore.planet.bn:/gluster_bricks/vmstore/vmstore
ignore_volume_errors=no
1 minute
Fencing : SSL or not?
by Nicolas Ecarnot
Hello,
As fence_idrac has never worked for us, and as fence_ipmilan has worked
nicely since years, we are using fence_ipmilan with the lanplus=1 option
and we're happy with it.
We upgraded to 4.3.0.4 and we're witnessing that we cannot fence our
hosts anymore :
2019-02-18 09:42:08,678+01 ERROR
[org.ovirt.engine.core.bll.pm.FenceProxyLocator] (default task-11)
[2f78ed99-6703-4d92-b7cb-948c2d24b623] Can not run fence action on host
'xxxxx', no suitable proxy host was found.
and on the SPM :
fence_ipmilan: Failed: Unable to obtain correct plug status or plug is
not available
I found the suggested workaround here :
https://access.redhat.com/solutions/3349841
but no combination of
- lanplus={0,1}
- -z
- ssl=={0,1}
lead to no solution.
The package version is the same as what's described in the KB :
fence-agents-rhevm-4.2.1-11.el7_6.7.x86_64
What should I test now?
Thank you.
--
Nicolas ECARNOT
59 minutes
Quota Actual Consumption
by Andrey Rusakov
Hi,
Is it possible to configure oVirt to work Quota with actual vCPU Consumption.
Currently i got 1 VM with 2 vCPU and 8Gb.
In VM stats -
0% of CPU in use
0% of RAM in Use
(it is fresh Linux instalaltion)
In Quota consumption i can see
40% of CPU (vCPU quota = 5)
27% of RAM (Memory quota 30Gb)
Is it possible to adjust Quata management to work with Actual workload, instead of VM Configuration?
P.S. Seems like it is working for Storage, as i can see 0% Consumption in Quota (1Gb of thick disk with 20Gb Max and 200Gb in Quota)
1 hour, 35 minutes
Installing oVirt on Physical machine V4.3
by emmanualvnebu1＠gmail.com
Hi Guys,
I am trying to install oVirt node on a physical machine.
Once selected the option to install oVirt 4.3 it give a screen "dracut-initqueue time out And then get into emergency mode & gives dracut cmd line.
if I exit the command line, it continue boot into the installation window.
However it gives a message "Kickstart file /run/install/ks.config is missing" (Pane is dead).
Tried the same on vmware, it boot fine & am able to install it.
Can someone tell me how could I fix this?
3 hours, 10 minutes
Host choice when migrating VMs
by Nicolas Ecarnot
Hello,
I'm almost sure the following is useless as I think I know how it's
working, but as I'm preparing a major change in our infrastructure, I'd
rather be sure and not mess up. And also to be sure.
(Just to be sure)
For some reasons, and for the first time in our infra., one of our new
DC will temporary include heterogeneous hosts : some networks will be
available only on parts of them.
Please may someone confirm me that with every load balancing / VM
startup / VM migration / host choice, oVirt will smartly choose the
available host equipped with the adequate networks?
--
Nicolas ECARNOT
5 hours, 58 minutes
Re: Ovirt Glusterfs
by Strahil
I have done some testing and it seems that storhaug + ctdb + nfs-ganesha is showing decent performance in a 3 node hyperconverged setup.
Fuse mounts are hitting some kind of limit when mounting gluster -3.12.15 volumes.
Best Regards,
Strahil Nikolov
6 hours, 12 minutes
Novnc issues are not available to regular users
by xilazz＠126.com
Hi, I have a question. When I use ordinary users to log in the management platform, the VNC of the virtual machine can only be logged in through the virtviewer tool, and there is no novnc option.
8 hours, 51 minutes
