Exercise 3.2 - Add compute host
Hello,
I have problem with exercise 3.2. I log in to compute-node, run apt-get update and then install vim and git forllowing:
git clone https://git.openstack.org/openstack-dev/devstack
I then edit local.conf and copy the values from Lab Notes. I have:
[...]
HOST_IP=192.168.97.2 # <<-- compute-node IP address
FIRST_HOST=192.168.97.1 # <<-- devstack-cc IP address
[...]
I run ./devstack.sh on compute node and it completes successfully. When I log in to the devstack-cc node first as user developer1, add an instance, and then log out and log in as admin, I don't see seconnd hypervisor nor compute node anywhere.
What would be best way to find out if compute-node ever connects with first-node (devstack-cc)?
The IPs look correct but I don't see any connections.
Thank you!
Comments
I'm in the learning process myself so I hope I'm not misleading you, forgive me if I am ;
Based on other topics in this blog, it seems that they are recommending that we use a more stable devstack version rather than the one mentioned in our lab documents :
for cloning, the suggestion is to use :
git clone https://git.openstack.org/openstack-dev/devstack -b stable/liberty
for local.conf file, I suggest you remove or comment out the following line :
FIXED_RANGE=10.10...
This is what I used and it worked fine for me :
[[local|localrc]]
HOST_IP=192.168.97.2
FIRST_HOST=192.168.97.1
FLAT_INTERFACE=eth0
FIXED_NETWORK_SIZE=4096
MULTI_HOST=1
LOGFILE=/opt/stack/logs/stack.sh.log
ADMIN_PASSWORD=openstack
MYSQL_PASSWORD=DB-secret
RABBIT_PASSWORD=MQ-secret
SERVICE_PASSWORD=SERVICE-secret
SERVICE_TOKEN=ALongStringUsuallyHere
DATABASE_TYPE=mysql
SERVICE_HOST=$FIRST_HOST
MYSQL_HOST=$FIRST_HOST
RABBIT_HOST=$FIRST_HOST
GLANCE_HOSTPORT=$FIRST_HOST:9292
ENABLED_SERVICES=n-cpu,n-net,c-vol
NOVA_VNC_ENABLED=True
NOVNCPROXY_URL="http://$FIRST_HOST:6080/vnc_auto.html"
VNCSERVER_LISTEN=$HOST_IP
VNCSERVER_PROXYCLIENT_ADDRESS=$VNCSERVER_LISTEN
Only difference to thre one on Lab Notes seem to be that yours is missing below two lines:
Unortunately no change. I removed FIXED_RAGE and FLOATING_RANGE from ~/devstack/local.conf ran ./unstack.sh && ./clean.sh && ./stack.sh
received:
[...]
For whatever reason, when connecting to devstack-cc, compute-node is neither available on compute nodes nor can I see more than one hypervisor.
Thanks.
Did you use the stable version of devstack ?
git clone https://git.openstack.org/openstack-dev/devstack -b stable/liberty
I tried it again and it works for me,...
I missed that. I did think it actually before I tried even first time, but then I decided just to follow the docs. I will try this next,
Yes that was the thing, thank you!
If any of the admins are reading this, please update the docs, since the Labs don't work if you follow the docs exactly as they are at the moment.
On page 2 of 3.2 lab excercise, for Task 1, item4 the command needs to be applied as below:
[email protected]: ~$ cd devstack
[email protected]: ~/devstack$ ./stack.sh
Hope this helps.
My before and after ip output is as below:
[email protected]:~$ cat ip.before.out
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group defau
lt
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 g
roup default qlen 1000
link/ether bc:76:4e:04:5d:26 brd ff:ff:ff:ff:ff:ff
inet 192.168.97.2/24 brd 192.168.97.255 scope global eth0
valid_lft forever preferred_lft forever
inet6 fe80::be76:4eff:fe04:5d26/64 scope link
valid_lft forever preferred_lft forever
[email protected]:~$ cat ip.after.out
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group defau
lt
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
inet 169.254.169.254/32 scope link 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 g
roup default qlen 1000
link/ether bc:76:4e:04:5d:26 brd ff:ff:ff:ff:ff:ff
inet 192.168.97.2/24 brd 192.168.97.255 scope global eth0
valid_lft forever preferred_lft forever
inet6 fe80::be76:4eff:fe04:5d26/64 scope link
valid_lft forever preferred_lft forever
3: virbr0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOW
N group default
link/ether 2a:ed:56:d4:db:69 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
[email protected]:~$
I tried all above mentioned ip to open a browser and point it at the public IP Address of your devstack-cc node.
Nothing is working. Which IP I should be pointing that too.
Hello,
Thank you for the note. I will update the prompt to represent to correct directory to be in. If you follow the commands you would change into the directory back in item 2. Thank you for the content note.
Hello,
The Horizon BUI is running on the devstack-cc. You have run the above commands on the compute-node instead. Typically the most numerous type of node are dedicated to running the virtual machines and would not have other services, such as Horizon, installed or running. Use the public IP of the devstack-cc node, and you should be able to access Horizon.
Let us know if this does not work.
Regards,
I am doing the course in October and still having this same problem, i have used cd >devstack and run the ./stack.sh script form there and all went well. The problem is i cannot access the BUI of the devstack-cc from its public IP which is 192.168.97.1 according to the instruction on the lab3.2 sheet. I tried a couple of clean install by running ./unstack.sh &./clean.sh as recommanded and everytime no horizon BUI on the public address of the devstack-cc.
Any help will be very appreciated,
regards
To access the BUI during lab3.2 you will need to use the public IP, not a private IP.
The address you pasted, 192.168.97.1, is not the public ip, but the internal, private IP. Follows is an example from a lab I just started. Note that eth2 shows the IP you mention. The public IP address would be found under eth0. Your IP will be different. I have modified the public IP with XXX.YYY to make it more obvious in the output. Use the full IP that shows on YOUR machine instead.
Thanks for the reply, for anyone who made the same mistake i made ( see above,) to fix this problem you need to exit the node environment by typing "EXIT" this will take you back to the devstack cc environment, here you can lookup the "eth0" which is the public ip for the destack, just issue this command "ip addr show eth0". This solved my problem.
Mybe there are other issues that will affect your labenvironment aswell so becarefull and keep learning!!
I realy do not mind the little mistakes here and there in the instructions because it makes the learning more effective. BUT the lab environment need to be more responsive and stable for this course to reach its potential.
I do hope this small input helps
thank you very much to the staff and all the learners here.