ajayaa / cloudguru Goto Github PK

need to add some entries to ml2_conf.ini file, and restart the service. but still after doing that, it doesn't work

This issue is to keep track of things to look into. These things might be required to understand and implement the cloud fully

1. vif plugging. Mention in puppet-nova/manifests/network/neutron.pp

maybe for now, we just need to keep this in mind, but we should be solving this..
all api accessess fails due to this (e.g. keystone token-get)

old values still persist in the db. need to set a flag which takes care of cleaning up of nova database

shouldn't be doing in prod, but necessary before that..

LOG:

checking http://169.254.169.254/2009-04-04/instance-id
failed 1/20: up 10.86. request failed
failed 2/20: up 13.50. request failed
failed 3/20: up 15.80. request failed
failed 4/20: up 18.05. request failed
failed 5/20: up 20.29. request failed
failed 6/20: up 23.37. request failed
failed 7/20: up 25.90. request failed
failed 8/20: up 28.15. request failed

It's not causing an error, but just the VM boot takes half a minute longer due to this.. Maybe adding nova metadata server will solve this problem..

this is for ha/load balancing between controller nodes. writing here so that we don't forget about it in future

Currently, we assume the user has generated them and kept ready for use

what is it, and how can it be beneficial, and is it required?

We need to run this after puppet is done.
rabbitmqctl set_permissions rabbituser "." "." ".*"

see link https://github.com/enovance/puppet-openstack-cloud/blob/master/manifests/identity.pp
to know how they moved keystone related stuff to a separate file. we should do this too, sooner rather than later..

This is related to mtu size it seems.
Go to the vm in compute node with xvnc4viewer and run:
sudo ip link set eth0 mtu 1400

For full details refer to
https://ask.openstack.org/en/question/30502/can-ping-vm-instance-but-cant-ssh-ssh-command-halts-with-no-output/

so that it's visible that the logs before this point is from a previous puppet run

e.g. "==============================================="

if proxy vars is set, puppet keystone automatically uses them to connect to keystone server, and fails..

On /var/log/apache2/keystone_wsgi_admin_error_ssl.log, it says

Daemon process called 'keystone' cannot be accessed by this WSGI application: /usr/lib/cgi-bin/keystone/admin

And on /var/log/apache2/keystone_wsgi_main_error_ssl.log, the error is:

AH01909: RSA certificate configured for keystone_wsgi_main:443 does NOT include an ID which matches the server name

Currently, it is a PKI (long) token

looks like amqp server is not running.

glance doesn't list amqp/rabbit/qpid as a dependency. Need to find out how and why the error is happening

this should be a parameter to mysql class. need to add it by finding it in mysql puppet class/module

otherwise nova and glance commands fail..

allowed_hosts, bind address needs to be changed/managed better