We have configured Neutron and ml2 agent to use MTU 1500 using linux bridge agent but the VM's cannot run apt update but if we set the MTU to 1450 in the VM network interface works well.
Here is my neutron.conf value and ml2_conf.ini.
neutron.conf:
global_physnet_mtu = 1550
ml2_conf.ini:
[DEFAULT]
path_mtu = 1550
physical_network_mtus = physnet1:1550
Vxlan:
+---------------------------+--------------------------------------+
| Field | Value |
+---------------------------+--------------------------------------+
| admin_state_up | UP |
| availability_zone_hints | |
| availability_zones | nova |
| created_at | 2021-03-30T17:40:59Z |
| description | |
| dns_domain | None |
| id | 4ddce1ef-8151-4c67-b3ca-7519f2ab54c3 |
| ipv4_address_scope | None |
| ipv6_address_scope | None |
| is_default | None |
| is_vlan_transparent | None |
| mtu | 1500 |
| name | gfgfggf |
| port_security_enabled | True |
| project_id | 2467cb8a053d40f0a0a8af43d89d1c32 |
| provider:network_type | vxlan |
| provider:physical_network | None |
| provider:segmentation_id | 2 |
| qos_policy_id | None |
| revision_number | 4 |
| router:external | Internal |
| segments | None |
| shared | False |
| status | ACTIVE |
| subnets | 9a312e58-e830-4dc5-bd6e-ad903230f6b4 |
| tags | |
| updated_at | 2021-03-30T17:57:31Z |
+---------------------------+--------------------------------------+
Anyone can help?
UPDATE: The issue was solved restarting the controller and all hypervisors.
Thanks.
The settings in ml2_conf.ini should be under the heading [ml2]. [DEFAULT] is used in neutron.conf.
Related
I simply ask if there is any way to change an openstack flavor name after its creation with VMs using it as flavor, for example consider this flavor:
+----------------------------+--------------------+
| Field | Value |
+----------------------------+--------------------+
| OS-FLV-DISABLED:disabled | False |
| OS-FLV-EXT-DATA:ephemeral | 0 |
| access_project_ids | None |
| description | None |
| disk | 0 |
| id | g1-1-1-0 |
| name | g1-1-1-0 |
| os-flavor-access:is_public | True |
| properties | |
| ram | 1024 |
| rxtx_factor | 1.0 |
| swap | |
| vcpus | 1 |
+----------------------------+--------------------+
How to change its name field to g1-small?
Is it possible to configure the mtu of a network while creating it with a heat template in Openstack?
I can set the mtu through the CLI openstack network --mtu <value> <network_id>
I see a reference to mtu in the heat guide but it doesn't appear to allow setting of the value.
Yes, it is possible to set the MTU with heat-templates. You have to use the value_specs-field, which allows you additional properties. See its really bad documentation here:
https://docs.openstack.org/heat/pike/template_guide/openstack.html#OS::Neutron::Subnet-prop-value_specs
I tested it with the following simple heat-template:
heat_template_version: 2013-05-23
description: tests network
resources:
demo-network:
type: OS::Neutron::Net
properties:
name: demo-network
value_specs: { mtu: 1400 }
It really created a network with an MTU of 1400 like set in the template. Checked it with a show-command:
root#test-node:~# openstack network show demo-network
+---------------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| Field | Value |
+---------------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------+
| admin_state_up | UP |
| availability_zone_hints | |
| availability_zones | |
| created_at | 2020-05-16T13:33:53Z |
| description | |
| dns_domain | None |
| id | 965d890f-0c37-400e-8ee0-ec6ce1fcf4e6 |
| ipv4_address_scope | None |
| ipv6_address_scope | None |
| is_default | None |
| is_vlan_transparent | None |
| location | cloud='', project.domain_id=, project.domain_name='Default', project.id='2dacaa8f6d3348f28e126b371bf1dfab', project.name='admin', region_name='RegionOne', zone= |
| mtu | 1400 |
| name | demo-network |
| port_security_enabled | True |
| project_id | 2dacaa8f6d3348f28e126b371bf1dfab |
| provider:network_type | vxlan |
| provider:physical_network | None |
| provider:segmentation_id | 60 |
| qos_policy_id | None |
| revision_number | 1 |
| router:external | Internal |
| segments | None |
| shared | False |
| status | ACTIVE |
| subnets | |
| tags | |
| updated_at | 2020-05-16T13:33:53Z |
+---------------------------+------------------------------------------------------------------------------------------------------------------------------------------------------------------+
This is my topology:
+-------------------+
| |
+---------+ ipmi switch +----+
| | | |
| +-------------------+ |
| |
| |
+---+---+ +---+---+
| | | |
| bmc | | bmc |
| | | |
+-----+-------+----------+ +-----+-------+----------+
| | | |
| | | |
| | | |
| host1 | | host2 |
| | | |
| | | |
| | | |
+-----+--------+---------+ +-----+--------+---------+
| | | |
| mgmt | | mgmt |
| nic | | nic |
+----+---+ +----+---+
| |
| |
| |
| |
| +-------------------+ |
| | | |
+----------+ mgmt switch +---+
| |
+-------------------+
Is there a way to detect bmc of host2 alive only via bmc of host1(via ipmi switch) not mgmt network(mgmt nic and mgmt switch)?
I searched man page of ipmitool and freeipmi but found nothing helpful
Under the assumption that the "bmc" NICs only do IPMI traffic and no other traffic (i.e. no normal IP traffic), I don't believe there is a way for bmc-host1 to perform IPMI traffic to bmc-host2 to detect that it is up. Client side tools in ipmitool and freeipmi are effectively doing IP traffic to a bmc on the other side. So there needs to be a way to reach bmc-host2 with normal IP traffic.
The Project -> NETWORK -> Network Topology always loading...
But the Project -> NETWORK -> Networks and Project -> NETWORK -> Routers all are good, will shows the information.
Finally I found nova-compute service did not start up:
[root#ha-node3 mysql]# nova service-list
+----+------------------+----------+----------+---------+-------+----------------------------+-----------------+
| Id | Binary | Host | Zone | Status | State | Updated_at | Disabled Reason |
+----+------------------+----------+----------+---------+-------+----------------------------+-----------------+
| 2 | nova-consoleauth | ha-node3 | internal | enabled | up | 2017-08-06T11:44:32.000000 | - |
| 5 | nova-conductor | ha-node3 | internal | enabled | up | 2017-08-06T11:44:32.000000 | - |
| 7 | nova-cert | ha-node3 | internal | enabled | up | 2017-08-06T11:44:32.000000 | - |
| 15 | nova-scheduler | ha-node3 | internal | enabled | up | 2017-08-06T11:44:32.000000 | - |
| 22 | nova-cert | ha-node1 | internal | enabled | up | 2017-08-06T11:43:40.000000 | - |
| 29 | nova-conductor | ha-node1 | internal | enabled | up | 2017-08-06T11:43:40.000000 | - |
| 32 | nova-consoleauth | ha-node1 | internal | enabled | up | 2017-08-06T11:43:40.000000 | - |
| 33 | nova-consoleauth | ha-node2 | internal | enabled | up | 2017-08-06T11:43:38.000000 | - |
| 36 | nova-scheduler | ha-node1 | internal | enabled | up | 2017-08-06T11:43:41.000000 | - |
| 40 | nova-conductor | ha-node2 | internal | enabled | up | 2017-08-06T11:43:39.000000 | - |
| 44 | nova-cert | ha-node2 | internal | enabled | up | 2017-08-06T11:43:39.000000 | - |
| 46 | nova-scheduler | ha-node2 | internal | enabled | up | 2017-08-06T11:43:48.000000 | - |
| 49 | nova-compute | ha-node2 | nova | enabled | down | 2017-08-06T08:57:50.000000 | - |
| 53 | nova-compute | ha-node1 | nova | enabled | down | 2017-08-06T08:57:30.000000 | - |
+----+------------------+----------+----------+---------+-------+----------------------------+-----------------+
So I restarted the nova-compute service:
systemctl start openstack-nova-compute.service
Is there any way to find the possible fields for the nova show --fields command? Currently I can fill the --fields parameter with anything and it never complains, just fails silently.
The docs say use the show command, but that reports a machines characteristics.
the nova show returns may different from version to version, here is an example:
$ nova show b3cdc6c0-85a7-4904-ae85-71918f734048
+-------------------------------------+-------------------------------------+
| Property | Value |
+-------------------------------------+-------------------------------------+
| OS-EXT-STS:task_state | scheduling |
| image | cirros-0.3.2-x86_64-uec |
| OS-EXT-STS:vm_state | building |
| OS-EXT-SRV-ATTR:instance_name | instance-00000002 |
| flavor | m1.small |
| id | b3cdc6c0-85a7-4904-ae85-71918f734048|
| security_groups | [{u'name': u'default'}] |
| user_id | 376744b5910b4b4da7d8e6cb483b06a8 |
| OS-DCF:diskConfig | MANUAL |
| accessIPv4 | |
| accessIPv6 | |
| progress | 0 |
| OS-EXT-STS:power_state | 0 |
| OS-EXT-AZ:availability_zone | nova |
| config_drive | |
| status | BUILD |
| updated | 2013-07-16T16:25:34Z |
| hostId | |
| OS-EXT-SRV-ATTR:host | None |
| key_name | KeyPair01 |
| OS-EXT-SRV-ATTR:hypervisor_hostname | None |
| name | myCirrosServer |
| adminPass | tVs5pL8HcPGw |
| tenant_id | 66265572db174a7aa66eba661f58eb9e |
| created | 2013-07-16T16:25:34Z |
| metadata | {u'KEY': u'VALUE'} |
+-------------------------------------+-------------------------------------+
The property column is all the fields available to specify to --fields