[Yahoo-eng-team] [Bug 1709115] [NEW] IPV6 only network DHCP Namespace missing default route
Public bug reported: I noticed if I create an iPV6 only network using DHCPv6 stateful the qdhcp namespeace does not contain a default route to make recursive DNS requests possible as they need to route to external networks. The below shows what I see in the routing table of the qdhcp namespace with IPV6: # ip -6 ro ::9:9::/64 dev ns-e3832e4a-bd proto kernel metric 256 pref medium fe80::/64 dev ns-e3832e4a-bd proto kernel metric 256 pref medium # As we can see there is no default route so it is not possible to dns- masq to perform lookup: # ping6 2001:4860:4860::8844 connect: Network is unreachable >From within an instance on the same network/subnet if I set the resolv.conf to use the above external (google) ipv6 resolver I am able to reach it and get an address as expected as the route is getting set in the instance correctly. If we contrast this with IPv4 I can see a default route as shown below inside the qdhcp namespace: # ip -4 ro default via 172.16.0.1 dev ns-e3832e4a-bd 172.16.0.0/24 dev ns-e3832e4a-bd proto kernel scope link src 172.16.0.11 # I find it very strange we are missing this default gateway for IPV6 in the qDHCP name space. I did some further testing and found this is also the case in the LBAAS (v2) name space - I can see that a packet ipv6 only request would route into the LBAAS name space, haproxy would proxy it onto the instance, the instance would respond to haproxy but then the connection would drop as the lbaas namespace does not contain the default route. This is on the newton release of neutron, if you need any further details just let me know and I will be happy to help. ** Affects: neutron Importance: Undecided Status: New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to neutron. https://bugs.launchpad.net/bugs/1709115 Title: IPV6 only network DHCP Namespace missing default route Status in neutron: New Bug description: I noticed if I create an iPV6 only network using DHCPv6 stateful the qdhcp namespeace does not contain a default route to make recursive DNS requests possible as they need to route to external networks. The below shows what I see in the routing table of the qdhcp namespace with IPV6: # ip -6 ro ::9:9::/64 dev ns-e3832e4a-bd proto kernel metric 256 pref medium fe80::/64 dev ns-e3832e4a-bd proto kernel metric 256 pref medium # As we can see there is no default route so it is not possible to dns- masq to perform lookup: # ping6 2001:4860:4860::8844 connect: Network is unreachable From within an instance on the same network/subnet if I set the resolv.conf to use the above external (google) ipv6 resolver I am able to reach it and get an address as expected as the route is getting set in the instance correctly. If we contrast this with IPv4 I can see a default route as shown below inside the qdhcp namespace: # ip -4 ro default via 172.16.0.1 dev ns-e3832e4a-bd 172.16.0.0/24 dev ns-e3832e4a-bd proto kernel scope link src 172.16.0.11 # I find it very strange we are missing this default gateway for IPV6 in the qDHCP name space. I did some further testing and found this is also the case in the LBAAS (v2) name space - I can see that a packet ipv6 only request would route into the LBAAS name space, haproxy would proxy it onto the instance, the instance would respond to haproxy but then the connection would drop as the lbaas namespace does not contain the default route. This is on the newton release of neutron, if you need any further details just let me know and I will be happy to help. To manage notifications about this bug go to: https://bugs.launchpad.net/neutron/+bug/1709115/+subscriptions -- Mailing list: https://launchpad.net/~yahoo-eng-team Post to : yahoo-eng-team@lists.launchpad.net Unsubscribe : https://launchpad.net/~yahoo-eng-team More help : https://help.launchpad.net/ListHelp
[Yahoo-eng-team] [Bug 1709104] [NEW] cloud-init with only ipv6 Failes to POST encrypted password
Public bug reported: Release = newton When launching a Windows instance utilizing an IPv6-only network, cloudinit fails to post the encrypted (generated) password to the metadata service due to the absence of the metadata service on non- dual-stack/IPv4 probably ** Affects: neutron Importance: Undecided Status: New ** Description changed: - Release = neutron + Release = newton When launching a Windows instance utilizing an IPv6-only network, cloudinit fails to post the encrypted (generated) password to the metadata service due to the absence of the metadata service on non- dual-stack/IPv4 probably -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to neutron. https://bugs.launchpad.net/bugs/1709104 Title: cloud-init with only ipv6 Failes to POST encrypted password Status in neutron: New Bug description: Release = newton When launching a Windows instance utilizing an IPv6-only network, cloudinit fails to post the encrypted (generated) password to the metadata service due to the absence of the metadata service on non- dual-stack/IPv4 probably To manage notifications about this bug go to: https://bugs.launchpad.net/neutron/+bug/1709104/+subscriptions -- Mailing list: https://launchpad.net/~yahoo-eng-team Post to : yahoo-eng-team@lists.launchpad.net Unsubscribe : https://launchpad.net/~yahoo-eng-team More help : https://help.launchpad.net/ListHelp
[Yahoo-eng-team] [Bug 1708458] [NEW] Expose instance system_metadata in compute API
Public bug reported: Via the nova compute API I can see it is possible to GET metadata for an instance but it does not seem to be possible to get system_metadata. It is useful to be able to GET the system metadata fro an instance if you want to query the point in time properties that where inherited from an image during the launch. I noticed in change https://review.openstack.org/#/c/7045/5/nova/db/api.py there is a db api definition to read this from the database but this does not seem to be exposed in the compute api. I assume something like the below in compute/api.py would be a starting point to expose this. def get_instance_system_metadata(self, context, instance): """Get all system metadata associated with an instance.""" return self.db.instance_system_metadata_get(context, instance.uuid) ** Affects: nova Importance: Undecided Status: New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Compute (nova). https://bugs.launchpad.net/bugs/1708458 Title: Expose instance system_metadata in compute API Status in OpenStack Compute (nova): New Bug description: Via the nova compute API I can see it is possible to GET metadata for an instance but it does not seem to be possible to get system_metadata. It is useful to be able to GET the system metadata fro an instance if you want to query the point in time properties that where inherited from an image during the launch. I noticed in change https://review.openstack.org/#/c/7045/5/nova/db/api.py there is a db api definition to read this from the database but this does not seem to be exposed in the compute api. I assume something like the below in compute/api.py would be a starting point to expose this. def get_instance_system_metadata(self, context, instance): """Get all system metadata associated with an instance.""" return self.db.instance_system_metadata_get(context, instance.uuid) To manage notifications about this bug go to: https://bugs.launchpad.net/nova/+bug/1708458/+subscriptions -- Mailing list: https://launchpad.net/~yahoo-eng-team Post to : yahoo-eng-team@lists.launchpad.net Unsubscribe : https://launchpad.net/~yahoo-eng-team More help : https://help.launchpad.net/ListHelp
[Yahoo-eng-team] [Bug 1692007] [NEW] Dual Stack IPV4/6 ARP bleed
Public bug reported: Version = newton (2:9.2.0-0ubuntu1~cloud0) DVR (HA mode) with ipv4 and ipv6 l2_population = True I noticed when using dual stack ipv4 and ipv6 Linux guests are working fine but windows guests seemed to have a problem with their ipv4 connectivity. Upon investigation I found an ARP issue in that both the ipv4 and ipv6 interface of the virtual router are responding to arp requests the below shows a capture from the tap interface of the guest when the arp table in the guest is flushed. 12:04:58.273446 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 10.0.0.1 tell 10.0.0.23, length 28 12:04:58.273776 ARP, Ethernet (len 6), IPv4 (len 4), Reply 10.0.0.1 is-at fa:16:3e:43:9d:58, length 28 12:04:58.273790 ARP, Ethernet (len 6), IPv4 (len 4), Reply 10.0.0.1 is-at fa:16:3e:19:96:e6, length 28 If I look at the active router interfaxces I can see mac ending 9d:58 is the ipv4 interface and mac ending 96:e6 is the ipv6 interfaces as shown below: # ip a 1: lo:mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1 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: qr-12e41bc2-68@if1021: mtu 8950 qdisc noqueue state UP group default qlen 1000 link/ether fa:16:3e:43:9d:58 brd ff:ff:ff:ff:ff:ff link-netnsid 0 inet 10.0.0.1/8 brd 10.255.255.255 scope global qr-12e41bc2-68 valid_lft forever preferred_lft forever inet6 fe80::f816:3eff:fe43:9d58/64 scope link valid_lft forever preferred_lft forever 6: qr-bd91567c-81@if1143: mtu 8950 qdisc noqueue state UP group default qlen 1000 link/ether fa:16:3e:19:96:e6 brd ff:ff:ff:ff:ff:ff link-netnsid 0 inet6 :9:1::1/64 scope global valid_lft forever preferred_lft forever inet6 fe80::f816:3eff:fe19:96e6/64 scope link valid_lft forever preferred_lft forever 7: qr-03c27e46-4b@if1159: mtu 8950 qdisc noqueue state UP group default qlen 1000 link/ether fa:16:3e:8e:f7:bd brd ff:ff:ff:ff:ff:ff link-netnsid 0 inet6 :9:2::1/64 scope global valid_lft forever preferred_lft forever inet6 fe80::f816:3eff:fe8e:f7bd/64 scope link valid_lft forever preferred_lft forever We should not see two arp responses here, we should only see one from fa:16:3e:43:9d:58. Turns out that Linux guests populates the arp table based upon the first response and windows guests based upon the latest response, this explains to me why windows guests are failing and Linux are working as the first arp response is valid and the second one is invalid - but I think we have a bigger issue here as I should not be getting an arp response for the ipv6 interface. ** Affects: neutron Importance: Undecided Status: New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to neutron. https://bugs.launchpad.net/bugs/1692007 Title: Dual Stack IPV4/6 ARP bleed Status in neutron: New Bug description: Version = newton (2:9.2.0-0ubuntu1~cloud0) DVR (HA mode) with ipv4 and ipv6 l2_population = True I noticed when using dual stack ipv4 and ipv6 Linux guests are working fine but windows guests seemed to have a problem with their ipv4 connectivity. Upon investigation I found an ARP issue in that both the ipv4 and ipv6 interface of the virtual router are responding to arp requests the below shows a capture from the tap interface of the guest when the arp table in the guest is flushed. 12:04:58.273446 ARP, Ethernet (len 6), IPv4 (len 4), Request who-has 10.0.0.1 tell 10.0.0.23, length 28 12:04:58.273776 ARP, Ethernet (len 6), IPv4 (len 4), Reply 10.0.0.1 is-at fa:16:3e:43:9d:58, length 28 12:04:58.273790 ARP, Ethernet (len 6), IPv4 (len 4), Reply 10.0.0.1 is-at fa:16:3e:19:96:e6, length 28 If I look at the active router interfaxces I can see mac ending 9d:58 is the ipv4 interface and mac ending 96:e6 is the ipv6 interfaces as shown below: # ip a 1: lo: mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1 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: qr-12e41bc2-68@if1021: mtu 8950 qdisc noqueue state UP group default qlen 1000 link/ether fa:16:3e:43:9d:58 brd ff:ff:ff:ff:ff:ff link-netnsid 0 inet 10.0.0.1/8 brd 10.255.255.255 scope global qr-12e41bc2-68 valid_lft forever preferred_lft forever inet6 fe80::f816:3eff:fe43:9d58/64 scope link valid_lft forever preferred_lft forever 6: qr-bd91567c-81@if1143:
[Yahoo-eng-team] [Bug 1691427] [NEW] AttributeError: 'DvrLocalRouter' object has no attribute 'ha_state'
Public bug reported: I have noticed the below in the newutron l3 agent log of compute nodes running in DVR mode, It seems to log the same message every 60 seconds or so. Restarting the agent will stop the log until log rotate is called the next day and then the return in the same fashion. 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task [req-5c48816b-0705-438a-b823-eca41cfe6c35 - - - - -] Error during L3NATAgentWithStateReport.periodic_sync_routers_task 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task Traceback (most recent call last): 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/oslo_service/periodic_task.py", line 220, in run_periodic_tasks 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task task(self, context) 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/neutron/agent/l3/agent.py", line 552, in periodic_sync_routers_task 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task self.fetch_and_sync_all_routers(context, ns_manager) 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/osprofiler/profiler.py", line 154, in wrapper 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task return f(*args, **kwargs) 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/neutron/agent/l3/agent.py", line 586, in fetch_and_sync_all_routers 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task r['id'], r.get(l3_constants.HA_ROUTER_STATE_KEY)) 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/neutron/agent/l3/ha.py", line 120, in check_ha_state_for_router 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task if ri and current_state != TRANSLATION_MAP[ri.ha_state]: 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task AttributeError: 'DvrLocalRouter' object has no attribute 'ha_state' 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task [req-5c48816b-0705-438a-b823-eca41cfe6c35 - - - - -] Error during L3NATAgentWithStateReport.periodic_sync_routers_task 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task Traceback (most recent call last): 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/oslo_service/periodic_task.py", line 220, in run_periodic_tasks 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task task(self, context) 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/neutron/agent/l3/agent.py", line 552, in periodic_sync_routers_task 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task self.fetch_and_sync_all_routers(context, ns_manager) 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/osprofiler/profiler.py", line 154, in wrapper 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task return f(*args, **kwargs) 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/neutron/agent/l3/agent.py", line 586, in fetch_and_sync_all_routers 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task r['id'], r.get(l3_constants.HA_ROUTER_STATE_KEY)) 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task File "/usr/lib/python2.7/dist-packages/neutron/agent/l3/ha.py", line 120, in check_ha_state_for_router 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task if ri and current_state != TRANSLATION_MAP[ri.ha_state]: 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task AttributeError: 'DvrLocalRouter' object has no attribute 'ha_state' 2017-05-17 10:37:50.102 27502 ERROR oslo_service.periodic_task ** Affects: neutron Importance: Undecided Status: New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to neutron. https://bugs.launchpad.net/bugs/1691427 Title: AttributeError: 'DvrLocalRouter' object has no attribute 'ha_state' Status in neutron: New Bug description: I have noticed the below in the newutron l3 agent log of compute nodes running in DVR mode, It seems to log the same message every 60 seconds or so. Restarting the agent will stop the log until log rotate is called the next day and then the return in the same fashion. 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task [req-5c48816b-0705-438a-b823-eca41cfe6c35 - - - - -] Error during L3NATAgentWithStateReport.periodic_sync_routers_task 2017-05-17 10:37:10.432 27502 ERROR oslo_service.periodic_task Traceback (most recent call last): 2017-05-17 10:37:10.432 27502 ERROR
[Yahoo-eng-team] [Bug 1686729] [NEW] Creating object storage container causes user to be logged out
Public bug reported: Version = openstack-dashboard3:11.0.1-0ubuntu1~cloud0 Ceph version = 10.2.7 When using ceph RGW swift interface for open stack and the open stack dashboard version above to create a swift container the dashboard does a number of curl requests to check if the bucket name already exists to prevent the user from trying to create a bucket with the same name as an existing bucket. In most cases this works as expected, however if I try to create a bucket that starts with the same name as an existing bucket that has the ACL set to private I am unexpectedly logged out of the dashboard. In my tests I have open stack user 'paul' and project 'paul that owns a private swift bucket called 'paul' I then as a second user 'sean' and project 'sean' try to create a swift container called 'paul1' this will result in me getting logged out of the dashboard, The below shows the log file for when I try and create this bucket: `` REQ: curl -i https://rgw.domain.com/swift/v1/p/ -X GET -H "X-Auth-Token: {hidden}" RESP STATUS: 400 Bad Request RESP HEADERS: {u'Date': u'Thu, 27 Apr 2017 13:22:01 GMT', u'Content-Length': u'17', u'Content-Type': u'text/plain; charset=utf-8', u'Accept-Ranges': u'bytes', u'X-Trans-Id': u'{hidden}'} RESP BODY: InvalidBucketName REQ: curl -i https://rgw.domain.com/swift/v1/pa/ -X GET -H "X-Auth-Token: {hidden}" RESP STATUS: 400 Bad Request RESP HEADERS: {u'Date': u'Thu, 27 Apr 2017 13:22:02 GMT', u'Content-Length': u'17', u'Content-Type': u'text/plain; charset=utf-8', u'Accept-Ranges': u'bytes', u'X-Trans-Id': u'{hidden}'} RESP BODY: InvalidBucketName REQ: curl -i https://rgw.domain.com/swift/v1/pau/ -X GET -H "X-Auth-Token: {hidden}" RESP STATUS: 404 Not Found RESP HEADERS: {u'Date': u'Thu, 27 Apr 2017 13:22:04 GMT', u'Content-Length': u'12', u'Content-Type': u'text/plain; charset=utf-8', u'Accept-Ranges': u'bytes', u'X-Trans-Id': u'{hidden}'} RESP BODY: NoSuchBucket REQ: curl -i https://rgw.domain.com/swift/v1/paul/ -X GET -H "X-Auth-Token: {hidden}" RESP STATUS: 401 Unauthorized RESP HEADERS: {u'Date': u'Thu, 27 Apr 2017 13:22:04 GMT', u'Content-Length': u'12', u'Content-Type': u'text/plain; charset=utf-8', u'Accept-Ranges': u'bytes', u'X-Trans-Id': u'{hidden}'} RESP BODY: AccessDenied Logging out user "sean `` As you can see this works until the 401 is received by horizon from the rgw when checking bucket 'paul' I believe this is because the bucket ACL of Paul (created by user Paul) is set to ACL private as I don't have the same issue when the ACL is set to public or when the ACL is private and I try and create the bucket 'paul1' as the user 'paul' ** Affects: horizon Importance: Undecided Status: New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Dashboard (Horizon). https://bugs.launchpad.net/bugs/1686729 Title: Creating object storage container causes user to be logged out Status in OpenStack Dashboard (Horizon): New Bug description: Version = openstack-dashboard3:11.0.1-0ubuntu1~cloud0 Ceph version = 10.2.7 When using ceph RGW swift interface for open stack and the open stack dashboard version above to create a swift container the dashboard does a number of curl requests to check if the bucket name already exists to prevent the user from trying to create a bucket with the same name as an existing bucket. In most cases this works as expected, however if I try to create a bucket that starts with the same name as an existing bucket that has the ACL set to private I am unexpectedly logged out of the dashboard. In my tests I have open stack user 'paul' and project 'paul that owns a private swift bucket called 'paul' I then as a second user 'sean' and project 'sean' try to create a swift container called 'paul1' this will result in me getting logged out of the dashboard, The below shows the log file for when I try and create this bucket: `` REQ: curl -i https://rgw.domain.com/swift/v1/p/ -X GET -H "X-Auth-Token: {hidden}" RESP STATUS: 400 Bad Request RESP HEADERS: {u'Date': u'Thu, 27 Apr 2017 13:22:01 GMT', u'Content-Length': u'17', u'Content-Type': u'text/plain; charset=utf-8', u'Accept-Ranges': u'bytes', u'X-Trans-Id': u'{hidden}'} RESP BODY: InvalidBucketName REQ: curl -i https://rgw.domain.com/swift/v1/pa/ -X GET -H "X-Auth-Token: {hidden}" RESP STATUS: 400 Bad Request RESP HEADERS: {u'Date': u'Thu, 27 Apr 2017 13:22:02 GMT', u'Content-Length': u'17', u'Content-Type': u'text/plain; charset=utf-8', u'Accept-Ranges': u'bytes', u'X-Trans-Id': u'{hidden}'} RESP BODY: InvalidBucketName REQ: curl -i https://rgw.domain.com/swift/v1/pau/ -X GET -H "X-Auth-Token: {hidden}" RESP STATUS: 404 Not Found RESP HEADERS: {u'Date': u'Thu, 27 Apr 2017 13:22:04 GMT', u'Content-Length': u'12', u'Content-Type': u'text/plain; charset=utf-8', u'Accept-Ranges': u'bytes', u'X-Trans-Id':
[Yahoo-eng-team] [Bug 1633405] [NEW] Lock/Unlock status not visible in dashboard or action log
Public bug reported: It appears that when the Lock Instance / Unlock Instance options are used on a compute instance, nothing is indicated within the dashboard or action log to suggest this is the case. Steps to repeat: Log into dashboard Click instances >From the drop down on the right of a given instance, select Lock instance Try to power on/off/restart the instance, You will get the error: Error: Unable to shut off instance: , with no indication of the reason. Click into the instance and click the action log, there is no record there of the instance being locked. This makes it difficult for an end user to determine why they cannot interact with their instance. Version info: ii python-django-horizon 2:9.1.0-0ubuntu1all Django module providing web based interaction with OpenStack ii openstack-dashboard-ubuntu-theme 2:9.1.0-0ubuntu1all Ubuntu theme for the OpenStack dashboard ** Affects: horizon Importance: Undecided Status: New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Dashboard (Horizon). https://bugs.launchpad.net/bugs/1633405 Title: Lock/Unlock status not visible in dashboard or action log Status in OpenStack Dashboard (Horizon): New Bug description: It appears that when the Lock Instance / Unlock Instance options are used on a compute instance, nothing is indicated within the dashboard or action log to suggest this is the case. Steps to repeat: Log into dashboard Click instances From the drop down on the right of a given instance, select Lock instance Try to power on/off/restart the instance, You will get the error: Error: Unable to shut off instance: , with no indication of the reason. Click into the instance and click the action log, there is no record there of the instance being locked. This makes it difficult for an end user to determine why they cannot interact with their instance. Version info: ii python-django-horizon 2:9.1.0-0ubuntu1all Django module providing web based interaction with OpenStack ii openstack-dashboard-ubuntu-theme 2:9.1.0-0ubuntu1all Ubuntu theme for the OpenStack dashboard To manage notifications about this bug go to: https://bugs.launchpad.net/horizon/+bug/1633405/+subscriptions -- Mailing list: https://launchpad.net/~yahoo-eng-team Post to : yahoo-eng-team@lists.launchpad.net Unsubscribe : https://launchpad.net/~yahoo-eng-team More help : https://help.launchpad.net/ListHelp
[Yahoo-eng-team] [Bug 1528877] [NEW] Libvirt can't honour user-supplied dev names
Public bug reported: OpenStack liberty (Ubuntu 14.04) When creating a new instance via the dashboard and selecting 'boot from image (creates new volume)' the below is logged into nova-compute.log on the host the instance is deployed to: "Ignoring supplied device name: /dev/vda. Libvirt can't honour user- supplied dev names" In the virsh XML for the instance I can see the target dev name is sda, this seems to have changed from how kilo was behaving. I also noticed that if I then try and add an extra volume to this instance the below error is placed in the nova-compute.log and the volume is not attached. caf420922780362 - - -] Exception during message handling: internal error: unable to execute QEMU command 'device_add': Duplicate ID 'scsi0-0-0-0' for device 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher Traceback (most recent call last): 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_messaging/rpc/dispatcher.py", line 142, in _dispatch_and_reply 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher executor_callback)) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_messaging/rpc/dispatcher.py", line 186, in _dispatch 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher executor_callback) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_messaging/rpc/dispatcher.py", line 129, in _do_dispatch 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher result = func(ctxt, **new_args) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/exception.py", line 89, in wrapped 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher payload) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 195, in __exit__ 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher six.reraise(self.type_, self.value, self.tb) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/exception.py", line 72, in wrapped 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher return f(self, context, *args, **kw) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 378, in decorated_function 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher kwargs['instance'], e, sys.exc_info()) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 195, in __exit__ 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher six.reraise(self.type_, self.value, self.tb) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 366, in decorated_function 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher return function(self, context, *args, **kwargs) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 4609, in attach_volume 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher do_attach_volume(context, instance, driver_bdm) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_concurrency/lockutils.py", line 254, in inner 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher return f(*args, **kwargs) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 4607, in do_attach_volume 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher bdm.destroy() 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 195, in __exit__ 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher six.reraise(self.type_, self.value, self.tb) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 4604, in do_attach_volume 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher return self._attach_volume(context, instance, driver_bdm) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 4627, in _attach_volume 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher self.volume_api.unreserve_volume(context, bdm.volume_id) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File
[Yahoo-eng-team] [Bug 1528877] Re: Libvirt can't honour user-supplied dev names
** Also affects: ubuntu Importance: Undecided Status: New ** No longer affects: ubuntu -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Compute (nova). https://bugs.launchpad.net/bugs/1528877 Title: Libvirt can't honour user-supplied dev names Status in OpenStack Compute (nova): New Bug description: OpenStack liberty (Ubuntu 14.04) When creating a new instance via the dashboard and selecting 'boot from image (creates new volume)' the below is logged into nova- compute.log on the host the instance is deployed to: "Ignoring supplied device name: /dev/vda. Libvirt can't honour user- supplied dev names" In the virsh XML for the instance I can see the target dev name is sda, this seems to have changed from how kilo was behaving. I also noticed that if I then try and add an extra volume to this instance the below error is placed in the nova-compute.log and the volume is not attached. caf420922780362 - - -] Exception during message handling: internal error: unable to execute QEMU command 'device_add': Duplicate ID 'scsi0-0-0-0' for device 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher Traceback (most recent call last): 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_messaging/rpc/dispatcher.py", line 142, in _dispatch_and_reply 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher executor_callback)) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_messaging/rpc/dispatcher.py", line 186, in _dispatch 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher executor_callback) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_messaging/rpc/dispatcher.py", line 129, in _do_dispatch 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher result = func(ctxt, **new_args) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/exception.py", line 89, in wrapped 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher payload) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 195, in __exit__ 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher six.reraise(self.type_, self.value, self.tb) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/exception.py", line 72, in wrapped 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher return f(self, context, *args, **kw) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 378, in decorated_function 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher kwargs['instance'], e, sys.exc_info()) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 195, in __exit__ 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher six.reraise(self.type_, self.value, self.tb) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 366, in decorated_function 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher return function(self, context, *args, **kwargs) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 4609, in attach_volume 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher do_attach_volume(context, instance, driver_bdm) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_concurrency/lockutils.py", line 254, in inner 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher return f(*args, **kwargs) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 4607, in do_attach_volume 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher bdm.destroy() 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 195, in __exit__ 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher six.reraise(self.type_, self.value, self.tb) 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 4604, in do_attach_volume 2015-12-23 15:21:55.698 4621 ERROR oslo_messaging.rpc.dispatcher return
[Yahoo-eng-team] [Bug 1502876] [NEW] Error: You are not allowed to delete volume: $X
Public bug reported: When using Ceph as the back end storage a snapshot of a volume is dependent upon the original volume meaning it is an illegal operation to delete the volume without first deleting the snapshot. Should you try and delete such a volume horizon {kilo} reports back the error "Error: You are not allowed to delete volume:" - This error message is confusing and leaves the end user unsure why the volume can not be removed. I think we should handle this by better checking what cinder API reports back and displaying a more user friendly error message. The below is an example of the error that comes back from cinder, passing this up in horizon would be very helpful. $ cinder delete 34feea60-fb50-4fbe-8136-885f5553a8a8 Delete for volume 34feea60-fb50-4fbe-8136-885f5553a8a8 failed: Invalid volume: Volume still has 1 dependent snapshots. (HTTP 400) (Request-ID: req-320be24f-ce7a-4a79-9d6f-9166821c42e9) ERROR: Unable to delete any of specified volumes. $ ** Affects: horizon Importance: Undecided Status: New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Dashboard (Horizon). https://bugs.launchpad.net/bugs/1502876 Title: Error: You are not allowed to delete volume: $X Status in OpenStack Dashboard (Horizon): New Bug description: When using Ceph as the back end storage a snapshot of a volume is dependent upon the original volume meaning it is an illegal operation to delete the volume without first deleting the snapshot. Should you try and delete such a volume horizon {kilo} reports back the error "Error: You are not allowed to delete volume:" - This error message is confusing and leaves the end user unsure why the volume can not be removed. I think we should handle this by better checking what cinder API reports back and displaying a more user friendly error message. The below is an example of the error that comes back from cinder, passing this up in horizon would be very helpful. $ cinder delete 34feea60-fb50-4fbe-8136-885f5553a8a8 Delete for volume 34feea60-fb50-4fbe-8136-885f5553a8a8 failed: Invalid volume: Volume still has 1 dependent snapshots. (HTTP 400) (Request-ID: req-320be24f-ce7a-4a79-9d6f-9166821c42e9) ERROR: Unable to delete any of specified volumes. $ To manage notifications about this bug go to: https://bugs.launchpad.net/horizon/+bug/1502876/+subscriptions -- Mailing list: https://launchpad.net/~yahoo-eng-team Post to : yahoo-eng-team@lists.launchpad.net Unsubscribe : https://launchpad.net/~yahoo-eng-team More help : https://help.launchpad.net/ListHelp