[Openstack] Nova-network multi-NIC
Dear openstackers, I'm missing a piece of information when I read this page http://docs.openstack.org/trunk/openstack-compute/admin/content/existing-ha-networking-options.html Scenario 1. My compute host has only one NIC but I'd like my VMs to be multi-NIC. I'd like to use VLANs in the host for the different virtual networks (I understand from the above page this is possible?). So in Scenario 1, the host may have multiple VLANs, each one with its own bridge, and the VMs obviously can connect to any of those virtual networks. Scenario 2. My compute host has multiple NICs and my VMs will be multi-NIC. One bridge/physical NIC on the host (and thus one virtual network / physical NIC). This case seems to me a bit more straightforward. Then I have three questions: 1. Are both scenarios supported by nova-network? If I read the page above, specially HA Option 3, I imply this is true 2. How would my nova.conf look like in both cases? There is usually only one entry for flat_network in it. 3. Will nova-network configure the VLANs for me in Scenario 1? Or will it only create the virtual bridges and attach to VLAN interfaces I must create manually? Please enlighten me. Viktor ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Grizzly multi-node setup
Hi all, After hunting ghosts this weekend, I finally found the problem. - Both my host and controller had firewalls set up blocking AMQP messages - RabbitMQ binds to **all** network interfaces through :::5672. Previously I believed this to be only v6 addresses. I realize now that if the flag IPV6_V6ONLY is not set, the IPv6 socket will serve both IPv4 and IPv6 traffic. It can be misleading when reading the output of netstat but that is not an OpenStack problem Thank you, Viktor On Sun, May 5, 2013 at 3:35 PM, Viktor Viking wrote: > Hi Ashok, > > In nova.conf I have "rabbit_host=192.168.1.2" both in the controller and > in the compute node. > > The output of netstat for listening ports is: > > tcp00 :::5672 :::*LISTEN27352/beam > > As I said, I have other connections with ESTABLISHED state on 5672 > Viktor > > > On Sun, May 5, 2013 at 3:09 PM, Ashok Kumaran wrote: > >> Can you post the value set for rabbitmq_host ? >> >> And also the netstat -ntlp Output >> >> Sent from my iPhone >> >> On 05-May-2013, at 6:29 PM, Viktor Viking >> wrote: >> >> Hi all, >> >> I'm trying to deploy a multi-node Grizzly (one controller, N compute >> nodes running nova-compute and nova-network). >> >> The controller seems to be doing fine. >> >> When starting the nova services in the compute node, I get the following >> error in the log: >> >> 2013-05-05 14:28:43.450 INFO nova.openstack.common.rpc.common >> [ref-f007f3f4-5223-3a3a-add-e059209849039 None None] Reconnecting to AMQP >> server on 192.168.1.2:5672 >> >> 2013-05-05 14:28:43.468 ERROR nova.openstack.common.rpc.common >> [ref-f007f3f4-5223-3a3a-add-e059209849039 None None] AMQP server on >> 192.168.1.2:5672 is unreachable: timed out. Trying again in 3 seconds. >> >> a. I can ping 192.168.1.2 from my compute node >> b. netstat -an in the controller shows me port 5672 has an ESTABLISHED >> connection >> >> I have not changed any rabbitmq conf (so guest/guest is being used) >> >> Any help is appreciated. >> Viktor >> >> ___ >> Mailing list: https://launchpad.net/~openstack >> Post to : openstack@lists.launchpad.net >> Unsubscribe : https://launchpad.net/~openstack >> More help : https://help.launchpad.net/ListHelp >> >> > ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] slow terminate jobs..
Hi Steve, >> the updated rabbitmq did default to an ipv6 address which I had to fix >> in the rabbit-env file. Could you tell me how you did that? I tried option "NODE_IP_ADDRESS=0.0.0.0" in rabbitmq-env but it insists on binding to IPv6 address. netstat -na | grep 5672 tcp00:::5672:::*LISTEN Thank you Viktor On Fri, May 3, 2013 at 10:04 PM, Steve Heistand wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > so its enabled and running. not sure what this all means. > I launched a few instances the messages when up and is holding steady. > Still taking forever to launch more then 1 image. > > the updated rabbitmq did default to an ipv6 address which I had to fix > in the rabbit-env file. might have missed something else that needs > tweaked between the older default version and this one. > > perhaps unrelated note that the updated openstack/rabbit seems > to have broken my ability to create cinder volumes. the server > nodes show up as down to the controller. Will work on that next. > > s > > > On 05/03/2013 11:43 AM, eric_e_sm...@dell.com wrote: > > Have you by chance looked at the management plugin for RabbitMQ? > > http://www.rabbitmq.com/management.html > > > > You can get great information from the management CLI / Web UI and > quickly determine > > if RabbitMQ is having a problem. > > > > -Original Message- From: Steve Heistand [mailto: > steve.heist...@nasa.gov] > > Sent: Friday, May 03, 2013 1:28 PM To: Smith, Eric E Cc: > > openstack@lists.launchpad.net Subject: Re: [Openstack] slow terminate > jobs.. > > > > well there are a lot (68) beam.smp processes running, well in existence > only a few at > > a time are consuming cpu time. Aside from that Im not sure how to answer > the > > question. or does this perhaps indicate a problem: > > > > root@cloudfe:/var/log/rabbitmq# tail > /var/log/rabbitmq/rab...@cloudfe.log > > > > =INFO REPORT 3-May-2013::11:14:36 === starting TCP connection > <0.14589.260> from > > 10.0.1.16:40015 > > > > =INFO REPORT 3-May-2013::11:19:13 === accepted TCP connection on > [::]:5672 from > > 10.0.1.16:40233 > > > > 5 minutes from starting connection to accepted connection. > > > > thanks > > > > s > > > > On 05/03/2013 11:22 AM, eric_e_sm...@dell.com wrote: > >> How does RabbitMQ look? Are the messages being consumed in a timely > fashion? > > > >> -Original Message- From: Openstack > >> [mailto:openstack-bounces+eric_e_smith=dell@lists.launchpad.net] > On Behalf Of > >> Steve Heistand Sent: Friday, May 03, 2013 1:14 PM To: > >> openstack@lists.launchpad.net Subject: [Openstack] slow terminate > jobs.. > > > >> so in addition to my problem of really long startup times, the process > of trying to > >> terminate a number of instances is taking ages as well. currently into > an 18 > >> instance terminate and its been 45 minutes. As there probably isnt much > data > >> movement involved with terminating a job Im wondering if there is a > >> keystone/nova/mysql/apache/something deadlock going on. normal CLI > commands are > >> taking a fairly long time to complete. 10 seconds for a nova > service-list (only 16 > >> compute nodes) 8 seconds for a userland nova list (20 instances > running) 7 seconds > >> for a keystone tenant-list > > > >> not really sure there is a question in all this aside from something > like has > >> anyone else seen such slowness? > > > >> s > > > > > >> ___ Mailing list: > >> https://launchpad.net/~openstack Post to : > openstack@lists.launchpad.net > >> Unsubscribe : https://launchpad.net/~openstack More help : > >> https://help.launchpad.net/ListHelp > > > > > > > > - -- > > Steve Heistand NASA Ames Research Center > email: steve.heist...@nasa.gov Steve Heistand/Mail Stop 258-6 > ph: (650) 604-4369 Bldg. 258, Rm. 232-5 > Scientific & HPC ApplicationP.O. Box 1 > Development/OptimizationMoffett Field, CA 94035-0001 > > "Any opinions expressed are those of our alien overlords, not my own." > > # For Remedy# > #Action: Resolve# > #Resolution: Resolved # > #Reason: No Further Action Required # > #Tier1: User Code # > #Tier2: Other # > #Tier3: Assistance # > #Notification: None # > -BEGIN PGP SIGNATURE- > Version: GnuPG v2.0.14 (GNU/Linux) > > iEYEARECAAYFAlGEGDMACgkQoBCTJSAkVrGGBgCfYDas8SRch3o4xYxTNT8Hn5fC > Db8An1hjCfzTLqcD6zKUP0uM16Ju6mNH > =Kmq4 > -END PGP SIGNATURE- > > ___ > Mailing list: https://launchpad.net/~openstack > Post to : openstack@lists.launchpad.net > Unsubscribe : https://launchpad.net/~openstack > More h
[Openstack] Grizzly multi-node setup
Hi all, I'm trying to deploy a multi-node Grizzly (one controller, N compute nodes running nova-compute and nova-network). The controller seems to be doing fine. When starting the nova services in the compute node, I get the following error in the log: 2013-05-05 14:28:43.450 INFO nova.openstack.common.rpc.common [ref-f007f3f4-5223-3a3a-add-e059209849039 None None] Reconnecting to AMQP server on 192.168.1.2:5672 2013-05-05 14:28:43.468 ERROR nova.openstack.common.rpc.common [ref-f007f3f4-5223-3a3a-add-e059209849039 None None] AMQP server on 192.168.1.2:5672 is unreachable: timed out. Trying again in 3 seconds. a. I can ping 192.168.1.2 from my compute node b. netstat -an in the controller shows me port 5672 has an ESTABLISHED connection I have not changed any rabbitmq conf (so guest/guest is being used) Any help is appreciated. Viktor ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Glance cannot authenticate
Hi Razique, Thank you for the tip. My problem was misconfiguration of admin_user/admin_password at /etc/glance/glance-api.conf. Best, Viktor On Tue, Apr 30, 2013 at 10:15 PM, Razique Mahroua wrote: > ok problem solved here. > As I was using an HA cluster between two nodes, I forgot to add a link > between the public IP of my APIs and the glance services - so Glance was > running on one node as the IP was running on the other one > You can use tcpdump in order to diagnose when it fails : > $ tcpdump -nvv -s0 -A -o ethX (eth1 in my case) tcp and port 5000 > > regards > > *Razique Mahroua** - **Nuage & Co* > razique.mahr...@gmail.com > Tel : +33 9 72 37 94 15 > > > Le 30 avr. 2013 à 19:31, Razique Mahroua a > écrit : > > Funny, > I'm facing the same issue today - are glance api/ registry conf files good > ? (especially the [paste_deploy] flavor=keystone for both api and registry) > what $ nova -debug image-show outputs? > > > > *Razique Mahroua** - **Nuage & Co* > razique.mahr...@gmail.com > Tel : +33 9 72 37 94 15 > > > > Le 30 avr. 2013 à 18:03, Viktor Viking a > écrit : > > Dear Openstackers, > > I'm facing authentication problems when running glance commands (e.g., > glance image-list) > > I have my environment variables set in the following way: > > OS_REGION_NAME=RegionOne > OS_PASSWORD=openstack > OS_AUTH_URL=http://192.168.0.1:5000/v2.0 > OS_USERNAME=admin > OS_TENANT_NAME=demo > > With these env variables set, I can run "keystone endpoint-list" and I > seehttp://192.168.0.1:5000/v2.0"; > as my publicurl for the keystone endpoint. So far so good. This also proves > the user=admin, pass=openstack, tenant=demo are correct. > > The problem happens when I try to use glance (for anything). "glance > image-list" for example returns an exception plus: > > NotAuthenticated : 401 Unauthorized > This server could not verify that you are authorized to access the > document you requested. Either you supplied the wrong credentials (e.g., > bad password), or your browser does not understand how to supply the > credentials required. > Authentication required > (HTTP 500) > > I don't have SERVICE_TOKEN variables set. > > I have seen this https://answers.launchpad.net/keystone/+question/212102but > it didn't solve my problem either. > > Does anyone have a clue that might enlighten me? > > Thank you, > Viktor > > > > ___ > Mailing list: https://launchpad.net/~openstack > Post to : openstack@lists.launchpad.net > Unsubscribe : https://launchpad.net/~openstack > More help : https://help.launchpad.net/ListHelp > > > > <>___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
[Openstack] Glance cannot authenticate
Dear Openstackers, I'm facing authentication problems when running glance commands (e.g., glance image-list) I have my environment variables set in the following way: OS_REGION_NAME=RegionOne OS_PASSWORD=openstack OS_AUTH_URL=http://192.168.0.1:5000/v2.0 OS_USERNAME=admin OS_TENANT_NAME=demo With these env variables set, I can run "keystone endpoint-list" and I seehttp://192.168.0.1:5000/v2.0"; as my publicurl for the keystone endpoint. So far so good. This also proves the user=admin, pass=openstack, tenant=demo are correct. The problem happens when I try to use glance (for anything). "glance image-list" for example returns an exception plus: NotAuthenticated : 401 Unauthorized This server could not verify that you are authorized to access the document you requested. Either you supplied the wrong credentials (e.g., bad password), or your browser does not understand how to supply the credentials required. Authentication required (HTTP 500) I don't have SERVICE_TOKEN variables set. I have seen this https://answers.launchpad.net/keystone/+question/212102but it didn't solve my problem either. Does anyone have a clue that might enlighten me? Thank you, Viktor ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Keystone Grizzly install
Community, That did the trick. Just install repoze-lru and keystone can start. -- Viktor On Thu, Apr 25, 2013 at 12:09 AM, Viktor Viking wrote: > Hi Dolph, > > Now I got an exception. It seems like I am missing "repoze.lru". I will > download and install it. I will let you know if it works. > > Thank you, > Viktor > > > > On Wed, Apr 24, 2013 at 11:25 PM, Dolph Mathews > wrote: > >> What happens when you run keystone-all directly? >> >> >> -Dolph >> >> >> On Wed, Apr 24, 2013 at 4:23 PM, Viktor Viking < >> viktor.viking...@gmail.com> wrote: >> >>> Community, >>> >>> I am trying to install Keystone Grizzly following these instructions: >>> http://docs.openstack.org/trunk/openstack-compute/install/yum/content/install-keystone.html >>> >>> When I try to start the service (before db sync), I get the following >>> error message: "Starting keytonestartproc: exit status of parent of >>> /usr/bin/keystone-all 1 failed". >>> >>> /var/log/keystone/keystone.log is not giving me any clue wrt what is >>> wrong. >>> >>> Could anyone let me know where to look? >>> Viktor >>> >>> ___ >>> Mailing list: https://launchpad.net/~openstack >>> Post to : openstack@lists.launchpad.net >>> Unsubscribe : https://launchpad.net/~openstack >>> More help : https://help.launchpad.net/ListHelp >>> >>> >> > ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Keystone Grizzly install
Hi Dolph, Now I got an exception. It seems like I am missing "repoze.lru". I will download and install it. I will let you know if it works. Thank you, Viktor On Wed, Apr 24, 2013 at 11:25 PM, Dolph Mathews wrote: > What happens when you run keystone-all directly? > > > -Dolph > > > On Wed, Apr 24, 2013 at 4:23 PM, Viktor Viking > wrote: > >> Community, >> >> I am trying to install Keystone Grizzly following these instructions: >> http://docs.openstack.org/trunk/openstack-compute/install/yum/content/install-keystone.html >> >> When I try to start the service (before db sync), I get the following >> error message: "Starting keytonestartproc: exit status of parent of >> /usr/bin/keystone-all 1 failed". >> >> /var/log/keystone/keystone.log is not giving me any clue wrt what is >> wrong. >> >> Could anyone let me know where to look? >> Viktor >> >> ___ >> Mailing list: https://launchpad.net/~openstack >> Post to : openstack@lists.launchpad.net >> Unsubscribe : https://launchpad.net/~openstack >> More help : https://help.launchpad.net/ListHelp >> >> > ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
[Openstack] Keystone Grizzly install
Community, I am trying to install Keystone Grizzly following these instructions: http://docs.openstack.org/trunk/openstack-compute/install/yum/content/install-keystone.html When I try to start the service (before db sync), I get the following error message: "Starting keytonestartproc: exit status of parent of /usr/bin/keystone-all 1 failed". /var/log/keystone/keystone.log is not giving me any clue wrt what is wrong. Could anyone let me know where to look? Viktor ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Horizon
Hi Emilien, I had edited keystone-data.sh and changed the password to "secret". But your pointers helped me. I had also changed the value "SERVICE_TENANT_NAME" to "MY_TENANT" as you can see in the mysql table. Apparently that was not a good idea. I dropped the tables, changed it back to "service", run the script again and… voilá. I can login now. Thank you for your help, Viktor On Wed, Feb 6, 2013 at 7:06 PM, Emilien Macchi wrote: > Hi Viktor, > > >- If you have used keystone-data.sh script, the password should be >"password". >- Horizon does not use novarc neither MySQL. It only uses OpenStack >APIs. >- Django log files can by found on /var/log/apache2/error.log & >access.log > > > Cheers, > > Emilien Macchi > > // eNovance Inc. http://enovance.com > // ✉ emil...@enovance.com ☎ +33 (0)1 49 70 99 80 > // 10 rue de la Victoire 75009 Paris > > > -- > *From: *"Viktor Viking" > *To: *openstack@lists.launchpad.net > *Sent: *Wednesday, February 6, 2013 1:47:02 PM > *Subject: *[Openstack] Horizon > > > Hi, > > After a successful installation of Folsom, I'm trying to use Dashboard. > I've had no success login in. > > If I just type a random user/passwd string, I get a message: "Invalid > username or password". > > If I type admin:secret (a correct user:password combination), the login > page reloads and nothing happens. > > Some info on my setup: > > I followed this guide for Folsom installation: > > > http://docs.openstack.org/folsom/basic-install/content/basic-install-folsom.pdf > > My novarc file: > > export OS_TENANT_NAME=admin > export OS_USERNAME=admin > export OS_PASSWORD=secret > export OS_AUTH_URL="http://localhost:5000/v2.0/"; > export SERVICE_ENDPOINT="http://localhost:35357/v2.0"; > export SERVICE_TOKEN=secret > > Keystone.log doesn't show anything out of the ordinary. > > mysql> select id,name from user; > +--+-+ > | id | name| > +--+-+ > | 6294f0f5e16d49b9992c211e9cac6855 | admin | > | 88f6d25dec6f40f1a9cddf2b20aad4b2 | cinder | > | 4c0f5a439a3d44368e8d4af395049ff2 | demo| > | c7ee8554321d4fe2bba341a99ee98dc9 | glance | > | 520ddc424a30447aa47992400382a167 | nova| > | 6e44c5aa1edc4fe5919286ecdda196a0 | quantum | > | f4a54be5b3154f85bc4df89a61c5360d | swift | > | 6dfec4512b7f4eaa9dc214d7232244fd | viktor | > +--+-+ > > mysql> select * from tenant; > > +--+++ > | id | name | extra >| > > +--+++ > | 779c87a9c37f4f63b0fc0aab236a5e1e | demo | {"enabled": > true, "description": null} | > | 8d4286a5249e43e9aa56864357737884 | invisible_to_admin | {"enabled": > true, "description": null} | > | cf03bbe4daaf4579bd005a4e64da3334 | MYTENANT | {"enabled": > true, "description": null} | > | f822dc409d684062b003ba99efa12672 | admin | {"enabled": > true, "description": null} | > > +--+++ > > I realize dashboard doesn't use a database anymore as it did in Essex. Is > that right? BTW, where can I find django log files? > > Any tips about what is going wrong? > > Thank you, > Viktor > > ___ > Mailing list: https://launchpad.net/~openstack > Post to : openstack@lists.launchpad.net > Unsubscribe : https://launchpad.net/~openstack > More help : https://help.launchpad.net/ListHelp > ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
[Openstack] Horizon
Hi, After a successful installation of Folsom, I'm trying to use Dashboard. I've had no success login in. If I just type a random user/passwd string, I get a message: "Invalid username or password". If I type admin:secret (a correct user:password combination), the login page reloads and nothing happens. Some info on my setup: I followed this guide for Folsom installation: http://docs.openstack.org/folsom/basic-install/content/basic-install-folsom.pdf My novarc file: export OS_TENANT_NAME=admin export OS_USERNAME=admin export OS_PASSWORD=secret export OS_AUTH_URL="http://localhost:5000/v2.0/"; export SERVICE_ENDPOINT="http://localhost:35357/v2.0"; export SERVICE_TOKEN=secret Keystone.log doesn't show anything out of the ordinary. mysql> select id,name from user; +--+-+ | id | name| +--+-+ | 6294f0f5e16d49b9992c211e9cac6855 | admin | | 88f6d25dec6f40f1a9cddf2b20aad4b2 | cinder | | 4c0f5a439a3d44368e8d4af395049ff2 | demo| | c7ee8554321d4fe2bba341a99ee98dc9 | glance | | 520ddc424a30447aa47992400382a167 | nova| | 6e44c5aa1edc4fe5919286ecdda196a0 | quantum | | f4a54be5b3154f85bc4df89a61c5360d | swift | | 6dfec4512b7f4eaa9dc214d7232244fd | viktor | +--+-+ mysql> select * from tenant; +--+++ | id | name | extra | +--+++ | 779c87a9c37f4f63b0fc0aab236a5e1e | demo | {"enabled": true, "description": null} | | 8d4286a5249e43e9aa56864357737884 | invisible_to_admin | {"enabled": true, "description": null} | | cf03bbe4daaf4579bd005a4e64da3334 | MYTENANT | {"enabled": true, "description": null} | | f822dc409d684062b003ba99efa12672 | admin | {"enabled": true, "description": null} | +--+++ I realize dashboard doesn't use a database anymore as it did in Essex. Is that right? BTW, where can I find django log files? Any tips about what is going wrong? Thank you, Viktor ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp