[one-users] showstopper -- can't create VMs after upgrade to OpenNebula 4.4
Hello ONE team, So... after the upgrade to 4.4 OpenNebula on openSuse Linux 12.3 (x86_64), I can no longer create VMs. The log seems to indicate a SSH login-failure (requiring passsword?) Seen below at the end of this message, it's clear that 'oneadmin' user can do passwordless logins... Please let me know what may be going on; the log is not that informative. I did run (as 'oneadmin') the Ruby/ONE clone command by hand, with the results, below... (thanks in advance for any insights - this is a *showstopper*...) oneadmin@sealion:~ cat /var/log/one/61.log | dpaste https://dpaste.de/6I9O/raw oneadmin@sealion:~/vms oneds list | dpaste https://dpaste.de/b0Tj/raw oneadmin@sealion:~ onevm show 61 -x | dpaste https://dpaste.de/mgT3/raw oneadmin@sealion:~ onevm list ID USER GROUPNAMESTAT UCPUUMEM HOST TIME 48 oneadmin oneadmin ttylinux-00 unkn 14192M sealion.in 4d 21h31 49 oneadmin oneadmin Vyatta Router 6 unkn0768M sealion.in 4d 21h27 59 oneadmin oneadmin ttylinxr-0001 fail0 0K 0d 00h00 61 oneadmin oneadmin ttylnx-raw-02 fail0 0K 0d 00h00 oneadmin@sealion:~/vms /var/lib/one/remotes/tm/shared/clone sealion.ine.corp:/var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 sealion.ine.corp:/var/lib/one//datastores/0/61/disk.0 21 | dpaste https://dpaste.de/oZJw/raw -- ssh attempts: all OK oneadmin@sealion:~/vms ssh oneadmin@sealion Last login: Tue Dec 10 10:05:59 2013 from 10.7.1.30 oneadmin@sealion:~ exit logout Connection to sealion closed. oneadmin@sealion:~/vms ssh sealion Last failed login: Tue Dec 10 10:07:00 EST 2013 from sealion.ine.corp on ssh:notty There were 12 failed login attempts since the last successful login. Last login: Tue Dec 10 09:45:29 2013 from 10.7.1.30 sealion.ine.corp:2526 $ exit Connection to sealion closed. oneadmin@sealion:~/vms ssh oneadmin@localhost Last login: Tue Dec 10 10:13:13 2013 from sealion.ine.corp oneadmin@sealion:~ exit logout Connection to localhost closed. oneadmin@sealion:~/vms ssh localhost Last login: Tue Dec 10 10:13:20 2013 from sealion.ine.corp sealion.ine.corp:2527 $ exit Connection to localhost closed ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] Fwd: Re: showstopper -- can't create VMs after upgrade to OpenNebula 4.4
oops... now, to the list! Original Message Subject:Re: [one-users] showstopper -- can't create VMs after upgrade to OpenNebula 4.4 Date: 12/10/2013 11:23 AM From: Mark Biggers mbigg...@ine.commailto:mbigg...@ine.com To: Ruben S. Montero rsmont...@opennebula.orgmailto:rsmont...@opennebula.org On 12/10/2013 11:00 AM, Ruben S. Montero wrote: Hello Ruben, thanks much for the quick! response: So it seems more a problem with a non-existing file. The problem is with cp: cp: cannot create regular file '/var/lib/one/datastores/0/61/disk.0': No such file or directory So can you double check: 1.- All files can be accessed in the hosts /var/lib/one/datastores/0 and /var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 These look fine to me, other than missing folders under datastores (by VM id), when the clone command runs... oneadmin@sealion:~ ll /var/lib/one/datastores/0 /var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 | dpaste https://dpaste.de/2coQ/raw oneadmin@sealion:~ ll -d /var/lib/one/datastores/[012] | dpaste https://dpaste.de/wfR8/raw 2.- If you are using SSH can you check that the datastores are configured properly. 3.- If no progress send the configuration of the datasotres onedatastore show 0 and 1, and vm.log file. Not sure how to do that, here's the configs. I have not touched the Datastores configs, ever - other than by creating VMs, I suppose. oneadmin@sealion:~ oneds list ID NAMESIZE AVAIL CLUSTER IMAGES TYPE DS TM 0 system 22.1G 61% - 0 sys -shared 1 default22.1G 61% - 3 img fs shared 2 files 22.1G 61% - 0 fil fs ssh oneadmin@sealion:~ oneds show 0 | dpaste https://dpaste.de/v53t/raw oneadmin@sealion:~ oneds show 1 | dpaste https://dpaste.de/OIj7/raw oneadmin@sealion:~ oneds show 2 | dpaste https://dpaste.de/WYA6/raw # images templates oneadmin@sealion:~ oneimage show 9 | dpaste https://dpaste.de/Avna/raw oneadmin@sealion:~ onetemplate list | dpaste https://dpaste.de/LUBa/raw oneadmin@sealion:~ onetemplate show 48 | dpaste https://dpaste.de/Z48S/raw oneadmin@sealion:~ onetemplate show 52 | dpaste https://dpaste.de/VH3C/raw On Tue, Dec 10, 2013 at 4:28 PM, Mark Biggers mbigg...@ine.commailto:mbigg...@ine.com wrote: Hello ONE team, So... after the upgrade to 4.4 OpenNebula on openSuse Linux 12.3 (x86_64), I can no longer create VMs. The log seems to indicate a SSH login-failure (requiring passsword?) Seen below at the end of this message, it's clear that 'oneadmin' user can do passwordless logins... Please let me know what may be going on; the log is not that informative. I did run (as 'oneadmin') the Ruby/ONE clone command by hand, with the results, below... (thanks in advance for any insights - this is a *showstopper*...) oneadmin@sealion:~ cat /var/log/one/61.log | dpaste https://dpaste.de/6I9O/raw oneadmin@sealion:~/vms oneds list | dpaste https://dpaste.de/b0Tj/raw oneadmin@sealion:~ onevm show 61 -x | dpaste https://dpaste.de/mgT3/raw oneadmin@sealion:~ onevm list ID USER GROUPNAMESTAT UCPUUMEM HOST TIME 48 oneadmin oneadmin ttylinux-00 unkn 14192M sealion.inhttp://sealion.in 4d 21h31 49 oneadmin oneadmin Vyatta Router 6 unkn0768M sealion.inhttp://sealion.in 4d 21h27 59 oneadmin oneadmin ttylinxr-0001 fail0 0K 0d 00h00 61 oneadmin oneadmin ttylnx-raw-02 fail0 0K 0d 00h00 oneadmin@sealion:~/vms /var/lib/one/remotes/tm/shared/clone sealion.ine.corp:/var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 sealion.ine.corp:/var/lib/one//datastores/0/61/disk.0 21 | dpaste https://dpaste.de/oZJw/raw -- ssh attempts: all OK oneadmin@sealion:~/vms ssh oneadmin@sealion Last login: Tue Dec 10 10:05:59 2013 from 10.7.1.30 oneadmin@sealion:~ exit logout Connection to sealion closed. oneadmin@sealion:~/vms ssh sealion Last failed login: Tue Dec 10 10:07:00 EST 2013 from sealion.ine.corp on ssh:notty There were 12 failed login attempts since the last successful login. Last login: Tue Dec 10 09:45:29 2013 from 10.7.1.30 sealion.ine.corp:2526 $ exit Connection to sealion closed. oneadmin@sealion:~/vms ssh oneadmin@localhost Last login: Tue Dec 10 10:13:13 2013 from sealion.ine.corp oneadmin@sealion:~ exit logout Connection to localhost closed. oneadmin@sealion:~/vms ssh localhost Last login: Tue Dec 10 10:13:20 2013 from sealion.ine.corp sealion.ine.corp:2527 $ exit Connection to localhost closed ___ Users mailing list Users@lists.opennebula.orgmailto:Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- -- Ruben
Re: [one-users] Fwd: Re: showstopper -- can't create VMs after upgrade to OpenNebula 4.4
Already did all that... sealion is just an alias in /etc/hosts, for sealion.ine.corp. Perhaps this StackEx note, is related to my issue?? With the ONE (ruby) command clone? http://unix.stackexchange.com/questions/37884/how-to-copy-with-scp-between-two-servers-using-key-auth Here's my failure, on another attempt -- onevm delete --recreate 62 Tue Dec 10 17:32:30 2013 [TM][E]: clone: Command scp -r sealion.ine.corp:/var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 sealion.ine.corp:/var/lib/one//datastores/0/62/disk.0 failed: Permission denied (publickey,keyboard-interactive). Tue Dec 10 17:32:30 2013 [TM][E]: Error copying sealion.ine.corp:/var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 to sealion.ine.corp:/var/lib/one//datastores/0/62/disk.0 Tue Dec 10 17:32:30 2013 [TM][I]: ExitCode: 1 Tue Dec 10 17:32:30 2013 [TM][E]: Error executing image transfer script: Error copying sealion.ine.corp:/var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 to sealion.ine.corp:/var/lib/one//datastores/0/62/disk.0 Tue Dec 10 17:32:31 2013 [DiM][I]: New VM state is FAILED Should the clone ing create a *source hostname* for the scp command (as above), and why? How would the ssh pubkey authentication work, in this case?? (If I drop the source hostname in the scp command above, the scp works fine...) Thank you for your assistance, mark On 12/10/2013 05:27 PM, Ruben S. Montero wrote: Ok, I was mislead by the manual clone output and that of vm.log. So yes it is a permission problem. In your tests you are using oneadmin@sealion:~/vms ssh oneadmin@sealion BUT the name of the host is sealion.ine.corp Can you double check with the same hostname, just in case you are not facing a hostname resolution issue. Cheers On Tue, Dec 10, 2013 at 8:25 PM, Mark Biggers mbigg...@ine.commailto:mbigg...@ine.com wrote: oops... now, to the list! Original Message Subject:Re: [one-users] showstopper -- can't create VMs after upgrade to OpenNebula 4.4 Date: 12/10/2013 11:23 AM From: Mark Biggers mbigg...@ine.commailto:mbigg...@ine.com To: Ruben S. Montero rsmont...@opennebula.orgmailto:rsmont...@opennebula.org On 12/10/2013 11:00 AM, Ruben S. Montero wrote: Hello Ruben, thanks much for the quick! response: So it seems more a problem with a non-existing file. The problem is with cp: cp: cannot create regular file '/var/lib/one/datastores/0/61/disk.0': No such file or directory So can you double check: 1.- All files can be accessed in the hosts /var/lib/one/datastores/0 and /var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 These look fine to me, other than missing folders under datastores (by VM id), when the clone command runs... oneadmin@sealion:~ ll /var/lib/one/datastores/0 /var/lib/one/datastores/1/20288f3556d15e65b1dfdc540a450226 | dpaste https://dpaste.de/2coQ/raw oneadmin@sealion:~ ll -d /var/lib/one/datastores/[012] | dpaste https://dpaste.de/wfR8/raw 2.- If you are using SSH can you check that the datastores are configured properly. 3.- If no progress send the configuration of the datasotres onedatastore show 0 and 1, and vm.log file. Not sure how to do that, here's the configs. I have not touched the Datastores configs, ever - other than by creating VMs, I suppose. oneadmin@sealion:~ oneds list ID NAMESIZE AVAIL CLUSTER IMAGES TYPE DS TM 0 system 22.1G 61% - 0 sys -shared 1 default22.1G 61% - 3 img fs shared 2 files 22.1G 61% - 0 fil fs ssh oneadmin@sealion:~ oneds show 0 | dpaste https://dpaste.de/v53t/raw oneadmin@sealion:~ oneds show 1 | dpaste https://dpaste.de/OIj7/raw oneadmin@sealion:~ oneds show 2 | dpaste https://dpaste.de/WYA6/raw # images templates oneadmin@sealion:~ oneimage show 9 | dpaste https://dpaste.de/Avna/raw oneadmin@sealion:~ onetemplate list | dpaste https://dpaste.de/LUBa/raw oneadmin@sealion:~ onetemplate show 48 | dpaste https://dpaste.de/Z48S/raw oneadmin@sealion:~ onetemplate show 52 | dpaste https://dpaste.de/VH3C/raw On Tue, Dec 10, 2013 at 4:28 PM, Mark Biggers mbigg...@ine.commailto:mbigg...@ine.com wrote: Hello ONE team, So... after the upgrade to 4.4 OpenNebula on openSuse Linux 12.3 (x86_64), I can no longer create VMs. The log seems to indicate a SSH login-failure (requiring passsword?) Seen below at the end of this message, it's clear that 'oneadmin' user can do passwordless logins... Please let me know what may be going on; the log is not that informative. I did run (as 'oneadmin') the Ruby/ONE clone command by hand, with the results, below... (thanks in advance for any insights - this is a *showstopper*...) oneadmin@sealion:~ cat /var/log/one/61.log | dpaste https://dpaste.de/6I9O/raw
[one-users] update... Re: (RESEND) need to create Flows? for openvswitch-based ONE (4.2) setup -- (passed on ebtables)
Hello Valetin and ONE team: I have done some work to make the TTYLinux image (from ONE template) fully boot, in addition to having the Marketplace Vyatta router boot OK. They can ping and SSH to each other, on their 10.0.0.0/24 network. They cannot see any external network. What I am attempting to do is have at least complete network-routing, from the external network, to the VMs internal network, and back again. I have made little progress, so any steps to help make this possible, are greatly appreciated. (I am currently looking at establishing OpenFlow flows, across the OpenVSwitch, to route packets... not sure about that, even). Here's what I have, now: oneadmin@sealion:~ onehost show 8 | dpaste https://dpaste.de/5hd5/raw oneadmin@sealion:~ onevm show --all 48 | dpaste https://dpaste.de/OEmz/raw oneadmin@sealion:~ onevm show --all 49 | dpaste https://dpaste.de/LW8n/raw oneadmin@sealion:~ onevnet show --all 10 | dpaste https://dpaste.de/nupU.raw # the overview: oneadmin@sealion:~ onevnet show 10 VIRTUAL NETWORK 10 INFORMATION ID : 10 NAME : ovsnet_02 USER : oneadmin GROUP : oneadmin CLUSTER: - TYPE : RANGED BRIDGE : vbr0 VLAN : Yes VLAN ID: 1 USED LEASES: 2 PERMISSIONS OWNER : um- GROUP : --- OTHER : --- VIRTUAL NETWORK TEMPLATE DNS=10.0.0.1 GATEWAY=10.0.0.1 NETWORK_ADDRESS=10.0.0.0/24 NETWORK_MASK=255.255.255.0 RANGE IP_START : 10.0.0.1 IP_END : 10.0.0.254 USED LEASES LEASE=[ MAC=02:00:0a:00:00:02, IP=10.0.0.2, IP6_LINK=fe80::400:aff:fe00:2, USED=1, VID=48 ] LEASE=[ MAC=02:00:0a:00:00:03, IP=10.0.0.3, IP6_LINK=fe80::400:aff:fe00:3, USED=1, VID=49 ] VIRTUAL MACHINES ID USER GROUPNAMESTAT UCPUUMEM HOST TIME 48 oneadmin oneadmin ttylinux-00 runn 14192M sealion.in 0d 22h11 49 oneadmin oneadmin Vyatta Router 6 runn0768M sealion.in 0d 22h07 ## the current OpenVSwitch setup (this needs to be a Gist!)The external network is now 10.7.1.0/24 # sealion.ine.corp:2396 $ cat boot-fscking-br0.sh | dpaste https://dpaste.de/YO5O oneadmin@sealion:~ netstat -nr | dpaste https://dpaste.de/mUUK/raw oneadmin@sealion:~ BRIDGE_DEV=vbr0 oneadmin@sealion:~ sudo ovs-vsctl show | dpaste https://dpaste.de/toFu/raw oneadmin@sealion:~ sudo ovs-ofctl show $BRIDGE_DEV | dpaste https://dpaste.de/nLAy/raw oneadmin@sealion:~ sudo ovs-ofctl dump-flows $BRIDGE_DEV | dpaste https://dpaste.de/m4PT/raw Would it be advantageous to install another Ethernet port on my W-530 notebook, and have that be the OpenVSwitch hardware port? thank you! (MarkB1 on #opennebula, freenode.net) On 11/22/2013 07:47 AM, Valentin Bud wrote: Hello Mark, On Thu, Nov 21, 2013 at 01:01:17PM -0600, Mark Biggers wrote: Hello Valentin, thanks for the reply. On 11/21/2013 03:30 AM, Valentin Bud wrote: Hello Mark, Before pointing you to the problem I think your config has you should also check that you have routing enabled in the machine Are you speaking of ip route routes? Or some sysconf variable? Or, route(s) on the VMs themselves?? I was actually speaking about IP Forwarding, to be precise. That is attained via a sysctl. It should be set to 1. To make the changes persistent after reboot you have /etc/sysctl.conf. -- Valentin Bud http://databus.pro | valen...@databus.promailto:valen...@databus.pro ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] need to create Flows? for openvswitch-based ONE (4.2) setup -- (passed on ebtables)
Hello ONE team, I have passed on the ebtables configuration for networking in 4.2 ONE. We'll need OpenVSwitch anyways to manage the VMs VLANS, so I have moved on. I *think* I have an almost working OpenVSwitch configuration. Must I manually create flows for each VM/MAC-addr to enable IP traffic, across the OVS vbridge (vbr0), in this case? The info on my new (OVS networking) setup, is included, at the end of this message. Thank you. (The platform is still openSUSE 12.3 on a Thinkpad W530...) On 11/19/2013 05:43 AM, Jaime Melis wrote: Hi Mark, I have the feeling the NAT policies are interfering with this. Can you try without applying NAT rules? On Wed, Nov 13, 2013 at 9:08 PM, Mark Biggers mbigg...@ine.commailto:mbigg...@ine.com wrote: The subject says it all. I am available on IRC -- see my signature, and Google chat. I can get no networking across a bridge working, for the ONE ebtables model. === edited out -- Jaime Melis Project Engineer OpenNebula - Flexible Enterprise Cloud Made Simple www.OpenNebula.orghttp://www.OpenNebula.org | jme...@opennebula.orgmailto:jme...@opennebula.org Script started on Wed Nov 20 16:27:05 2013 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # netstat -nr Kernel IP routing table DestinationGatewayGenmaskFlagsMSS Window irtt Iface 0.0.0.0192.168.1.10.0.0.0UG 0 0 0 vbr0 67.139.46.149192.168.1.1255.255.255.255 UGH 0 0 0 vbr0 127.0.0.00.0.0.0255.255.255.0U 0 0 0 lo 127.0.0.00.0.0.0255.0.0.0U 0 0 0 lo 192.168.1.00.0.0.0255.255.255.0U 0 0 0 vbr0 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # ip addr 1: lo: LOOPBACK,UP,LOWER_UP mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo 2: eth0: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 3c:97:0e:ab:0a:de brd ff:ff:ff:ff:ff:ff 3: wlan0: BROADCAST,MULTICAST mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 6c:88:14:da:0b:44 brd ff:ff:ff:ff:ff:ff 4: ovs-system: BROADCAST,MULTICAST mtu 1500 qdisc noop state DOWN link/ether 0a:0e:fd:bb:5a:8a brd ff:ff:ff:ff:ff:ff 7: vbr0: BROADCAST,PROMISC,UP,LOWER_UP mtu 1500 qdisc noqueue state UNKNOWN link/ether 3c:97:0e:ab:0a:de brd ff:ff:ff:ff:ff:ff inet 192.168.1.250/24 scope global vbr0 12: vnet0: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 500 link/ether fe:00:0a:00:00:03 brd ff:ff:ff:ff:ff:ff 13: vnet1: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 500 link/ether fe:00:0a:00:00:04 brd ff:ff:ff:ff:ff:ff r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # BRIDGE_DEV=vbr0 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # sudo ovs-ofctl dump-desc $BRIDGE_DEV OFPST_DESC reply (xid=0x2): Manufacturer: Nicira, Inc. Hardware: Open vSwitch Software: 1.11.0 Serial Num: None DP Description: None r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # sudo ovs-vsctl show 001119d6-32d7-4db8-8015-229b271cca6a Bridge vbr0 Controller ptcp: fail_mode: standalone Port vnet0 tag: 0 Interface vnet0 Port vnet1 tag: 0 Interface vnet1 Port eth0 Interface eth0 Port vbr0 Interface vbr0 type: internal ovs_version: 1.11.0 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # sudo ovs-ofctl show $BRIDGE_DEV OFPT_FEATURES_REPLY (xid=0x2): dpid:3c970eab0ade n_tables:254, n_buffers:256 capabilities: FLOW_STATS TABLE_STATS PORT_STATS QUEUE_STATS ARP_MATCH_IP actions: OUTPUT SET_VLAN_VID SET_VLAN_PCP STRIP_VLAN SET_DL_SRC SET_DL_DST SET_NW_SRC SET_NW_DST SET_NW_TOS SET_TP_SRC SET_TP_DST ENQUEUE 1(eth0): addr:3c:97:0e:ab:0a:de config: 0 state: STP_FORWARD current: 1GB-FD COPPER AUTO_NEG advertised: 10MB-HD 10MB-FD 100MB-HD 100MB-FD 1GB-FD COPPER AUTO_NEG supported: 10MB-HD 10MB-FD 100MB-HD 100MB-FD 1GB-FD COPPER AUTO_NEG speed: 1000 Mbps now, 1000 Mbps max 2(vnet0): addr:fe:00:0a:00:00:03 config: 0 state: 0 current: 10MB-FD COPPER speed: 10 Mbps now, 0 Mbps max 3(vnet1): addr:fe:00:0a:00:00:04 config: 0 state: 0 current: 10MB-FD COPPER speed: 10 Mbps now, 0 Mbps max LOCAL(vbr0): addr:3c:97:0e:ab:0a:de config: 0 state: 0 speed: 0 Mbps now, 0 Mbps max OFPT_GET_CONFIG_REPLY (xid=0x4): frags=normal miss_send_len=0 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # sudo ovs-ofctl dump-flows $BRIDGE_DEV NXST_FLOW reply (xid=0x4): cookie=0x0, duration=8382.092s, table=0, n_packets=4, n_bytes=240, idle_age=8381, priority=4,in_port=2,dl_src=02:00:0a:00:00:03 actions=NORMAL cookie=0x0
[one-users] (RESEND) need to create Flows? for openvswitch-based ONE (4.2) setup -- (passed on ebtables)
Hello ONE team,(RESENT -- dropped in bad onevm show NN data, earlier -- corrected, below!) I have passed on the ebtables configuration for networking in 4.2 ONE. We'll need OpenVSwitch anyways to manage the VMs VLANS, so I have moved on. I *think* I have an almost working OpenVSwitch configuration. Must I manually create flows for each VM/MAC-addr to enable IP traffic, across the OVS vbridge (vbr0), in this case? The info on my new (OVS networking) setup, is included, at the end of this message. Thank you. (The platform is still openSUSE 12.3 on a Thinkpad W530...) On 11/19/2013 05:43 AM, Jaime Melis wrote: Hi Mark, I have the feeling the NAT policies are interfering with this. Can you try without applying NAT rules? On Wed, Nov 13, 2013 at 9:08 PM, Mark Biggers mbigg...@ine.commailto:mbigg...@ine.com wrote: The subject says it all. I am available on IRC -- see my signature, and Google chat. I can get no networking across a bridge working, for the ONE ebtables model. === edited out -- Jaime Melis Project Engineer OpenNebula - Flexible Enterprise Cloud Made Simple www.OpenNebula.orghttp://www.OpenNebula.org | jme...@opennebula.orgmailto:jme...@opennebula.org Script started on Wed Nov 20 16:27:05 2013 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # netstat -nr Kernel IP routing table DestinationGatewayGenmaskFlagsMSS Window irtt Iface 0.0.0.0192.168.1.10.0.0.0UG 0 0 0 vbr0 67.139.46.149192.168.1.1255.255.255.255 UGH 0 0 0 vbr0 127.0.0.00.0.0.0255.255.255.0U 0 0 0 lo 127.0.0.00.0.0.0255.0.0.0U 0 0 0 lo 192.168.1.00.0.0.0255.255.255.0U 0 0 0 vbr0 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # ip addr 1: lo: LOOPBACK,UP,LOWER_UP mtu 65536 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo 2: eth0: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 3c:97:0e:ab:0a:de brd ff:ff:ff:ff:ff:ff 3: wlan0: BROADCAST,MULTICAST mtu 1500 qdisc noop state DOWN qlen 1000 link/ether 6c:88:14:da:0b:44 brd ff:ff:ff:ff:ff:ff 4: ovs-system: BROADCAST,MULTICAST mtu 1500 qdisc noop state DOWN link/ether 0a:0e:fd:bb:5a:8a brd ff:ff:ff:ff:ff:ff 7: vbr0: BROADCAST,PROMISC,UP,LOWER_UP mtu 1500 qdisc noqueue state UNKNOWN link/ether 3c:97:0e:ab:0a:de brd ff:ff:ff:ff:ff:ff inet 192.168.1.250/24 scope global vbr0 12: vnet0: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 500 link/ether fe:00:0a:00:00:03 brd ff:ff:ff:ff:ff:ff 13: vnet1: BROADCAST,MULTICAST,UP,LOWER_UP mtu 1500 qdisc pfifo_fast state UNKNOWN qlen 500 link/ether fe:00:0a:00:00:04 brd ff:ff:ff:ff:ff:ff r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # BRIDGE_DEV=vbr0 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # sudo ovs-ofctl dump-desc $BRIDGE_DEV OFPST_DESC reply (xid=0x2): Manufacturer: Nicira, Inc. Hardware: Open vSwitch Software: 1.11.0 Serial Num: None DP Description: None r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # sudo ovs-vsctl show 001119d6-32d7-4db8-8015-229b271cca6a Bridge vbr0 Controller ptcp: fail_mode: standalone Port vnet0 tag: 0 Interface vnet0 Port vnet1 tag: 0 Interface vnet1 Port eth0 Interface eth0 Port vbr0 Interface vbr0 type: internal ovs_version: 1.11.0 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # sudo ovs-ofctl show $BRIDGE_DEV OFPT_FEATURES_REPLY (xid=0x2): dpid:3c970eab0ade n_tables:254, n_buffers:256 capabilities: FLOW_STATS TABLE_STATS PORT_STATS QUEUE_STATS ARP_MATCH_IP actions: OUTPUT SET_VLAN_VID SET_VLAN_PCP STRIP_VLAN SET_DL_SRC SET_DL_DST SET_NW_SRC SET_NW_DST SET_NW_TOS SET_TP_SRC SET_TP_DST ENQUEUE 1(eth0): addr:3c:97:0e:ab:0a:de config: 0 state: STP_FORWARD current: 1GB-FD COPPER AUTO_NEG advertised: 10MB-HD 10MB-FD 100MB-HD 100MB-FD 1GB-FD COPPER AUTO_NEG supported: 10MB-HD 10MB-FD 100MB-HD 100MB-FD 1GB-FD COPPER AUTO_NEG speed: 1000 Mbps now, 1000 Mbps max 2(vnet0): addr:fe:00:0a:00:00:03 config: 0 state: 0 current: 10MB-FD COPPER speed: 10 Mbps now, 0 Mbps max 3(vnet1): addr:fe:00:0a:00:00:04 config: 0 state: 0 current: 10MB-FD COPPER speed: 10 Mbps now, 0 Mbps max LOCAL(vbr0): addr:3c:97:0e:ab:0a:de config: 0 state: 0 speed: 0 Mbps now, 0 Mbps max OFPT_GET_CONFIG_REPLY (xid=0x4): frags=normal miss_send_len=0 r...@sealion.ine.corp:onemailto:r...@sealion.ine.corp:one # sudo ovs-ofctl dump-flows $BRIDGE_DEV NXST_FLOW reply (xid=0x4): cookie=0x0, duration=8382.092s, table=0, n_packets=4, n_bytes=240, idle_age=8381
[one-users] ttylinux image from Marketplace fails to boot...
... while a Marketplace Vyatta V65R1 Router (KVM) image, boots just fine.** Here are the details, below. oneadmin onevnet show 8 VIRTUAL NETWORK 8 INFORMATION ID : 8 NAME : cloud USER : oneadmin GROUP : oneadmin CLUSTER: - TYPE : RANGED BRIDGE : br0 VLAN : Yes VLAN ID: vnet_default USED LEASES: 2 PERMISSIONS OWNER : um- GROUP : --- OTHER : --- VIRTUAL NETWORK TEMPLATE DNS=192.168.1.1 GATEWAY=192.168.122.1 ID=3 NETMASK=255.255.255.0 NETWORK_ADDRESS=192.168.122.0/24 NETWORK_MASK=255.255.255.0 RANGE IP_START : 192.168.122.1 IP_END : 192.168.122.254 USED LEASES LEASE=[ MAC=02:00:c0:a8:7a:02, IP=192.168.122.2, IP6_LINK=fe80::400:c0ff:fea8:7a02, USED=1, VID=27 ] LEASE=[ MAC=02:00:c0:a8:7a:03, IP=192.168.122.3, IP6_LINK=fe80::400:c0ff:fea8:7a03, USED=1, VID=29 ] VIRTUAL MACHINES ID USER GROUPNAMESTAT UCPUUMEM HOST TIME 27 oneadmin oneadmin vyatta-router runn0 1024M host01 0d 02h25 29 oneadmin oneadmin TTY Linux NOW runn 99256M host01 0d 01h52 oneadmin onevm show 29 VIRTUAL MACHINE 29 INFORMATION ID : 29 NAME: TTY Linux NOW USER: oneadmin GROUP : oneadmin STATE : ACTIVE LCM_STATE : RUNNING RESCHED : No HOST: host01 START TIME : 11/13 11:35:22 END TIME: - DEPLOY ID : one-29 VIRTUAL MACHINE MONITORING USED MEMORY : 256M USED CPU: 99 NET_TX : 0K NET_RX : 0K PERMISSIONS OWNER : um- GROUP : --- OTHER : --- VM DISKS ID TARGET IMAGE TYPE SAVE SAVE_AS 0 vdattylinux - kvm file NO - VM NICS ID NETWORK VLAN BRIDGE IP MAC 0 cloud yes br0 192.168.122.3 02:00:c0:a8:7a:03 fe80::400:c0ff:fea8:7a03 VIRTUAL MACHINE HISTORY SEQ HOSTACTION REAS STARTTIME PROLOG 0 host01 none none 11/13 11:35:30 0d 01h49m 0h00m01s USER TEMPLATE SCHED_REQUIREMENTS=ID=\5\ VIRTUAL MACHINE TEMPLATE CONTEXT=[ DISK_ID=1, ETH0_DNS=192.168.1.1, ETH0_GATEWAY=192.168.122.1, ETH0_IP=192.168.122.3, ETH0_MASK=255.255.255.0, ETH0_NETWORK=192.168.122.0/24, NETWORK=YES, SSH_PUBLIC_KEY=ssh-rsa ... , TARGET=vdb ] CPU=0.4 GRAPHICS=[ LISTEN=0.0.0.0, PORT=5929, TYPE=VNC ] MEMORY=256 OS=[ ARCH=i686, BOOT=hd ] TEMPLATE_ID=27 VMID=29 ** That Vyatta VR VM gets no working route to the br0 bridge (another issue), and is not pingable at its address from the ONE host (combined host/node, developer desktop). Should I just forget about getting even basic ONE network (ebtables) working, on openSUSE 12.3? I am getting nowhere with it... I Am Not A Network Engineer. The ONE docs assume a lot of Linux NetEngr background, and/or have some critical details missing. -- I will send a follow-up email on this, I need spoon-feeding, sorry. thank you. Mark Biggers mbigg...@ine.com ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] *no go*, using Linux br0 to contact ONE VMs -- (was) RE: VM networking configuration CentOS
Yet another Mark, here: This is perhaps a related question. I completely missed the CONTEXT = [ NETWORK = YES ] in my ONE templates, and thereby was screwed. Fixed, thanks for this ( this needs to be **better documented**, please... ) However, I still can't get the ebtables / brN (Linux bridging) to work, for ping / ssh / VNC going , to my VM network 192.168.122.0. Notes below. Need the ONE team's help, thank you. Here's how I configured the br0 bridge, on openSUSE 12.3 Linux : sudo ifdown br0 || echo no way to down bridge br0 :) sudo brctl delbr br0 || echo no bridge br0 :) sudo ifconfig eth0 promisc up sudo brctl show sudo brctl addbr br0 sudo brctl show sudo ifconfig eth0 0.0.0.0 up sudo brctl addif br0 eth0 sudo brctl show sudo ifconfig br0 192.168.1.250 # previous addr of eth0 sudo route add default gw 192.168.1.1 sudo route add -net 192.168.122.0 gw 192.168.1.250 netmask 255.255.255.0 supporting info = r...@sealion.ine.corp:~ # ip addr | dpaste http://dpaste.de/csGz oneadmin@sealion:~ onevnet show -x 4 | dpaste http://dpaste.de/594n oneadmin@sealion:~ onetemplate show 18 -x | dpaste http://dpaste.de/OIya oneadmin@sealion:~ onevm show 20 -x | dpaste http://dpaste.de/pXEZ r...@sealion.ine.corp:~ # cat typescript | dpaste ## pinging, netstats http://dpaste.de/9OwD mark From: users-boun...@lists.opennebula.org [users-boun...@lists.opennebula.org] On Behalf Of Jaime Melis [jme...@opennebula.org] Sent: Wednesday, November 06, 2013 10:50 AM To: Mark Kusch Cc: Users OpenNebula Subject: Re: [one-users] VM networking configuration CentOS Hi Mark, I think the problem is that you are not using the CONTEXT = [ NETWORK = YES ] option. If you don't, the VM will infer its IP from the MAC, converting the last 4 bytes of the MAC to decimal: MAC=54:52:00:c0:de:0a, = 00 . c0 . de . 0a = 0.192.222.10 If you add the CONTEXT bit from before to your template everything should work OK. cheers, Jaime On Mon, Oct 21, 2013 at 4:14 PM, Mark Kusch mark.ku...@silpion.demailto:mark.ku...@silpion.de wrote: Hi list! Problem solved... Still it FEELS like NetworkManager broke networking in the VMs, still not confirmed. ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
[one-users] my DevOps install notes for openSUSE 12.3 on Github --- (WAS: First post - Error on New install of Open Nebula on OpenSuSE 12.3)
Although my attempt at configuration of ONE 4.2 on openSUSE 12.3 still does not work for me (VMs don't instantiate), here's my contribution, in a restructuredText document. https://gist.github.com/biggers/7251574 Perhaps this will help others, trying to do what I am attempting to do -- just bring up ONE 4.2 for a Developer/DevOps Cloud + Host node. In my case, I must write some Cloud automation code in Python, against the OCCI APIs. Any changes or comments appreciated! Thank you. mark Mark Biggers INE, Inc 2530 Meridian Parkway, Suite 2061 Durham, NC 27713 Internets: Freenode.net IRC: markb1, #trizpug, #trilug (now #opennebula) ..Google chat: mbiggers.01 ATT gmail.com ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] my DevOps install notes for openSUSE 12.3 on Github --- (WAS: First post - Error on New install of Open Nebula on OpenSuSE 12.3)
Thanks Ruben, I appreciate the help. Here's what I have (see the dpaste.de links). It's consistent, regardless of any VM image that is instantiated ... cat /var/log/13.log | dpaste # http://dpaste.de/5yts oneadmin@sealion:~ onevm show 13 --all -x | dpaste http://dpaste.de/y1XS cat /var/log/libvirt/qemu/one-13.log | dpaste http://dpaste.de/Zpr5 mark From: Ruben S. Montero [rsmont...@opennebula.org] Sent: Thursday, October 31, 2013 12:30 PM To: Mark Biggers Cc: users@lists.opennebula.org Subject: Re: [one-users] my DevOps install notes for openSUSE 12.3 on Github --- (WAS: First post - Error on New install of Open Nebula on OpenSuSE 12.3) Hi Mike Thank you very much for sharing, great doc! Could we get more details about still does not work for me (VMs don't instantiate)? So we can clear any issue and complete the document? Thanks Ruben On Thu, Oct 31, 2013 at 4:36 PM, Mark Biggers mbigg...@ine.commailto:mbigg...@ine.com wrote: Although my attempt at configuration of ONE 4.2 on openSUSE 12.3 still does not work for me (VMs don't instantiate), here's my contribution, in a restructuredText document. https://gist.github.com/biggers/7251574 Perhaps this will help others, trying to do what I am attempting to do -- just bring up ONE 4.2 for a Developer/DevOps Cloud + Host node. In my case, I must write some Cloud automation code in Python, against the OCCI APIs. Any changes or comments appreciated! Thank you. mark Mark Biggers INE, Inc 2530 Meridian Parkway, Suite 2061 Durham, NC 27713 Internets: Freenode.net IRC: markb1, #trizpug, #trilug (now #opennebula) ..Google chat: mbiggers.01 ATT gmail.com ___ Users mailing list Users@lists.opennebula.orgmailto:Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org -- -- Ruben S. Montero, PhD Project co-Lead and Chief Architect OpenNebula - Flexible Enterprise Cloud Made Simple www.OpenNebula.orghttp://www.OpenNebula.org | rsmont...@opennebula.orgmailto:rsmont...@opennebula.org | @OpenNebula ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] my DevOps install notes for openSUSE 12.3 on Github --- (WAS: First post - Error on New install of Open Nebula on OpenSuSE 12.3)
Hello Ruben:
** Yow, I should have seen this one! ( this .conf had OS[ KERNEL = /vmlinuz ] )
cd /etc/one/vmm_exec
diff -ub vmm_exec_kvm.conf.dist vmm_exec_kvm.conf | dpaste
SEE http://dpaste.de/Q4HI
** Had a failure to VM instantiate (onevm delete --recreate), due to ebtables
path:
# Note: hard-coded paths in this Ruby script, to Linux networking tools...
(from grep) /var/lib/one/remotes/vnm/OpenNebulaNetwork.rb
31 -
32 - COMMANDS = {
33 : :ebtables = sudo /sbin/ebtables,
34 + :iptables = sudo /sbin/iptables,
35 + :brctl= sudo /sbin/brctl,
!! WORKAROUND
cd /usr/sbin; for cmd in ebtables iptables; do sudo ln -s /usr/sbin/$cmd
/sbin/; done
** My first VM is instantiated, and running. Now to fix the missing
VNC-console proxy, for my VM (as seen from Sunstone UI)
thank you,
mark
Mark Biggers http://www.ine.com
From: Ruben S. Montero [rsmont...@opennebula.org]
Sent: Thursday, October 31, 2013 3:07 PM
To: Mark Biggers
Cc: users@lists.opennebula.org
Subject: Re: [one-users] my DevOps install notes for openSUSE 12.3 on Github
--- (WAS: First post - Error on New install of Open Nebula on OpenSuSE 12.3)
Hi Mark
It seems that your domain is trying to boot from /vmlinuz that cannot be found
in the host. This primitive is not in your original template as shown in the
template of VM 13 (there is no OSKERNEL).
So this is most probably coming from the driver configuration file. Could you
check (and remove) /etc/vmm_exec/vmm_exec_kvm.conf for a line setting KERNEL
to /vmlinuz?
Cheers
Ruben
On Thu, Oct 31, 2013 at 5:57 PM, Mark Biggers
mbigg...@ine.commailto:mbigg...@ine.com wrote:
Thanks Ruben, I appreciate the help. Here's what I have (see the
dpaste.dehttp://dpaste.de links). It's consistent, regardless of any VM
image that is instantiated ...
cat /var/log/13.log | dpaste
# http://dpaste.de/5yts
oneadmin@sealion:~ onevm show 13 --all -x | dpaste
http://dpaste.de/y1XS
cat /var/log/libvirt/qemu/one-13.log | dpaste
http://dpaste.de/Zpr5
mark
From: Ruben S. Montero
[rsmont...@opennebula.orgmailto:rsmont...@opennebula.org]
Sent: Thursday, October 31, 2013 12:30 PM
To: Mark Biggers
Cc: users@lists.opennebula.orgmailto:users@lists.opennebula.org
Subject: Re: [one-users] my DevOps install notes for openSUSE 12.3 on Github
--- (WAS: First post - Error on New install of Open Nebula on OpenSuSE 12.3)
Hi Mike
Thank you very much for sharing, great doc!
Could we get more details about still does not work for me (VMs don't
instantiate)? So we can clear any issue and complete the document?
Thanks
Ruben
On Thu, Oct 31, 2013 at 4:36 PM, Mark Biggers
mbigg...@ine.commailto:mbigg...@ine.commailto:mbigg...@ine.commailto:mbigg...@ine.com
wrote:
Although my attempt at configuration of ONE 4.2 on openSUSE 12.3 still does not
work for me (VMs don't instantiate), here's my contribution, in a
restructuredText document.
https://gist.github.com/biggers/7251574
Perhaps this will help others, trying to do what I am attempting to do -- just
bring up ONE 4.2 for a Developer/DevOps Cloud + Host node. In my case, I
must write some Cloud automation code in Python, against the OCCI APIs.
Any changes or comments appreciated! Thank you.
mark
Mark Biggers
INE, Inc
2530 Meridian Parkway, Suite 2061
Durham, NC 27713
Internets: Freenode.net IRC: markb1, #trizpug, #trilug (now #opennebula)
..Google chat: mbiggers.01 ATT
gmail.comhttp://gmail.com
___
Users mailing list
Users@lists.opennebula.orgmailto:Users@lists.opennebula.orgmailto:Users@lists.opennebula.orgmailto:Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
--
--
Ruben S. Montero, PhD
Project co-Lead and Chief Architect
OpenNebula - Flexible Enterprise Cloud Made Simple
www.OpenNebula.orghttp://www.OpenNebula.org |
rsmont...@opennebula.orgmailto:rsmont...@opennebula.org | @OpenNebula
___
Users mailing list
Users@lists.opennebula.org
http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
Re: [one-users] First post - Error on New install of Open Nebula on OpenSuSE 12.3 -- here's the Template-dump
Hello Carlos and all. Luc Bell has been posting, for me. Here's the dump of my latest onetemplate instantiate attempt.Not sure who's maintaining the openSUSE 12.3 packages, but I have had to to a *lot* of configuration, with ongoing trial error (since Docs vary a bit), just to attempt to bring up a working DevOps Cloud host on a Thinkpad W530 http://dpaste.de/bTkZ/raw In my very humble opinion, there needs to be a DevOps ONE 4.x workbook with a completely worked out, tested scenario -- with variations as needed, for at least Ubuntu, Fedora and openSUSE (OK, maybe also CentOS 6.x, but that has very old package releases, for a Developer...) thank you, mark --- Hi, What OpenNebula version are you using? Can you paste the contents of your VM template? Use the 'onevm show --all' command so it prints all internal attributes. Regards -- Carlos MartÃn, MSc Project Engineer OpenNebula - Flexible Enterprise Cloud Made Simple www.OpenNebula.org | cmartin at opennebula.org | @OpenNebulahttp://twitter.com/opennebulacmartin at opennebula.org On Tue, Oct 29, 2013 at 3:23 PM, Luc Bell lucrbell at gmail.com wrote: Hello, my colleague is getting the following error after implementing OpenNebula on OpenSuse12.3 when attempting to provision a VM via KVM. We double checked that these params are correct, but still get the error and the VM does not provision. It fails with this error: 1. Wed Oct 23 17:36:31 2013 [DiM][I]: New VM state is DONE. 2. Fri Oct 25 10:22:38 2013 [DiM][I]: New VM state is ACTIVE. 3. Fri Oct 25 10:22:38 2013 [LCM][I]: New VM state is PROLOG. 4. Fri Oct 25 10:22:38 2013 [TM][E]: missing DISK mandatory attributes (SOURCE, TM_MAD, CLONE, DATASTORE_ID) for VM 4, DISK 0 5. Fri Oct 25 10:22:38 2013 [DiM][I]: New VM state is FAILED 6. Fri Oct 25 10:29:44 2013 [DiM][I]: New VM state is CLEANUP. 7. Fri Oct 25 10:29:45 2013 [DiM][I]: New VM state is PENDING 8. Fri Oct 25 10:30:08 2013 [DiM][I]: New VM state is ACTIVE. 9. Fri Oct 25 10:30:08 2013 [LCM][I]: New VM state is PROLOG. 10. Fri Oct 25 10:30:08 2013 [TM][E]: missing DISK mandatory attributes (SOURCE, TM_MAD, CLONE, DATASTORE_ID) for VM 4, DISK 0 11. Fri Oct 25 10:30:08 2013 [DiM][I]: New VM state is FAILED 12. ~ Thank you in advance for any guidance. ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org - Mark Biggers INE.com 2530 Meridian Parkway, Suite 2061 (temp) Durham, NC 27713 ___ Users mailing list Users@lists.opennebula.org http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
