Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2016-12-16

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:19 raunak joined #fuel
00:24 raunak joined #fuel
00:25 gomarivera joined #fuel
00:26 zla joined #fuel
00:38 Julien-zte joined #fuel
00:39 Julien-zte joined #fuel
01:08 raunak joined #fuel
01:26 Julien-zte joined #fuel
01:27 Julien-zte joined #fuel
02:03 mdnadeem joined #fuel
02:32 kbaegis joined #fuel
02:33 kbaegis Hey. Has anyone had problems with cephx & fuel 9.1?
02:34 kbaegis '/Stage[main]/Ceph::Osds/Exec[ceph-disk activate-all]/returns) change from notrun to 0 1 failed: Command exceeded timeout'
02:37 Egyptian[Home] joined #fuel
02:48 ilbot3 joined #fuel
02:48 Topic for #fuel is now Fuel 9.0 (Mitaka) https://www.fuel-infra.org/ | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel/
03:08 Egyptian[Home] joined #fuel
03:16 fatdragon joined #fuel
03:24 Egyptian[Home] joined #fuel
03:25 Egyptian[Home] left #fuel
03:27 kbaegis joined #fuel
03:42 fatdragon joined #fuel
03:46 kbaegis joined #fuel
03:53 fatdragon joined #fuel
03:58 fatdragon joined #fuel
04:40 diogogmt joined #fuel
05:02 mdnadeem joined #fuel
05:04 DaTaz joined #fuel
05:05 DaTaz hello every one i see status partial deployment
05:05 DaTaz what is that mean
05:24 Vijayendra joined #fuel
05:37 gomarivera joined #fuel
05:42 kbaegis joined #fuel
06:31 gomarivera joined #fuel
06:48 portdirect_ joined #fuel
07:06 vkulanov joined #fuel
07:45 fatdragon joined #fuel
09:00 athinkingmeat joined #fuel
09:45 fatdragon joined #fuel
09:52 Vijayendra_ joined #fuel
09:52 priya joined #fuel
10:28 chardegen left #fuel
11:41 Egyptian[Home] joined #fuel
11:42 Egyptian[Home] left #fuel
12:47 fatdragon joined #fuel
12:53 gomarivera joined #fuel
14:00 kelepirci joined #fuel
14:00 kelepirci hello all
14:02 kelepirci is there any problem upstream code?
14:02 kelepirci I have just seem following error on Mirantis Download page?
14:02 kelepirci Due to an unexpected change in upstream code, the Mirantis OpenStack distribution is currently unavailable.
14:11 goldenfri On a fairly vanilla fuel 10 install I get - "No project admin with domain Default found" Any one know how to fix this?
14:15 Vijayendra_ joined #fuel
14:40 AlexAvadanii joined #fuel
14:42 Vijayendra joined #fuel
14:48 fatdragon joined #fuel
14:55 gomarivera joined #fuel
15:46 ashtokolov GMAzrael: the updated release schedule is here https://wiki.openstack.org/wiki/Fuel/Release_Schedule
15:48 GMAzrael ashtokolov: Much appreciated!!!
15:49 GMAzrael ashtokolov: Is newton being skipped?
15:57 ashtokolov GMAzrael: Fuel Newton aka Fuel 10 was released Oct20
15:57 GMAzrael So MOS 10 will skill Newton and go to Mitaka?
15:57 ashtokolov GMAzrael: Community ISO is available here https://wiki.openstack.org/wiki/Fuel#Where_to_get_Fuel
15:59 ashtokolov GMAzrael: I'm sorry I have no information about MOS10 but is must be based on Newton release according to our versioning
16:12 raunak joined #fuel
16:16 raunak_ joined #fuel
16:23 raunak_ joined #fuel
16:24 Jainesh joined #fuel
16:25 Jainesh Hi
16:25 Jainesh anyone have issue with Fuel 8 and 9 where Dell Nodes are merged together?
16:27 Jainesh Anyone there?
16:30 evgenyl Jainesh: what do you mean by "merged together"?
16:31 Jainesh I boot 1 node, and shows up in fuel. Boot second node, and it overwrites the first node in fuel
16:31 Jainesh Fuel always shows 1 node. but you can see that the mac address and details keep going back and forth between first and second node
16:31 Jainesh no ip conflict as these boot using pxe
16:32 Jainesh btw I am using dell c6100
16:40 evgenyl Jainesh: That is pretty strange...
16:40 evgenyl Jainesh: What about mac addresses for admin network, are they different?
16:40 evgenyl Jainesh: As far as I remember mac addresses are used to determine identity of the node.
16:45 Jainesh it is weird... as all mac address are different and the UUID is also different.
16:46 Jainesh I even have the same issue if I use C6220
16:48 fatdragon joined #fuel
16:52 johnavp19891 joined #fuel
16:55 evgenyl Jainesh: try running `fuel node` on master, there is a field with mac address, does it match nodes' macs in admin network?
16:56 gomarivera joined #fuel
16:59 Jainesh looks like it is using the MAC Address of the BMC of the node
17:02 Jainesh sorry. it is the correct mac address
17:02 evgenyl Jainesh: if you bring up different node, will this ma be changed?
17:02 evgenyl mac8
17:02 evgenyl mac*
17:03 Jainesh yes, and the ip address changes as well
17:04 Jainesh but the name stays the same "Untitled (28:c4)"
17:06 raunak joined #fuel
17:07 evgenyl Jainesh: How do you determine that they change? By running `fuel node`?
17:07 Jainesh yes
17:07 gomarivera joined #fuel
17:07 evgenyl Jainesh: I double checked and Fuel indeed uses mac address to identify the node on first boot.
17:09 Jainesh since it uses mac address, then not sure why Fuel only sees 1 node only and the mac/ip address keeps changing
17:10 Jainesh could be a bad fuel installation?
17:10 evgenyl MiroslavAnashkin: Have you seen such behavior?
17:11 Jainesh would the snapshot from fuel help?
17:12 DeMiNe0 joined #fuel
17:29 evgenyl Jainesh: yes, it would help, could you please create a ticket which describes your issue https://bugs.launchpad.net/fuel/+filebug ?
17:30 Jainesh doing that now
17:32 fatdragon joined #fuel
17:41 fatdragon joined #fuel
17:42 fatdragon joined #fuel
17:44 Zer0Byte__ joined #fuel
17:48 Jainesh https://bugs.launchpad.net/fuel/+bug/1650646
17:48 Jainesh added
17:55 evgenyl Jainesh: what are macs of these two nodes?
17:58 Jainesh 00:26:6C:FE:57:4C and 00:26:6C:FE:28:C4
18:08 xarses joined #fuel
18:29 Jainesh evgenyl noticed that they both have the same runtime uuid : runtime uuidd897c7cd-7db7-4f48-a475-f1607ab8457e
18:31 Jainesh evgenyl is Mellanox ConnectX-2 compatible with Fuel 8 or 9? it shows up as 10Gb rather than 40Gb. It is using the mlx4_en driver.
18:45 evgenyl Jainesh: Cannot say about Mellanox compatibility.
18:46 evgenyl Jainesh: runtime_uuid shouldn't affect the way the node is mapped to Nailgun representation.
19:14 Jainesh could it be an issue with the fuel installation?
19:15 evgenyl Jainesh: it's a bug in Fuel https://bugs.launchpad.net/fuel/+bug/1650646/comments/3
19:15 Jainesh really
19:15 evgenyl Jainesh: But appears on melanox because https://bugs.launchpad.net/fuel/+bug/1650646/comments/4
19:16 evgenyl Jainesh: You can find this file in your container `dockerctl shell nailgun`, and `find /usr/lib --name node.py`
19:17 Jainesh let me check
19:17 evgenyl Jainesh: after that add change mac=meta.get('mac') to mac=meta.get('mac', "").lower() and restart the container.
19:18 Jainesh let me try that...
19:18 Jainesh le tyou know shortly
19:27 Jainesh same results... creating another snapshot for review
19:28 evgenyl Jainesh: have you restarted it after applying the change?
19:29 Jainesh yes we restarted it
19:29 johnavp1989 joined #fuel
19:32 evgenyl Jainesh: can you send lines of code you changed from the container, so I can double-check.
19:34 Jainesh yes... uploading the screenshot of changes
19:34 Jainesh uploaded
19:38 evgenyl Jainesh: Ok, looks good, what command did you use to restart the container? Send me result of `docker ps | grep nailgun|`
19:38 evgenyl Jainesh: Ok, looks good, what command did you use to restart the container? Send me result of `docker ps | grep nailgun`
19:44 evgenyl Jainesh: ok, I got what is the problem. It cannot find an interface for new node and attempts to find by interfaces, find 00:01.
19:45 evgenyl Jainesh: Let me provide you with new patch. But Fuel folks will have to figure out more generic solution for the problem.
19:47 Jainesh sure sounds good to me
19:49 evgenyl Jainesh: https://github.com/openstack/fuel-web/blob/stable/8.0/nailgun/nailgun/objects/node.py#L138 here replace `[n["mac"].lower() for n in interfaces]` with `[n["mac"].lower() for n in interfaces if n["mac"].lower() != "00:00:00:00:00:01"]`
19:49 Jainesh 1 sec to make the changes.
19:50 HeOS joined #fuel
19:57 Jainesh made changes and restarted nailgun.
19:57 Jainesh rebooting nodes
20:02 evgenyl Jainesh: I'm just curious, do you know what interface with mac `00:00:00:00:00:01` is used for in case of mellanox cards?
20:02 Jainesh mellanox cards was suppose to be used for the private/management/storage to my ceph (Dell C2100). These are connected to mellanox IS5035Q Switches.
20:03 Jainesh these cards are QDR ConnectX-2 VPI.
20:03 Jainesh sadly not sure if they are supported with Fuel, as all documents says to install mellanox fuel plugin that uses connectx-3
20:10 Jainesh same issue. uploading screenshot of changes
20:15 evgenyl Jainesh: hmm, and you rebooted both nodes with mellanox cards?
20:15 Jainesh rebooted all nodes including the ones with the mellanox cards
20:16 evgenyl Jainesh: ok, upload diagnostic snapshot once more.
20:16 Jainesh here is an error  [2016-12-16T20:16:14.808003 #8010] DEBUG -- : Response: status: 409
20:16 Jainesh body: {"message": "Node with mac 00:26:2D:0B:A6:8C already exists - doing nothing", "errors": []}
20:17 evgenyl Jainesh: oh, interesting.
20:17 evgenyl Jainesh: could you please upload api.log only?
20:17 Jainesh where is that located
20:18 evgenyl /var/log/docker-logs/nailgun/api.log
20:18 evgenyl There are all requests from agents.
20:21 Jainesh [2016-12-16T20:16:52.701439 #8143] DEBUG -- : Response: status: 404 body: {"message": "Can't find node: {u'status': u'discover', u'ip': u'10.20.1.9', u'is _agent': True, u'platform_name': u'ROGUE12', u'mac': u'00:26:2D:0B:A6:8C', u'meta': {u'pci_devices': {u'product': u'ROGUE12 (N/A)', u'hand
20:22 evgenyl Jainesh: Could you please login to both nodes and check if file `/etc/nailgun_uid` exists?
20:22 evgenyl Jainesh: If exists, what is the content of file `/etc/nailgun_uid`.
20:23 Jainesh getting it
20:24 Jainesh no file exists
20:26 evgenyl Jainesh: Hmm, something really strange happens, ok, send fresh snapshot, I will try to look into it.
20:28 Jainesh sure. 1 sec to get that
20:38 Jainesh uploaded
20:45 johnavp1989 joined #fuel
20:49 evgenyl Jainesh: Could you please send me output of `ip a` from the node with mellanox card?
20:53 Jainesh root@bootstrap:~# ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default     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
20:55 yellowcircle joined #fuel
20:55 gomarivera joined #fuel
20:55 Jainesh root@bootstrap:~# ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default     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: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP group default qlen 1000     link/ether 00:26:2d:
20:58 evgenyl ashtokolov: Hm, are you sure that node with mac `00:26:6C:FE:28:C4` is booted?
20:59 evgenyl ashtokolov: Sorry, not for you.
20:59 evgenyl Jainesh: it was for you.
20:59 evgenyl Jainesh: I see only requests from node with mac `00:26:6C:FE:57:4C`
21:00 Jainesh will try again
21:06 Jainesh just powering off all the nodes, and deleting the discovered node. waiting for it to go offline in GUI
21:11 johnavp19891 joined #fuel
21:22 Jainesh 1 node shows up... waiting for second node to get registered... let you know the results
21:38 Jainesh new nodes don't show up
21:39 Jainesh perf samples too long (2502 > 2500), lowering kernal.perf_event_max_smaples_rate to 50000
22:22 xarses_ joined #fuel
22:23 andreww joined #fuel
22:35 johnavp19891 joined #fuel
22:48 xarses_ joined #fuel
22:56 gomarivera joined #fuel
23:50 francois joined #fuel
23:57 gomarivera joined #fuel

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary