Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2014-08-15

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

All times shown according to UTC.

Time Nick Message
00:08 KodiakF joined #fuel
00:20 KodiakF2 joined #fuel
00:26 rmoe joined #fuel
00:45 lordd_ joined #fuel
00:49 adanin joined #fuel
01:45 jaypipes joined #fuel
02:06 jaypipes joined #fuel
02:33 Krast joined #fuel
02:49 adanin joined #fuel
03:10 Krast joined #fuel
03:13 rongze joined #fuel
03:40 xarses joined #fuel
05:13 xarses joined #fuel
05:40 Longgeek joined #fuel
05:51 adanin joined #fuel
06:09 Krast joined #fuel
06:14 Baboune Fuel 5.0.1 RC, Ubuntu, All nodes were deployed with an error.  What does this message imply: "DEBUG [7ffadabfd700] (node) Node Untitled (2e:ac) has provisioning or error status - status not updated by agent"?
06:15 Baboune from nailgun log
06:18 Baboune also during node setup: "anaconda.syslog:12:30:53,561 WARNING NetworkManager: <warn> error requesting auth for org.freedesktop.NetworkManager.enable-disable-wwan: (26) Remote Exception invoking org.freedesktop.PolicyKit1.Authority.CheckAuthorization() on /org/freedesktop/PolicyKit1/Authority at name org.freedesktop.PolicyKit1: org.freedesktop.DBus.Error.Spawn.ExecFailed: Cannot launch daemon, file not found or permissions invalid"
06:31 adanin joined #fuel
06:38 Baboune About deployment error, I removed 7 nodes (only have 8 left) and the error disapeared. Any clues?  There are no meaningful error messages aside from the one mentioned above
06:59 xarses Baboune: anything in in the asute log?
07:01 e0ne joined #fuel
07:02 Baboune @xarses noting special no
07:03 Baboune 2014/08/14 14:09:24 Cannot connect to the Docker daemon. Is 'docker -d' running on this host? Container astute is not running. Cannot attac
07:03 Baboune then this error disapears
07:03 Baboune 2014/08/14 14:11:14 dial unix /var/run/docker.sock: no such file or directory 2014/08/14 14:11:14 dial unix /var/run/docker.sock: no such file or directory 2014/08/14 14:11:14 dial unix /var/run/docker.sock: no such file or directory Attaching to container fuel-core-5.0.1-astute... 2014/08/14 14:11:14 dial unix /var/run/docker.sock: no such file or directory checking container astute checking with command "shell_container astute ps aux
07:03 xarses oh, well if astute is missing, that will be quite the problem
07:04 xarses since it goes out and does all this
07:05 xarses anyway there is a RC2 around for 5.0.1 in part due to some serious astute issues
07:06 xarses although I haven't heard about it being missing
07:06 xarses you should be able to supervisorctl restart docker-astute
07:06 xarses and it should reload the container for you
07:06 xarses Baboune: ^^
07:13 e0ne joined #fuel
07:19 e0ne joined #fuel
07:24 e0ne joined #fuel
07:30 e0ne joined #fuel
07:36 teknoprep joined #fuel
07:37 Longgeek joined #fuel
07:51 Longgeek joined #fuel
08:00 e0ne joined #fuel
08:11 Baboune it just takes a long time to come up. later in the logs: try number 22 return code is 1 checking with command "curl -f -L -u naily:naily -s http://127.0.0.1:15672/api/exchanges | grep -qw 'naily_service'" try number 1 return code is 1 astute is ready. fuel-core-5.0.1-astute is already running. Attaching to container fuel-core-5.0.1-astute...
08:25 HeOS joined #fuel
08:34 Baboune A question about DNS.  During the health checks, the DNS check fails.  This is expected as the Fuel Master does not have a valid resolver.  How do you configure the DNS masq in 5.x?  In 4.x, it was sufficient to run a DNS masq on the Fuel Master, and to update dnsmasq.conf, but in 5.1, the DNSmasq seems to be running in one of the docker containers.
08:50 Longgeek joined #fuel
08:53 Longgeek joined #fuel
08:59 ArminderS joined #fuel
09:14 e0ne joined #fuel
09:23 DaveJ__ joined #fuel
09:47 Longgeek joined #fuel
10:04 e0ne joined #fuel
11:00 Longgeek joined #fuel
11:12 evg Baboune: do you really mean 5.1? In 5.0.x dnsmask is on its place.
11:34 Longgeek_ joined #fuel
12:55 BillTheKat joined #fuel
12:57 BillTheKat I am having trouble with some bridges and I a trying to track down MAC adds on my fuel PXE network.  Is there a way for me to figure the MACs of all the stuff running on my fuel server?
13:12 gregory_pmd joined #fuel
13:12 GoKEV joined #fuel
13:13 evg BillTheKat: ip neig list
13:14 BillTheKat evg:
13:14 BillTheKat evg: tx
13:15 evg ip link list (stuff running on fuel node
13:17 BillTheKat evg: the ip neig list - shows me what I needed to see......... THANKS
13:23 e0ne joined #fuel
13:35 bookwar joined #fuel
13:38 twforeman joined #fuel
15:01 ppradhan joined #fuel
15:08 teknomega joined #fuel
15:58 rmoe joined #fuel
16:40 angdraug joined #fuel
17:18 ArminderS- joined #fuel
17:25 xarses joined #fuel
17:31 BillTheKat joined #fuel
18:17 gregory_pmd anyone around?
18:21 MiroslavAnashkin Simply ask your question. Even if there is nobody here, we read the logs of this chat and may answer later.
18:40 gregory_pmd oh, okay.
18:41 gregory_pmd I'm doing a nova install and some of my nodes aren't getting DHCP addresses allocated.
18:41 gregory_pmd I ran into this: https://bugs.launchpad.net/fuel/+bug/1323354 first and got that sorted with the update (downgrade to 2.6) listed in known issues.
18:43 gregory_pmd now however, when I'm PXE booting (which works) I get to the loader screen and then the install fails to bring up the interface eth0 even though it's link and good
18:43 gregory_pmd when the loader finishes and I get into the console, the interface is up and connected but with no IP address allocated.
18:49 MiroslavAnashkin What is your network card model? Is it Broadcom 5720?
18:51 gregory_pmd I think so.
18:52 gregory_pmd using bnx2x
18:52 gregory_pmd the really odd part is it was working yesterday during a test build.
18:53 gregory_pmd we blew everything away and restarted to document the process better and now I can't get either 720 identified in fuel
18:55 MiroslavAnashkin There is known issue with this card. It starts too slow. Default LINKDELAY parameter value in  /etc/init.d/setup-bootdev file in the initramfs.img file (bootstrap PXE image) may be not enough.
18:56 MiroslavAnashkin It may be also related to the link stability
18:59 MiroslavAnashkin BTW, Mirantis OpenStack 5.0.1 is already uploaded to Mirantis web site and will be available today for download. It includes updated bootstrap image.
19:00 MiroslavAnashkin While, it does not include increased delay time for NIC start.
19:00 gregory_pmd ooohh... well that fixes the first one.
19:01 TVR_ nice...
19:01 TVR_ I will have to see how well it handles hardware
19:01 gregory_pmd any workaround for this as of now?
19:01 e0ne joined #fuel
19:02 TVR_ what is the solution for "Error: Modified project information and members, but unable to modify project quotas." when you edit the quotas for Fuel 4.1?
19:03 MiroslavAnashkin For now you may try to include the LINKDELAY=30 value on lines 48 and 81 to /etc/init.d/setup-bootdev file inside the initramfs.img. If you would like to try this solution, I'll provide you instructions. However, I would recommend to use the latest, 5.0.1 bootstrap image
19:07 gregory_pmd once it's up and loaded, can't I just hard set the IP to something in the discovered node ip range and let let Fuel see it that way? (spitballing here)
19:07 e0ne joined #fuel
19:08 MiroslavAnashkin IP should be set in scope of PXE - so your hard IP should match with that assigned by master node
19:09 gregory_pmd woohoo that worked!
19:10 gregory_pmd still no idea why it worked yesterday and not today.
19:11 MiroslavAnashkin TVR_: Please check https://bugs.launchpad.net/horizon/+bug/1227102 and https://bugs.launchpad.net/horizon/+bug/1286099 first
19:12 MiroslavAnashkin Looks like today it needs more time to perform link speed calibration or whatever it does on start
19:13 MiroslavAnashkin It is normal for optical NICs and looks like getting normal for 10G NICs as well
19:18 gregory_pmd think something else is going on, cause if that was it I should just be able to do a 'dhclient -r' and snag a new IP.
19:19 TVR_ The website changed for Mirantis.... Also, the 10 node cluster size is now spelled out in several places in the contract form
19:41 xarses gregory_pmd: which network did you assign to your instance?
19:42 MiroslavAnashkin Not instance - IP address inside bootstrapped node
19:42 xarses sorry, maybe you are still trying to install openstack, not clear from the scroll back
19:44 xarses gregory_pmd: what is the output of 'dhclient -v -i eth0' (or whichever)
19:46 MiroslavAnashkin Network interface starts too slow on OS boot, so OS continues to boot thinking the NIC is off and NIC does not get an IP from master node during boot. We just resolved very similar ticket with the same NIC model.
19:47 TVR_ what is the CLI command to edit the default quota? So I (as example) "cinder quota-update --volumes 15 admin" and now I have "cinder quota-show admin" showing me 15 volumes I can create... but from the UI, I still see the admin project has the limit of 10 volumes, and it seems that is what causes the 11'th instance to fail create
19:48 MiroslavAnashkin There are both, `cinder quota-update` and `nova quota-update` if i remember well.
19:49 TVR_ editing ~/nova/quota.py did not seem to fix it from the UI... but often I have seen the CLI gives more info OR works
19:53 tatyana joined #fuel
19:55 dmitryme joined #fuel
19:59 e0ne joined #fuel
20:19 e0ne joined #fuel
20:43 gregory_pmd joined #fuel
20:43 gregory_pmd Sorry guys, had to switch locations.
20:45 gregory_pmd didn't get a chance to do the dhclient check before I left.  hard setting the IPs to something in the discovery range worked a treat.  tried renewing the dhclient, but it didna' seem to work.  either way the deploy went off without a hitch and we'll be redoing it again next week under 5.0.1.
20:45 gregory_pmd thanks for the help! you guys are rockstars!
20:45 gregory_pmd I'm out for the weekend.  Have a good one.
20:51 e0ne joined #fuel
21:04 kupo24z did 5.0.1 get approved for release today?
21:07 xarses kupo24z: yes, It should be on the website soon
21:08 kupo24z Cool, I see the Release notes have been updated
21:08 ArminderS- yay
21:09 ArminderS- busy weekend deploying 5.0.1
21:17 xarses kupo24z: Arminder- it sounds like it should be live on software.mirantis.com now
21:18 kupo24z yeh the buttons wernt working for me before heh
21:26 ArminderS- great, thanks
21:31 ArminderS- where are the download buttons, and whats this opt-in thing
21:31 ArminderS- even on login, we have to fill that form?
21:32 xarses ArminderS-: ya, if you don't have a login already
21:32 ArminderS- i did have and even if i got to manage account link, i can see i'm logged in
21:33 ArminderS- as it shows my profile and didn't ask for login
21:33 ArminderS- its just me?
21:33 ArminderS- waiting for that email to show up
21:33 xarses https://software.mirantis.com/new/#downloads
21:36 ArminderS- thanks xarses
21:37 xarses did that get you where you needed?
21:37 ArminderS- weird that i got that email, but not even link to is in that email, just getting starting page with virtualbox
21:37 Topic for #fuel is now Fuel 5.0.1 for Openstack: https://wiki.openstack.org/wiki/Fuel | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel/
21:37 ArminderS- problem with the overlays somewhere
21:38 ArminderS- download button doesn't work for me on crome
21:38 ArminderS- trying on ie
21:38 kupo24z Yeah same issue on chrome
21:39 xarses can you try clearing your cookies for mirantis.com?
21:39 ArminderS- it used to be flawless earlier
21:39 ArminderS- still we can live with it :)
21:39 ArminderS- ie doesn't even show the full page
21:41 ArminderS- wow, still same on IE 11
21:45 Arminder- if anyone has a direct link i can wget, please pm me the same
21:46 xarses joined #fuel
21:47 e0ne joined #fuel
21:48 ArminderS if anyone has a direct link i can wget, please pm me the same
21:49 ArminderS i have already tried from different places
21:50 xarses try doing https://software.mirantis.com/terms-and-conditions/ this first
21:52 ArminderS thanks, it worked
21:52 xarses yay
21:55 xarses kupo24z: did that work for you?
21:58 kupo24z xarses: is the 10 node thing new?
22:01 xarses what 10 node thing?
22:02 kupo24z "You can use the Mirantis OpenStack Software you download here for evaluation and test on up to ten nodes." Cant remember if thats new or not
22:03 xarses ya, but where is that?
22:04 kupo24z The link you sent, https://software.mirantis.com/terms-and-conditions/ 3.1
22:07 xarses I've not seen it before
22:08 kupo24z Is that a hard cap added in fuel now?
22:08 xarses no
22:09 xarses I've not heard about this, so if anything its a config for nailgun or something that can be changed
22:09 xarses we coudn't enforce a hard cap if we wanted to
22:09 xarses the code is all open source
22:13 kupo24z does this TOS specifically apply to all Mirantis products incl. Fuel?
22:13 kupo24z er TOC*
22:21 xarses I'll have to discuss this with some internal folks
22:22 kupo24z Okay thanks
22:23 e0ne joined #fuel
22:23 xarses I'd personally not worry about it at the moment, however that is not legal advice
22:23 xarses if you are worried about it, try one of the community builds
22:24 kupo24z Is there a Terms and conditions for the community releases?
22:24 kupo24z which im guessing are https://fuel-jenkins.mirantis.com/view/ISO/job/publish_fuel_community_iso/
22:25 xarses Not that I'm aware of. It would have to be there somewhere or once unpacked and I've not seen one
22:26 kupo24z Are they only for 5.1 or is there stable releases for 5.0.1 as well?
22:36 xarses kupo24z: apparently it's been that way since at least October according to the product manager
22:38 kupo24z xarses: and those terms cover Fuel GA releases as well?

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