*** masber has joined #openstack-sahara | 00:05 | |
*** dave-mcc_ has joined #openstack-sahara | 00:05 | |
*** dave-mccowan has quit IRC | 00:07 | |
*** deep-book-gk has joined #openstack-sahara | 01:39 | |
*** deep-book-gk has left #openstack-sahara | 01:42 | |
openstackgerrit | Hongtao Zhang proposed openstack/sahara master: Enable some off-by-default checks https://review.openstack.org/487379 | 01:59 |
---|---|---|
*** tuanluong has joined #openstack-sahara | 02:38 | |
*** ukaynar has quit IRC | 02:44 | |
*** ukaynar has joined #openstack-sahara | 02:46 | |
*** ukaynar has quit IRC | 02:48 | |
*** ukaynar has joined #openstack-sahara | 02:49 | |
openstackgerrit | zhongshengping proposed openstack/puppet-sahara master: Update openstackdocstheme>=1.16.0 https://review.openstack.org/489065 | 03:03 |
*** shuyingya has joined #openstack-sahara | 04:55 | |
*** pgadiya has joined #openstack-sahara | 05:10 | |
*** Poornima_K has joined #openstack-sahara | 05:11 | |
*** dhellmann has quit IRC | 05:28 | |
*** dhellmann has joined #openstack-sahara | 05:28 | |
*** Poornima_K has quit IRC | 05:41 | |
*** GK1wmSU has joined #openstack-sahara | 06:02 | |
openstackgerrit | Shu Yingya proposed openstack/sahara master: Fix UnicodeEncoding Error https://review.openstack.org/488387 | 06:03 |
*** GK1wmSU has left #openstack-sahara | 06:03 | |
*** rcernin has joined #openstack-sahara | 06:11 | |
*** _GK1wmSU has joined #openstack-sahara | 06:16 | |
*** _GK1wmSU has left #openstack-sahara | 06:18 | |
*** pcaruana has joined #openstack-sahara | 06:19 | |
*** tuanluong has quit IRC | 06:19 | |
*** Poornima_K has joined #openstack-sahara | 07:00 | |
*** Poornima_K has quit IRC | 07:06 | |
*** Poornima_K has joined #openstack-sahara | 07:08 | |
*** Poornima_K has quit IRC | 07:09 | |
*** Poornima_K has joined #openstack-sahara | 07:10 | |
*** ukaynar has quit IRC | 07:17 | |
*** tuanluong has joined #openstack-sahara | 07:18 | |
*** Poornima_K has quit IRC | 07:23 | |
*** dhellmann has quit IRC | 07:30 | |
*** dhellmann has joined #openstack-sahara | 07:31 | |
*** esikachev has joined #openstack-sahara | 07:38 | |
*** Poornima_K has joined #openstack-sahara | 08:39 | |
*** Poornima_K has quit IRC | 08:40 | |
*** Poornima_K has joined #openstack-sahara | 08:42 | |
*** Poornima_K has quit IRC | 08:43 | |
*** pgadiya has quit IRC | 09:17 | |
*** tosky has joined #openstack-sahara | 09:21 | |
*** pgadiya has joined #openstack-sahara | 09:33 | |
*** Poornima_K has joined #openstack-sahara | 09:37 | |
*** Poornima_K has quit IRC | 09:50 | |
*** esikachev has quit IRC | 10:09 | |
*** esikachev has joined #openstack-sahara | 10:09 | |
*** esikachev has quit IRC | 10:14 | |
*** pgadiya has quit IRC | 10:36 | |
*** esikachev has joined #openstack-sahara | 10:46 | |
*** zemuvier has joined #openstack-sahara | 10:48 | |
*** pgadiya has joined #openstack-sahara | 10:49 | |
*** tuanluong has quit IRC | 11:08 | |
*** ukaynar has joined #openstack-sahara | 11:09 | |
*** ukaynar has quit IRC | 11:19 | |
*** ukaynar has joined #openstack-sahara | 11:19 | |
*** ukaynar has quit IRC | 11:23 | |
*** ukaynar has joined #openstack-sahara | 11:23 | |
*** ukaynar has quit IRC | 11:25 | |
*** ukaynar has joined #openstack-sahara | 11:26 | |
*** ukaynar has quit IRC | 11:30 | |
*** ltosky[m] has quit IRC | 12:01 | |
*** shuyingya has quit IRC | 12:35 | |
*** ukaynar has joined #openstack-sahara | 12:57 | |
*** shuyingya has joined #openstack-sahara | 12:59 | |
*** lucasxu has joined #openstack-sahara | 13:02 | |
*** shuyingya has quit IRC | 13:04 | |
*** shuyingya has joined #openstack-sahara | 13:08 | |
*** jeremyfreudberg has joined #openstack-sahara | 13:10 | |
*** shuyingya has quit IRC | 13:11 | |
*** ukaynar has quit IRC | 13:12 | |
*** ukaynar has joined #openstack-sahara | 13:12 | |
*** shuyingya has joined #openstack-sahara | 13:13 | |
*** ukaynar has quit IRC | 13:17 | |
*** ukaynar has joined #openstack-sahara | 13:21 | |
*** shuyingya has quit IRC | 13:23 | |
*** esikachev has quit IRC | 13:28 | |
*** shuyingya has joined #openstack-sahara | 13:39 | |
*** shuyingya has quit IRC | 13:43 | |
*** pgadiya has quit IRC | 13:56 | |
*** ssmith has joined #openstack-sahara | 13:57 | |
tosky | I almost nailed the sahara-dashboard failure: it looks like an issue in horizon | 14:27 |
jeremyfreudberg | tosky, nice | 14:27 |
tosky | horizon bug coming, with fix and dependency | 14:27 |
tosky | and then discussion with the horizon team (which I already nagged :) | 14:28 |
ssmith | Good morning gentlemen. We now have ephemeral volumes working and the instances are launching but the heat stacks are getting stuck. Any hints on what we can look at? http://imgur.com/R4Nx3sa | 14:31 |
jeremyfreudberg | hey ssmith, glad you finally got the storage working... anything in heat-engine logs? | 14:32 |
*** openstackgerrit has quit IRC | 14:33 | |
*** ltosky[m] has joined #openstack-sahara | 14:39 | |
*** shuyingya has joined #openstack-sahara | 14:40 | |
*** shuyingya has quit IRC | 14:44 | |
*** openstackgerrit has joined #openstack-sahara | 14:52 | |
openstackgerrit | Luigi Toscano proposed openstack/sahara-dashboard master: Switch render() arguments to the new way https://review.openstack.org/488425 | 14:52 |
jeremyfreudberg | tosky, nice work | 14:53 |
tosky | whack-a-mole and yak shaving to the top | 14:54 |
jeremyfreudberg | :D | 14:54 |
*** zemuvier has quit IRC | 14:55 | |
*** ukaynar has quit IRC | 15:01 | |
*** rcernin has quit IRC | 15:03 | |
*** pcaruana has quit IRC | 15:04 | |
*** ukaynar has joined #openstack-sahara | 15:06 | |
*** ukaynar_ has joined #openstack-sahara | 15:08 | |
*** ukaynar has quit IRC | 15:11 | |
*** esikachev has joined #openstack-sahara | 15:14 | |
*** lucasxu has quit IRC | 15:26 | |
*** shuyingya has joined #openstack-sahara | 15:29 | |
*** shuyingya has quit IRC | 15:34 | |
tosky | tellesnobrega, jeremyfreudberg, esikachev: the horizon patch is going to land, so https://review.openstack.org/488425 should unlock sahara-dashboard | 15:44 |
tosky | and then we can ninja merge the request from the update bot for example | 15:44 |
tosky | now I can take a bre.. no, the OTHER patch | 15:44 |
tosky | :D | 15:44 |
*** jeremyfreudberg has quit IRC | 15:46 | |
*** ukaynar_ has quit IRC | 15:48 | |
*** ukaynar has joined #openstack-sahara | 15:49 | |
tomtomtom | I'm seeing this in the sahara logs, 2017-07-31 15:37:24.876 223 INFO keystonemiddleware.auth_token [-] Rejecting request anyone else experience this issue and solve it? According to my keystone auth section the username(sahara) and password are correct. | 15:52 |
*** ukaynar has quit IRC | 15:53 | |
tosky | tomtomtom: master? | 15:53 |
tosky | tomtomtom: how did you deploy it? | 15:53 |
tomtomtom | this is an openstack newton version deployed with openstack-ansible. | 15:54 |
tosky | so, jeremyfreudberg have seen that message, and I've seen it too, but in different context | 15:54 |
tosky | in my case (master deployment), haproxy tries to contact periodically (not sure if it's misconfigured or not) | 15:54 |
tosky | I don't remember what was the issue in Jeremy's deployment | 15:55 |
tosky | how much frequently do you see it? | 15:55 |
tomtomtom | every 10 seconds or so. | 15:56 |
tosky | as mentioned, I' | 15:56 |
tosky | grmf | 15:56 |
tosky | I'm still not sure if it's an issue in my case (haproxy); I'd suggest you to wait for Jeremy and see if he has more details | 15:57 |
tomtomtom | ok | 15:57 |
*** lucasxu has joined #openstack-sahara | 16:10 | |
*** ukaynar has joined #openstack-sahara | 16:11 | |
*** ukaynar has quit IRC | 16:18 | |
*** shuyingya has joined #openstack-sahara | 16:19 | |
*** shuyingya has quit IRC | 16:23 | |
*** shuyingya has joined #openstack-sahara | 16:34 | |
*** shuyingya has quit IRC | 16:39 | |
*** esikachev has quit IRC | 16:44 | |
ssmith | No errors in heat logs and everything stops after creating the servers. Turning on heat debug logging | 17:04 |
*** jeremyfreudberg has joined #openstack-sahara | 17:14 | |
jeremyfreudberg | tomtomtom, about your issue | 17:15 |
jeremyfreudberg | the "Rejecting request" is not actual an error or misconfiguration of sahara itself | 17:15 |
jeremyfreudberg | as tosky said, in his case it has haproxy sending a request to the sahara service, in my case it was nagios | 17:16 |
jeremyfreudberg | but these request are not wellformed for a service using keystonemiddleware for auth | 17:16 |
jeremyfreudberg | hence, rejecting request | 17:16 |
jeremyfreudberg | nothing really to worry about | 17:16 |
ssmith | So it's stuck in the image above with heat message "Task create from HeatWaitCondition "sparkslave-wc-waiter" Stack "Hellboy5d8ed36df-sparkslave-wrdnopuuf7zq-0-sjjl2ht437x7" " repeating over and over | 17:17 |
jeremyfreudberg | ssmith, try heat_enable_wait_condition=False in DEFAULT of sahara.conf | 17:19 |
ssmith | Is there anyway to see what it's waiting on? | 17:19 |
jeremyfreudberg | ssmith, wait condition is used for getting status of vm (has it booted, is ssh ready) but in my deployment it never worked.... on the other hand, we knew of some deployments where it was necessary | 17:20 |
*** ukaynar has joined #openstack-sahara | 17:21 | |
*** shuyingya has joined #openstack-sahara | 17:23 | |
jeremyfreudberg | ssmith, your cloud is newton? | 17:24 |
*** shuyingya has quit IRC | 17:28 | |
*** ukaynar has quit IRC | 17:29 | |
*** tosky has quit IRC | 17:32 | |
*** esikachev has joined #openstack-sahara | 17:41 | |
*** esikachev has quit IRC | 17:45 | |
*** ukaynar has joined #openstack-sahara | 17:50 | |
*** esikachev has joined #openstack-sahara | 17:52 | |
*** tosky has joined #openstack-sahara | 17:56 | |
*** tosky has quit IRC | 17:56 | |
*** esikachev has quit IRC | 17:57 | |
*** tosky has joined #openstack-sahara | 17:57 | |
*** tosky has quit IRC | 17:58 | |
*** tosky has joined #openstack-sahara | 17:58 | |
*** ukaynar has quit IRC | 17:59 | |
*** ukaynar has joined #openstack-sahara | 17:59 | |
openstackgerrit | Evgeny Sikachev proposed openstack/sahara-ci-config master: [wip]Added ability to deploy zuul https://review.openstack.org/489289 | 18:01 |
*** shuyingya has joined #openstack-sahara | 18:03 | |
*** ukaynar has quit IRC | 18:03 | |
*** shuyingya has quit IRC | 18:07 | |
ssmith | jeremyfreudberg: Yes, it's Newton | 18:08 |
ssmith | Now all Heat stack complete but status in Sahara is "Waiting" | 18:08 |
jeremyfreudberg | ssmith, ok, if it's newton there was some bugfixes regarding heat condition which should be there (if you were mitaka or earlier, could be a problem). how long has sahara been on "Waiting" ? waiting means can't ssh in | 18:09 |
ssmith | I was in a 1/2 hour meeting and started it before then so 35 minutes | 18:11 |
ssmith | I can ssh into it with the supplied key (which is mine on the OS system) with user ubuntu | 18:12 |
jeremyfreudberg | ssmith, can you check instance log (see if there is a key with comment "Generated-By-Sahara" in cloud-init logs) | 18:13 |
ssmith | No there is not just "Generated-by-Nova " | 18:14 |
* jeremyfreudberg thinks | 18:19 | |
jeremyfreudberg | ssmith, not exactly sure, but as a first step, might as well check authorized_keys file of ubuntu user instead of cloud-init logs | 18:20 |
jeremyfreudberg | just to cover all our bases | 18:20 |
ssmith | Just one on the instance "generated-by-nova" | 18:21 |
*** ukaynar has joined #openstack-sahara | 18:21 | |
jeremyfreudberg | ok, that's unfortunate, but at least consistent | 18:21 |
jeremyfreudberg | are you using floating IP or internal IP ? | 18:21 |
jeremyfreudberg | in sahara configuration, I mean | 18:22 |
ssmith | both | 18:22 |
jeremyfreudberg | so use_floating_ips is True, but not all instances have floating? just to clarify | 18:22 |
ssmith | One master, one slave both with a private tenant IP both with Floating IPs | 18:23 |
jeremyfreudberg | ok | 18:25 |
ssmith | Which key should be assigned to the stack? I used one that was generated for myself on the pulldown list | 18:26 |
jeremyfreudberg | ssmith, generally there would be two | 18:26 |
jeremyfreudberg | one that sahara creates for itself | 18:26 |
jeremyfreudberg | and one that is the user's | 18:27 |
jeremyfreudberg | right now, i'm trying to remember how the "Generated-By-Sahara" keypair gets inserted, i know the user's keypair comes through the heat template itself, but the management one is something else | 18:27 |
jeremyfreudberg | so here's the answer about that | 18:30 |
jeremyfreudberg | user keypair is the keypair that nova knows about | 18:31 |
jeremyfreudberg | management keypair is inserted as part of the "User data" script | 18:31 |
jeremyfreudberg | so i guess to diagnose | 18:32 |
jeremyfreudberg | see if you can launch a regular instance with a script in "User data" | 18:32 |
jeremyfreudberg | ^ ssmith | 18:34 |
ssmith | Before trying that I was just looking through the logs and see " Failed fetching userdata from url http://169.254.169.254/2009-04-04/user-data | 18:41 |
ssmith | " so could that be the issue? | 18:41 |
jeremyfreudberg | ssmith, that's the issue | 18:41 |
jeremyfreudberg | to clarify, some problem with you cloud-init / nova metadata server | 18:46 |
tosky | and/or network setup | 18:47 |
jeremyfreudberg | tosky - true, ssmith ^ | 18:47 |
ssmith | On machines with non-floating sets 169.254.169.254 | 10.0.128.10 and on this tenant with both sets to 169.254.169.254 | 172.16.1.1 | 18:49 |
*** ukaynar has quit IRC | 18:50 | |
*** ukaynar has joined #openstack-sahara | 18:50 | |
jeremyfreudberg | ssmith, what exactly are you referring to? sorry | 18:53 |
*** ukaynar has quit IRC | 18:54 | |
*** ukaynar has joined #openstack-sahara | 18:55 | |
ssmith | NVM, I ssh into the spark master and it can telnet to 169.254.169.254 80 | 18:57 |
*** ukaynar has quit IRC | 19:06 | |
*** shuyingya has joined #openstack-sahara | 19:12 | |
*** shuyingya has quit IRC | 19:16 | |
*** esikachev has joined #openstack-sahara | 19:33 | |
esikachev | jeremyfreudberg: ping | 19:35 |
ssmith | jeremyfreudberg: would you have something that would be a good test for the user data script? | 19:35 |
jeremyfreudberg | woah, two pings at once | 19:35 |
esikachev | :) | 19:35 |
jeremyfreudberg | ssmith, just a shell script with a valid shebang and echo hello world | 19:35 |
jeremyfreudberg | if you can run that you can run anything | 19:35 |
jeremyfreudberg | esikachev, what's up? | 19:36 |
esikachev | jeremyfreudberg: everything is ok with HDDs on cloud for sahara-ci? | 19:36 |
esikachev | is too slow :( | 19:36 |
jeremyfreudberg | esikachev, probably it's not ok :( | 19:37 |
esikachev | this is happend periodicaly | 19:37 |
jeremyfreudberg | i see some filesystem junk messages in your instance log... | 19:38 |
ssmith | If I do a curl from the cluster sparkmaster I get this curl http://169.254.169.254/2009-04-04 | 19:38 |
ssmith | --> meta-data/ | 19:38 |
ssmith | no /user-data | 19:38 |
jeremyfreudberg | esikachev, unfortunately it happens a lot | 19:38 |
jeremyfreudberg | it can all be recovered and fixed, but it might happen a gain | 19:38 |
openstackgerrit | Evgeny Sikachev proposed openstack/sahara-ci-config master: [wip]Added ability to deploy zuul https://review.openstack.org/489289 | 19:40 |
jeremyfreudberg | esikachev, i sent a note for it to be recovered, hopefully fixed soon | 19:40 |
jeremyfreudberg | ssmith, very interesting | 19:41 |
esikachev | ok. thank you. i finished for today with instance | 19:41 |
jeremyfreudberg | ssmith, investigating | 19:43 |
ssmith | Wait, yes there is....it's just down on the input line curl http://169.254.169.254/2009-04-04 | 19:43 |
ssmith | meta-data/ | 19:43 |
ssmith | user-dataubuntu@hellboy-6-sparkmaster-0:~$ | 19:43 |
jeremyfreudberg | yep, there's no newline so hard to see | 19:44 |
ssmith | Then it just hangs there curl -v http://169.254.169.254/2009-04-04/user-data/ | 19:44 |
ssmith | * Hostname was NOT found in DNS cache | 19:44 |
ssmith | * Trying 169.254.169.254... | 19:44 |
ssmith | * Connected to 169.254.169.254 (169.254.169.254) port 80 (#0) | 19:44 |
ssmith | > GET /2009-04-04/user-data/ HTTP/1.1 | 19:44 |
ssmith | > User-Agent: curl/7.35.0 | 19:44 |
ssmith | > Host: 169.254.169.254 | 19:44 |
ssmith | > Accept: */* | 19:44 |
ssmith | > | 19:44 |
jeremyfreudberg | ssmith, not sure how to diagnose, really not something i'm very familiar with | 19:47 |
ssmith | Launched server from Horizon with a config script and user-data works | 19:56 |
*** esikachev has quit IRC | 19:57 | |
*** shuyingya has joined #openstack-sahara | 20:01 | |
jeremyfreudberg | ssmith, hmm, then i would expect it to work on sahara instances too | 20:02 |
*** shuyingya has quit IRC | 20:05 | |
ssmith | jeremyfreudberg: this is in the DB so looks like it's there. Looks hashed but does it look normal? http://imgur.com/j94gyaJ | 20:43 |
jeremyfreudberg | ssmith, you got this from SELECT management_private_key FROM <sahara db name>.clusters; ? | 20:44 |
jeremyfreudberg | or something else? | 20:44 |
ssmith | select user_data from instance | 20:47 |
jeremyfreudberg | ssmith, oh, the nova table | 20:47 |
jeremyfreudberg | not familiar with what it's supposed to look like in there | 20:47 |
jeremyfreudberg | and this is the entry referring to sahara-created instance? | 20:48 |
ssmith | Is there a way to paste in a pre-generated key pair on the horizon UI? | 20:48 |
jeremyfreudberg | ssmith, there is | 20:48 |
jeremyfreudberg | access and security -> key pairs -> import key pair | 20:49 |
jeremyfreudberg | not sure if that will help you much | 20:49 |
jeremyfreudberg | any way, checked my own nova db, my user_data column also looks like yours | 20:51 |
jeremyfreudberg | so not to worry about that specifically, ssmith | 20:51 |
jeremyfreudberg | ssmith, sorry have to go afk now. | 20:53 |
*** jeremyfreudberg has quit IRC | 20:53 | |
ssmith | ok, thanks | 20:53 |
*** shuyingya has joined #openstack-sahara | 21:00 | |
*** lucasxu has quit IRC | 21:02 | |
*** shuyingya has quit IRC | 21:04 | |
ssmith | So we've discovered that the spark instance can reach user-data if the mtu is lowered to 1500. We run on 10G networks so have set the mtu on the servers to 9000. But dang if we can find out where the metadata service is set to 1500 | 21:58 |
*** shuyingya has joined #openstack-sahara | 21:59 | |
*** shuyingya has quit IRC | 22:03 | |
tosky | in most of the case, sahara issues are $some_other_service issue | 22:17 |
ssmith | Wierd that highest mtu it will work with is 7340 | 22:24 |
*** shuyingya has joined #openstack-sahara | 22:38 | |
*** shuyingya has quit IRC | 22:43 | |
*** ukaynar has joined #openstack-sahara | 22:53 | |
openstackgerrit | Merged openstack/sahara-dashboard master: Switch render() arguments to the new way https://review.openstack.org/488425 | 23:30 |
tosky | uh, I haz the power | 23:38 |
tellesnobrega | tosky, you do :) | 23:38 |
tellesnobrega | I was going to ask you to check if you have +2 powers already | 23:38 |
tellesnobrega | but thought you wouldn't be online now | 23:38 |
tosky | ehm, yeah, I'm usually late | 23:39 |
tellesnobrega | I had the worse trip of my life today | 23:39 |
tosky | oh | 23:39 |
tellesnobrega | from my parent's trip to mine, it is usually a 1:30h drive | 23:40 |
tellesnobrega | it took 7h today | 23:40 |
tosky | the time of an intercontinental flight | 23:40 |
tosky | traffic jam? | 23:40 |
tellesnobrega | there was a protest due to an accident that happened friday on the highway | 23:40 |
tosky | I see | 23:41 |
tellesnobrega | I was stuck from 11:30h until 17h | 23:41 |
tosky | and a lot of people with you, I guess | 23:42 |
tellesnobrega | yes | 23:42 |
tosky | a bad mishap, but at least it's over | 23:56 |
tosky | the power seems to work, I ninja approved one change (maybe two) coming from the bots now that sahara-dashboard works | 23:56 |
tosky | and thanks! | 23:56 |
* tosky disappears | 23:56 | |
*** tosky has quit IRC | 23:56 | |
*** shuyingya has joined #openstack-sahara | 23:57 | |
*** jeremyfreudberg has joined #openstack-sahara | 23:57 | |
tellesnobrega | see ya | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!