| *** xuhaiwei has joined #tripleo | 00:00 | |
| xuhaiwei | morning | 00:00 |
|---|---|---|
| *** lazy_prince has joined #tripleo | 00:09 | |
| *** lazy_prince is now known as killer_prince | 00:09 | |
| *** rpodolyaka1 has joined #tripleo | 00:11 | |
| lifeless | 2014-04-10 00:13:08.972 7720 INFO nova.metadata.wsgi.server [-] 10.22.157.168 "GET /latest/meta-data/ HTTP/1.1" status: 500 len: 343 time: 30.1532509 | 00:14 |
| lifeless | jogo: ^ | 00:14 |
| *** rpodolyaka1 has quit IRC | 00:15 | |
| *** matsuhashi has joined #tripleo | 00:26 | |
| *** yamahata has quit IRC | 00:46 | |
| *** jrist has quit IRC | 00:52 | |
| *** jrist has joined #tripleo | 00:52 | |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-image-elements: Ensure reset-db is only run on the bootstrap host https://review.openstack.org/86479 | 00:56 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-image-elements: Ensure reset-db is only run on the bootstrap host https://review.openstack.org/86479 | 00:57 |
| *** yamahata has joined #tripleo | 01:03 | |
| *** kevinbenton has joined #tripleo | 01:05 | |
| *** hewbrocca has quit IRC | 01:05 | |
| greghaynes | SpamapS: just occured to me that https://review.openstack.org/#/c/84197/ is going to conflict pretty nastily with your software-config patch... | 01:06 |
| *** CaptTofu has quit IRC | 01:07 | |
| *** CaptTofu has joined #tripleo | 01:07 | |
| *** morazi has quit IRC | 01:10 | |
| *** CaptTofu has quit IRC | 01:11 | |
| *** killer_prince has quit IRC | 01:15 | |
| *** corrigac has joined #tripleo | 01:18 | |
| xuhaiwei | greghaynes: have you solved known the error reason? | 01:19 |
| *** wendar_ has joined #tripleo | 01:19 | |
| *** GheRivero_ has joined #tripleo | 01:19 | |
| *** SpamapS_ has joined #tripleo | 01:21 | |
| *** StevenK_ has joined #tripleo | 01:21 | |
| *** proffalk1n has joined #tripleo | 01:22 | |
| *** lynxman_ has joined #tripleo | 01:22 | |
| *** juice has quit IRC | 01:22 | |
| *** vipul has quit IRC | 01:22 | |
| *** spiffxp has quit IRC | 01:22 | |
| *** SpamapS has quit IRC | 01:22 | |
| *** proffalken has quit IRC | 01:22 | |
| *** StevenK has quit IRC | 01:22 | |
| *** antonym has quit IRC | 01:23 | |
| *** lynxman has quit IRC | 01:23 | |
| *** mordred has quit IRC | 01:23 | |
| *** dividehex has quit IRC | 01:23 | |
| *** ccorrigan has quit IRC | 01:23 | |
| *** wendar has quit IRC | 01:23 | |
| *** GheRivero has quit IRC | 01:23 | |
| *** vipuls has joined #tripleo | 01:24 | |
| *** mordred has joined #tripleo | 01:24 | |
| *** lynxman_ is now known as lynxman | 01:24 | |
| *** dividebin has joined #tripleo | 01:24 | |
| *** lynxman has quit IRC | 01:24 | |
| *** lynxman has joined #tripleo | 01:24 | |
| *** spiffxp has joined #tripleo | 01:24 | |
| *** StevenK_ is now known as StevenK | 01:24 | |
| *** juice has joined #tripleo | 01:24 | |
| *** antonym has joined #tripleo | 01:25 | |
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: Update neutron's selinux policies https://review.openstack.org/82980 | 01:31 |
| *** lazy_prince has joined #tripleo | 01:38 | |
| *** lazy_prince is now known as killer_prince | 01:38 | |
| *** CaptTofu has joined #tripleo | 01:41 | |
| *** nosnos has joined #tripleo | 01:48 | |
| *** nati_ueno has quit IRC | 02:06 | |
| *** CaptTofu has quit IRC | 02:20 | |
| *** spzala has quit IRC | 02:24 | |
| *** ramishra_ has joined #tripleo | 02:24 | |
| *** ramishra_ has quit IRC | 03:00 | |
| *** cody-somerville has quit IRC | 03:07 | |
| *** ramishra has joined #tripleo | 03:08 | |
| *** saurabhs has left #tripleo | 03:15 | |
| *** matsuhashi has quit IRC | 03:22 | |
| tchaypo | morning xuhaiwei | 03:22 |
| tchaypo | xuhaiwei: which error? | 03:22 |
| *** fandi has joined #tripleo | 03:24 | |
| *** nati_ueno has joined #tripleo | 03:28 | |
| *** matsuhashi has joined #tripleo | 03:32 | |
| *** CaptTofu has joined #tripleo | 03:33 | |
| xuhaiwei | tchaypo: hi | 03:35 |
| xuhaiwei | tchaypo: I pasted the error message here http://paste.openstack.org/show/75458/ | 03:36 |
| *** ramishra has quit IRC | 03:37 | |
| *** CaptTofu has quit IRC | 03:38 | |
| xuhaiwei | tchaypo: lifeless said it seemed to have something with garela | 03:40 |
| tchaypo | ah | 03:41 |
| tchaypo | ln: failed to create symbolic link '/lib/i686-linux-gnu/libssl.so.6': No such file or directory | 03:41 |
| tchaypo | i knwo that greghaynes made a change yesterday that switched from creating a symlink called libssl.so.10 to libssl.so.6, but then other things were looking for libssl.so.10 still. I'm wondering if he might have changed ti back | 03:41 |
| clarkb | heartbleed breaks everything | 03:42 |
| tchaypo | it would be in the history for the tripleo-image-elements repo - could you take a look there to see if you can see the changes from greghaynes ? | 03:42 |
| *** eghobo has joined #tripleo | 03:43 | |
| *** BadCub01 has quit IRC | 03:43 | |
| xuhaiwei | You mean I should change libssl.so.10 back to libssl.so.6 ? | 03:44 |
| tchaypo | I'm not sure. That's the first thing that comse to mind because I know greghaynes was working on that, but I don't know the details of what he was doing, or what the final state was. This could be completely unrelated, it's just the first thing I think of when I see that particular file mentioned | 03:45 |
| xuhaiwei | libssl.so.10 helps me solve another error | 03:46 |
| xuhaiwei | it changed it back, that error will happen again, i think | 03:47 |
| * StevenK ponders arguing for non-module imports on the list | 03:56 | |
| tchaypo | I've seen cases where I think they're abused, but lots of python modules are designed to be imported piecemeal. | 03:57 |
| tchaypo | datetime.datetime being one example of a module that really only contains one useful thing with the same name as the module; the DSLs mentioned on the thread being another | 03:58 |
| clarkb | StevenK: there is a thread you can dig up about it | 03:58 |
| tchaypo | there's an active thread going on right now, no need to dig | 03:58 |
| clarkb | StevenK: i think the issue is fairly well settled :) (not that we can't shift opinions over time) | 03:58 |
| tchaypo | personally I'd like to switch from 4-space indents to 2-space indents, but that's a lost cause | 03:59 |
| StevenK | tchaypo: HEATHEN | 03:59 |
| tchaypo | no-one anywhere agrees with me | 03:59 |
| clarkb | oh the current thread is the old thread just repinged | 03:59 |
| tchaypo | ZOMBIE THREAD | 03:59 |
| StevenK | http://bazaar.launchpad.net/~launchpad-pqm/launchpad/devel/view/head:/lib/lp/soyuz/model/binarypackagebuildbehaviour.py | 03:59 |
| clarkb | tchaypo: we should switch to tabs then you can set tab width to whatever you like | 03:59 |
| StevenK | clarkb: You too | 04:00 |
| StevenK | That ^ uses function and class importing, and I think it looks quite readable | 04:00 |
| clarkb | though I will never forget 3 space tabs, why anyone would use 3 spaces tabs I have no idea | 04:00 |
| tchaypo | :q | 04:01 |
| clarkb | StevenK: I think the key take away from that thread is you can ignore Hwhatever | 04:01 |
| tchaypo | clarkb: that's not just heathen, it's positively unnatural and against the laws of nature. | 04:02 |
| clarkb | and projects that don't obviously see benefit in enforcing the rule | 04:02 |
| clarkb | tchaypo: let me see if I can find it | 04:02 |
| StevenK | clarkb: I'd love to ignore H*, but upstream hacking enforces it | 04:02 |
| StevenK | So the braindamage extends far | 04:02 |
| clarkb | StevenK: what does upstream hacking matter? | 04:03 |
| clarkb | infra pretty much ignores H* everywhere and has no issues | 04:03 |
| StevenK | New projects get it by default | 04:03 |
| StevenK | And then it requires effort to turn off | 04:03 |
| tchaypo | my guess is that the difference is that infra isn't trying to make a thing that will be distributed as part of the bundled core | 04:04 |
| clarkb | tchaypo: let me introduce you to lush2 http://sourceforge.net/p/lush/code/HEAD/tree/lush2/src/main.c | 04:04 |
| clarkb | as an undergrad conned into debugging grad student code it was the bane of my existence | 04:04 |
| clarkb | and explicitly uses 3 space indentation levels | 04:04 |
| tchaypo | oh, but that's c | 04:05 |
| StevenK | WAT | 04:05 |
| tchaypo | i have no expectations of sanity from languags where indentation is a matter of religious preference | 04:05 |
| StevenK | So I tend to use 2 space indent for C | 04:05 |
| clarkb | tchaypo: StevenK: I don't think bundled core has anything to do with what hacking rules you use | 04:05 |
| tchaypo | goto fail; | 04:05 |
| StevenK | I will occasionally use it for Perl | 04:05 |
| tchaypo | is all i'm going to say about indentation in c | 04:05 |
| StevenK | Braceless ifs were a stupid idea | 04:06 |
| clarkb | tchaypo: even better it is C written by and for lispers | 04:07 |
| StevenK | clarkb: I know I keep tripping over H* rules | 04:07 |
| clarkb | StevenK: I think it was the second time the period at end of first line of commit got tripped on an infra repo that all of H* got ignored | 04:07 |
| StevenK | Haha | 04:08 |
| *** nosnos has quit IRC | 04:09 | |
| openstackgerrit | Steve Kowalik proposed a change to openstack/tripleo-incubator: Build os-cloud-config in setup-clienttools https://review.openstack.org/84935 | 04:09 |
| tchaypo | bah. | 04:12 |
| tchaypo | trying to fix indentation on some docs in devtest_overcloud.sh | 04:12 |
| tchaypo | I can leave it correctly indented for bash control flow, but that adds 4 spaces too many at start of line in the generated .rst | 04:12 |
| tchaypo | or i can fix the .rst by removing the indentation in the .sh | 04:13 |
| StevenK | tchaypo: Moving to 3 space indent to troll clarkb? | 04:13 |
| tchaypo | or, i can copy and paste 5 linutes and duplicate them inside a comment, creating maintenance debt. | 04:13 |
| tchaypo | OR i could add a new hashtag to extract-docs.awk | 04:14 |
| tchaypo | #dedent == don't add the usual 4 spaces at the start of the line. | 04:14 |
| StevenK | Can we run a2p over extract-docs.awk ? | 04:14 |
| tchaypo | dunno. Does a2p have a setting to produce 3-space-indent puke? | 04:15 |
| tchaypo | i mean, perl? | 04:15 |
| *** tzumainn has quit IRC | 04:17 | |
| StevenK | Haha | 04:17 |
| StevenK | Cute comments that say "# Here is a workaround because awk is dumb" are | 04:17 |
| StevenK | passed through unmodified. | 04:17 |
| StevenK | tchaypo: a2p does nothing about indenting | 04:18 |
| StevenK | tchaypo: 'man perlguts' when you're brave | 04:19 |
| tchaypo | meh. So I can make dedent work, but it only dedents once; if something needs to be twice dedented, I got nothing. | 04:21 |
| xuhaiwei | tchaypo: even I changed libssl.so.10 back to libssl.so.6, this error happened again df: Warning: cannot read table of mounted file systems: No such file or directory | 04:22 |
| StevenK | << in vi? :-P | 04:22 |
| tchaypo | Worth adding a partial tweak? | 04:22 |
| StevenK | xuhaiwei: That's a warning | 04:22 |
| xuhaiwei | and after that, error : ln: failed to create symbolic link '/lib/i686-linux-gnu/libssl.so.10': No such file or directory | 04:22 |
| xuhaiwei | here is still libssl.so.10 | 04:23 |
| StevenK | xuhaiwei: ls -l /lib/i686-linux-gnu/libssl.so.* ? | 04:23 |
| xuhaiwei | ls: cannot access /lib/i686-linux-gnu/libssl.so.*: No such file or directory | 04:24 |
| tchaypo | xuhaiwei: does /lib/i686-linux-gnu/ exist? | 04:25 |
| xuhaiwei | no | 04:25 |
| *** cwolferh_ has joined #tripleo | 04:27 | |
| tchaypo | so it's going to be hard to create a symlink inside it. I guess we need to figure out why it doesn't exist | 04:27 |
| StevenK | xuhaiwei: What about uname -m ? | 04:28 |
| StevenK | Wait. That has to be i686 | 04:28 |
| xuhaiwei | x86_64 | 04:28 |
| StevenK | Oh | 04:28 |
| StevenK | Wat | 04:28 |
| tchaypo | as a really basic question - are you on an i686 system? maybe you have /lib/i386-linux-gnu or /lib/amd_64-gnu-lib (my system has both) | 04:28 |
| * tchaypo took too long to type | 04:29 | |
| *** cwolferh has quit IRC | 04:29 | |
| *** fandi has quit IRC | 04:30 | |
| xuhaiwei | I have this: x86_64-linux-gnu | 04:31 |
| StevenK | steven@undermined:~% ls -ldh /lib/*linux-gnu | 04:31 |
| StevenK | drwxr-xr-x 5 root root 20K Apr 9 11:52 /lib/x86_64-linux-gnu | 04:31 |
| StevenK | amd64 is a Debian/Ubuntu ism | 04:32 |
| xuhaiwei | this script has to be run on i686? | 04:33 |
| StevenK | xuhaiwei: It should cope with both | 04:33 |
| xuhaiwei | now it doesn't work on amd64? | 04:34 |
| StevenK | It should be running on the seed | 04:34 |
| StevenK | xuhaiwei: Let me try locally | 04:35 |
| xuhaiwei | thanks | 04:35 |
| StevenK | Blah, which results in a new cloud image | 04:36 |
| StevenK | But it's actually downloading quickly today | 04:36 |
| StevenK | Bring on the NBN. But only after Turnbull has been drawn, quartered and hung. | 04:37 |
| *** ramishra has joined #tripleo | 04:38 | |
| tchaypo | there are nbnco pits in wolli creek now | 04:39 |
| tchaypo | afaict it's just the new buildings under development at this point, they aren't digging up the rest of the suburb | 04:39 |
| *** ramishra has quit IRC | 04:42 | |
| *** fandi has joined #tripleo | 04:43 | |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-incubator: Make it possible to just build images. https://review.openstack.org/85222 | 04:47 |
| *** nosnos has joined #tripleo | 04:47 | |
| *** ramishra has joined #tripleo | 04:50 | |
| *** akuznetsov has joined #tripleo | 04:54 | |
| tchaypo | StevenK: https://review.openstack.org/#/c/84935/ - gerrit still doesn't like your code even though the dependent change has landed | 04:56 |
| StevenK | tchaypo: See -infra | 04:57 |
| *** sdake has quit IRC | 04:57 | |
| *** akuznets_ has joined #tripleo | 05:01 | |
| *** akuznetsov has quit IRC | 05:03 | |
| tchaypo | time for me to grab a coffee and stretch the legs. | 05:04 |
| tchaypo | bbs | 05:04 |
| *** russell_h has quit IRC | 05:09 | |
| *** russell_h has joined #tripleo | 05:10 | |
| *** russell_h has joined #tripleo | 05:10 | |
| *** giulivo has joined #tripleo | 05:15 | |
| *** ilives has joined #tripleo | 05:19 | |
| *** CaptTofu has joined #tripleo | 05:21 | |
| lifeless | o/ | 05:24 |
| *** kevinbenton has quit IRC | 05:25 | |
| *** ilives has quit IRC | 05:25 | |
| *** ilives has joined #tripleo | 05:25 | |
| *** CaptTofu has quit IRC | 05:25 | |
| *** kevinbenton has joined #tripleo | 05:30 | |
| lifeless | StevenK: / tchaypo: are one ofyou hacking on the pass-through config stuff ? | 05:31 |
| *** eghobo has quit IRC | 05:37 | |
| *** eghobo has joined #tripleo | 05:38 | |
| *** ramishra has quit IRC | 05:40 | |
| *** cwolferh_ has quit IRC | 05:44 | |
| *** ilives has quit IRC | 05:47 | |
| *** ilives has joined #tripleo | 05:47 | |
| StevenK | lifeless: I am not, I am hacking on os-cloud-config | 05:48 |
| *** ilives has quit IRC | 05:49 | |
| *** eguz has joined #tripleo | 05:51 | |
| *** ilives has joined #tripleo | 05:53 | |
| *** eghobo has quit IRC | 05:55 | |
| *** kevinbenton has quit IRC | 06:05 | |
| *** ilives has quit IRC | 06:08 | |
| xuhaiwei | StevenK: have you made the error happened in your local environment? | 06:08 |
| StevenK | xuhaiwei: No, in fact. | 06:08 |
| StevenK | My local machine is bringing up the overcloud now | 06:08 |
| xuhaiwei | :( | 06:09 |
| tchaypo | lifeless: i am not, because I don't understand what's needed | 06:11 |
| tchaypo | lifeless: but since you're here | 06:11 |
| tchaypo | is https://review.openstack.org/#/c/83409/ one of the issues you were talking about? | 06:11 |
| *** ramishra has joined #tripleo | 06:13 | |
| lifeless | tchaypo: ok; did you read the etherpad ? | 06:13 |
| lifeless | tchaypo: it isn't | 06:14 |
| *** sdake has joined #tripleo | 06:15 | |
| *** kevinbenton has joined #tripleo | 06:16 | |
| *** lsmola has joined #tripleo | 06:16 | |
| *** ewindisch has quit IRC | 06:16 | |
| tchaypo | I'm not sure if I read it before; a lot of it looks familiar but I'm not sure if it's from there or from the thread | 06:20 |
| tchaypo | Can you point me at some of the changes you're referring to? | 06:20 |
| *** SpamapS_ is now known as SpamapS | 06:20 | |
| *** SpamapS has joined #tripleo | 06:20 | |
| SpamapS | hm | 06:20 |
| SpamapS | still getting config.json instead of local.json in my seed | 06:20 |
| SpamapS | which isn't just deprecated, at this point, that will flat out _break_ nova. | 06:21 |
| mordred | SpamapS: I blame vodka | 06:23 |
| lifeless | SpamapS: mordred: oh his | 06:24 |
| lifeless | hies | 06:24 |
| lifeless | so, the mysql-slow-query log has lots of statements < 10 seconds | 06:24 |
| lifeless | which is odd AIUI | 06:24 |
| lifeless | SpamapS: tomorrow morning has exploded for me | 06:24 |
| SpamapS | lifeless: we want anything 1.0s or more IIRC | 06:25 |
| lifeless | SpamapS: # Thread_id: 33 Schema: ovs_neutron Last_errno: 0 Killed: 0 | 06:26 |
| lifeless | # Query_time: 0.000217 Lock_time: 0.000081 Rows_sent: 1 Rows_examined: 2 Rows_affected: 0 Rows_read: 2 | 06:26 |
| lifeless | for instance | 06:26 |
| lifeless | is being logged on this server I'm poking at | 06:26 |
| lifeless | SpamapS: I'm looking into the 'neutron goes lalalalalala' thing, since turns out that that breaks 40 node nova-bm deploys | 06:26 |
| lifeless | SpamapS: (badly) | 06:26 |
| lifeless | SpamapS: anyhow tomorrow, I need to head to the office around 0645 | 06:27 |
| mordred | lifeless: do you have the slow log somewhere I can get to it? (don't say vpn, I can't do it) | 06:27 |
| lifeless | SpamapS: so - see you at the office | 06:27 |
| SpamapS | lifeless: ah, well that may be a different default for percona thing. | 06:27 |
| lifeless | mordred: its 500M | 06:27 |
| lifeless | SpamapS: the --help --verbose still says 10 for that | 06:27 |
| mordred | lifeless: jesus | 06:27 |
| SpamapS | lifeless: ah, cool, I will reinstate my car reservation then | 06:27 |
| lifeless | SpamapS: (sorry) | 06:27 |
| mordred | lifeless: you know about mysqldumpslow? | 06:27 |
| *** nati_ueno has quit IRC | 06:27 | |
| lifeless | mordred: yes | 06:27 |
| mordred | kk. just making sure | 06:28 |
| lifeless | mordred: reported everything as being 0 seconds long. | 06:28 |
| SpamapS | lifeless: 10 may be 10ms ... though 0.000217 would be 217us IIRC | 06:28 |
| lifeless | mordred: At which point I went 'fuckme I need help' | 06:28 |
| *** ewindisch has joined #tripleo | 06:28 | |
| mordred | SpamapS: does modern mysql have the microslow patch applied? | 06:28 |
| SpamapS | lifeless: percona-toolkit has some tools to turn that into stats | 06:28 |
| SpamapS | mordred: percona sure as hell does | 06:28 |
| SpamapS | so is 10 actually 10us ?? | 06:29 |
| lifeless | but I did find some lovely slow queries via regex | 06:29 |
| lifeless | # User@Host: nova[nova] @ localhost [] | 06:29 |
| lifeless | # Thread_id: 63 Schema: nova Last_errno: 0 Killed: 0 | 06:29 |
| lifeless | # Query_time: 37.476880 Lock_time: 0.000000 Rows_sent: 0 Rows_examined: 0 Rows_affected: 0 Rows_read: 0 | 06:29 |
| lifeless | # Bytes_sent: 11 | 06:29 |
| lifeless | SET timestamp=1397033436; | 06:29 |
| lifeless | commit; | 06:29 |
| lifeless | mordred: I'm generating a new one with problem reproduced | 06:29 |
| tchaypo | To be frank, when I read https://etherpad.openstack.org/p/tripleo-config-passthroughpart of my reaction is to glaze over because of all the references to things I don't understand (eg, I've never used moustache, have only a very simplistic understanding of heat); part of my reaction is to glaze over because the etherpad doesn't seem to have a plan, it seems to have several competing proposed plans that I | 06:30 |
| tchaypo | don't have think I understand well enough to choose between; and partly it's to glaze over because I'm not sure what the changes we're talking about look like | 06:30 |
| lifeless | ok, so I think spamaps has said alexis' proposal doesn't fly, and I agree | 06:31 |
| *** stewart has joined #tripleo | 06:31 | |
| stewart | lifeless: mordred said you may have qs for me | 06:31 |
| tchaypo | The first part of that is similar to how I felt trying follow greghaynes's percona work. The best thing I've found for fixing that so far is looking at reviews and going "huh, how does that work?" and tracing through the code and looking at how bits hang together, so that's what I've been doing | 06:32 |
| mordred | lifeless: look - I brought the cavalry | 06:32 |
| lifeless | tchaypo: ok | 06:33 |
| mordred | stewart: if lifeless generated a 50M slow query log - is there a percona thing that helps him find the suck? | 06:33 |
| lifeless | tchaypo: so I've deleted alexislee's commentary - it should be more approacahable | 06:33 |
| stewart | mordred: yes, there is. | 06:33 |
| lifeless | tchaypo: the stuff frmo stevebaker is talking about how we might do passthrough for the heat template layer | 06:33 |
| stewart | lifeless, mordred: let me find the url of docs of the thing. | 06:34 |
| lifeless | stewart: thats 500M | 06:34 |
| mordred | lifeless: jesus | 06:34 |
| tchaypo | lifeless: are there open reviews showing some of the problems we're trying to solve? | 06:34 |
| lifeless | tchaypo: most of the 170 that are open | 06:34 |
| lifeless | tchaypo: all of the ones in tripleo-heat-templates and tripleo-image-elements that say 'make X configurable' | 06:35 |
| tchaypo | I'm looking in the wrong place then. one moment... | 06:35 |
| stewart | lifeless: as long as MB and not GB, likely fine :) | 06:35 |
| lifeless | tchaypo: most of those *don't want* to be part of the URL | 06:35 |
| lifeless | bbaj | 06:35 |
| lifeless | tchaypo: part of the UI for the system | 06:35 |
| stewart | lifeless, mordred: http://www.percona.com/doc/percona-toolkit/2.2/pt-query-digest.html | 06:35 |
| tchaypo | right. I've been looking at tripleo-incubator rather than t-i-e. no wonder I'm confused. | 06:35 |
| *** jprovazn has joined #tripleo | 06:36 | |
| stewart | lifeless, mordred: it attempts to do magic to extract out parameter bits to queries and give you what class of queries cause pain. | 06:36 |
| lifeless | like, our job is to make consistent working deployments, not to explicitly model every bit of the openstack world: passthrough is intended to let folk with lots of tweaks do so without the UI for tripleo becoming 700 settings of madness | 06:36 |
| lifeless | stewart: so here's the symptoms | 06:36 |
| lifeless | stewart: we deploy lots of machines via nova baremetal | 06:36 |
| lifeless | stewart: and neutron goes to lala land | 06:37 |
| lifeless | stewart: e.g. | 06:37 |
| lifeless | stewart: https://etherpad.openstack.org/p/slow-bm-deploy | 06:37 |
| tchaypo | using my newfound knowledge from 30 seconds of reading about moustache, and the fact that it's come up in this conversation, I'm guessing that https://review.openstack.org/#/c/77929/2/elements/glance/os-config-applier/etc/glance/glance-api.conf is a moustache template | 06:37 |
| lifeless | it is | 06:38 |
| tchaypo | and looking at that file, it looks like what we want is to stop having to put everything into that template by hand | 06:39 |
| SpamapS | lifeless: it may not show as 1 bad query, but many queries done in succession.. "client side IN()" is what I call that. | 06:39 |
| tchaypo | but instaed have something that takes a key:value mapping and iterates over it, turning key:value into a "key = value" line? | 06:39 |
| lifeless | SpamapS: its a clear stall in strace | 06:39 |
| lifeless | SpamapS: see the etherpad | 06:39 |
| SpamapS | you and your etherpads | 06:39 |
| SpamapS | I have so many etherpad tabs open | 06:39 |
| SpamapS | etherpad probably thinks I'm a DDoS'er | 06:40 |
| *** cody-somerville has joined #tripleo | 06:40 | |
| lifeless | SpamapS: I'm thinking agent vs neutron-server lock fight | 06:40 |
| stewart | lifeless: stall in getting things from mysqld? | 06:41 |
| *** sdake has quit IRC | 06:41 | |
| lifeless | stewart: did you look at the etherpad :P | 06:41 |
| stewart | lifeless: as in, that's strace of something talknig to a mysqld, right.... | 06:41 |
| stewart | lifeless: ahh.. no... i see | 06:41 |
| lifeless | thats a strace of neutron-server | 06:42 |
| lifeless | which is talking to a mysqld | 06:42 |
| stewart | lifeless: ahh.. so wtf... 0 timeout to poll, that's fine... | 06:42 |
| lifeless | it sends a commit message | 06:42 |
| lifeless | reads | 06:42 |
| lifeless | sits there for 44 seconds then times out (how?!) | 06:42 |
| lifeless | then issues a rollback | 06:42 |
| SpamapS | lifeless: oh | 06:44 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Make libvirt volume pool configurable https://review.openstack.org/83409 | 06:44 |
| SpamapS | lifeless: this heartbeat.. | 06:44 |
| tchaypo | hah. I mention that change in channel and minutes later it's suddenly +A ;) | 06:45 |
| lifeless | we definitely has some IO contention with the image unpacking IMO | 06:45 |
| SpamapS | lifeless: that does look like something that may raise the gap locking problem. | 06:45 |
| stewart | lifeless: is that read took 44 seconds... or poll. or the gap between the two calls. | 06:45 |
| * stewart has to look at man page of strace | 06:45 | |
| lifeless | but this can occur even we're just doing 4 instances | 06:45 |
| SpamapS | stewart: the client was polling for 44 seconds. | 06:45 |
| lifeless | stewart: thats -r | 06:45 |
| *** mrunge has joined #tripleo | 06:45 | |
| lifeless | stewart: relative timestamp since the line before | 06:46 |
| stewart | SpamapS, lifeless: wtf... poll with 0 for timeout should return immediately... | 06:46 |
| lifeless | server has 96G of RAM and 24 cores | 06:47 |
| stewart | lifeless: ahh.. small machine (i may be working with big machines atm :) | 06:47 |
| lifeless | :P | 06:50 |
| lifeless | big enough for this, one things | 06:50 |
| stewart | lifeless: it's odd... i mean... it looks like it's either getting stuck on the read (i.e. waiting for COMMIT to happen) or something happening not related to either of that | 06:50 |
| stewart | lifeless: now.. COMMIT taknig a while could be possible... | 06:50 |
| lifeless | I'm trapped another in strace | 06:50 |
| lifeless | but I'm runnning a longer strace because I think I saw a different table | 06:51 |
| SpamapS | stewart: we're not very tuned, so it is entirely possible that we have some really simple contention going on. | 06:51 |
| stewart | lifeless: it could be interesting to see what that whole transaction is and what it modifies.... | 06:51 |
| SpamapS | lifeless: also, the thing you're stracing, is it using any os level threads? | 06:51 |
| lifeless | SpamapS: neutron-server | 06:51 |
| stewart | lifeless: if you grab percona-toolkit and run pt-query-digest on it, it may show up stuff. | 06:51 |
| lifeless | I need to sleep or I'll be shite tomorrow | 06:51 |
| SpamapS | lifeless: yeah me too actually | 06:52 |
| lifeless | stewart: will tomorrow | 06:52 |
| SpamapS | have ot be up in just over 5 hours | 06:52 |
| SpamapS | have to | 06:52 |
| stewart | lifeless, SpamapS: no worries. i should stop work at some point too :) | 06:53 |
| stewart | lifeless, SpamapS: ping me tomorrow if anything else can help with | 06:53 |
| *** ilives has joined #tripleo | 06:55 | |
| SpamapS | stewart: cool ty | 06:56 |
| * SpamapS heads off to bed | 06:56 | |
| lifeless | more straces stuff up | 06:57 |
| lifeless | note that the poll()'s are blocking the eventloop | 06:58 |
| lifeless | because its not threaded | 06:58 |
| lifeless | and thats not polling the wsgi socket | 06:58 |
| *** cody-somerville has quit IRC | 06:58 | |
| SpamapS | lifeless: yeah, well known problem that waiting on SQL queries screws everything up | 06:58 |
| SpamapS | "because cpython" | 06:59 |
| SpamapS | lifeless: I wonder if we're overwhelming the transaction log and thus waiting on fsyncs | 06:59 |
| lifeless | SpamapS: because !threads :P | 06:59 |
| stewart | lifeless: that's probably not a non-blacknig socket... | 07:00 |
| stewart | lifeless: maybe try strace -T to get duration of syscalls? | 07:00 |
| SpamapS | but, the GIL the GIL ... it kills orphans | 07:00 |
| stewart | SpamapS, lifeless: switching to nonblocking client lib would be good :) | 07:00 |
| lifeless | stewart: ya think :) | 07:02 |
| lifeless | stewart: good call | 07:03 |
| lifeless | 6.587080 poll([{fd=20, events=POLLIN|POLLPRI}], 1, 0) = 0 (Timeout) <0.000136> | 07:03 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Fix libssl/crypto symlinks on Fedora https://review.openstack.org/86394 | 07:04 |
| stewart | lifeless: the read call before it? | 07:04 |
| lifeless | 0.000052 read(20, "\7\0\0\1\0\0\0\0\10\0\0", 16384) = 11 <65.004145> | 07:04 |
| lifeless | 65.004337 poll([{fd=20, events=POLLIN|POLLPRI}], 1, 0) = 0 (Timeout) <0.000012> | 07:04 |
| stewart | lifeless: aha! | 07:04 |
| stewart | lifeless: likely the COMMIt that's taking forever | 07:04 |
| stewart | lifeless: with luck, that should be in the slow query log. | 07:05 |
| lifeless | 0.000046 write(20, "\7\0\0\0\3commit", 11) = 11 <0.000023> | 07:05 |
| lifeless | 0.000052 read(20, "\7\0\0\1\0\0\0\0\10\0\0", 16384) = 11 <65.004145> | 07:05 |
| lifeless | 65.004337 poll([{fd=20, events=POLLIN|POLLPRI}], 1, 0) = 0 (Timeout) <0.000012> | 07:05 |
| lifeless | the disk is a tad busy | 07:05 |
| lifeless | Device: rrqm/s wrqm/s r/s w/s rkB/s wkB/s avgrq-sz avgqu-sz await r_await w_await svctm %util | 07:05 |
| lifeless | sda 0.00 13815.37 0.00 481.84 0.00 58047.11 240.94 146.84 298.27 0.00 298.27 2.07 99.80 | 07:05 |
| *** ilives has quit IRC | 07:06 | |
| lifeless | # Thread_id: 165 Schema: ovs_neutron Last_errno: 0 Killed: 0 | 07:07 |
| SpamapS | lifeless: do these things have any battery backed write cache? | 07:07 |
| lifeless | # Query_time: 65.004015 Lock_time: 0.000000 Rows_sent: 0 Rows_examined: 0 Rows_affected: 0 Rows_read: 0 | 07:07 |
| stewart | lifeless: so that could mean a few things... | 07:07 |
| lifeless | # Bytes_sent: 11 | 07:07 |
| lifeless | use ovs_neutron; | 07:07 |
| lifeless | SET timestamp=1397113451; | 07:07 |
| lifeless | commit; | 07:07 |
| *** ilives has joined #tripleo | 07:07 | |
| stewart | lifeless: either something else is thrashing teh disk making it impossible for innodb to sneak in an iop | 07:07 |
| stewart | lifeless: or it's actually big txn | 07:07 |
| stewart | lifeless: have a look at the other queries in that thread id | 07:07 |
| lifeless | stewart: in etherpad | 07:08 |
| lifeless | so we could look at ionicing the qemu-convert op | 07:09 |
| stewart | lifeless: hrm... that shouldn't be causing COMMIT to take a long time... | 07:09 |
| *** CaptTofu has joined #tripleo | 07:09 | |
| lifeless | not sure it would be effective | 07:09 |
| stewart | lifeless: iotop enlightening at all? perhaps it is other things that are drowning it out. | 07:10 |
| stewart | lifeless: what IO scheduler? | 07:10 |
| lifeless | stewart: there's more selects but no updates in the log for it | 07:10 |
| *** cody-somerville has joined #tripleo | 07:10 | |
| lifeless | stewart: so iotop didn't show anything terribly bad | 07:10 |
| lifeless | stewart: we already throttle nova-bm to do one dd of an image (disk -> network) and one qemu-img convert (qcow2 -> raw) at a time | 07:11 |
| lifeless | I *think* we also throttle the swft retrieval, will have to follow up on that | 07:11 |
| lifeless | for instance, the dd looks like: | 07:12 |
| lifeless | 16743 be/4 root 0.00 B/s 967.04 K/s 0.00 % 99.99 % dd if=/mnt/state/var/lib/nova/instances/insta~0-25c1a8586588-lun-1-part3 bs=1M oflag=direct | 07:12 |
| SpamapS | innodb_log_file_size == 64M .. should be sufficient for any transaction neutron would do | 07:12 |
| lifeless | the write there is all net | 07:12 |
| lifeless | iscsi | 07:12 |
| lifeless | ok so sleep | 07:12 |
| lifeless | SpamapS: lets deep dive tomorrow | 07:13 |
| SpamapS | innodb_flush_log_at_trx_commit = 2 .. which means flushes only forced 1 time per second. | 07:13 |
| SpamapS | lifeless: true, we're just costing eachother cognitive ability by not sleeping | 07:13 |
| * SpamapS vows to go to sleep now | 07:13 | |
| lifeless | night all | 07:13 |
| *** CaptTofu has quit IRC | 07:14 | |
| stewart | lifeless, SpamapS: iscsi colud be making things interesting.... | 07:14 |
| tchaypo | night lifeless | 07:14 |
| * tchaypo is reading about heat SoftwareConfig | 07:15 | |
| lifeless | stewart: oh, and deadline | 07:15 |
| *** hashar has joined #tripleo | 07:17 | |
| *** proffalk1n has quit IRC | 07:21 | |
| *** proffalken has joined #tripleo | 07:22 | |
| *** proffalken has quit IRC | 07:22 | |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-incubator: Standardise location of environment password/rc files. https://review.openstack.org/83250 | 07:22 |
| *** proffalken has joined #tripleo | 07:24 | |
| *** zigo has quit IRC | 07:29 | |
| greghaynes | xuhaiwei: The error you are seeing makes me think you are running an out of date tripleo-image-elements | 07:32 |
| greghaynes | xuhaiwei: see https://review.openstack.org/#/c/86223/ | 07:33 |
| xuhaiwei | ok | 07:33 |
| *** jcoufal has joined #tripleo | 07:35 | |
| *** ramishra has quit IRC | 07:36 | |
| greghaynes | additionally, I think you are running a 32bit devtest, not 64bit | 07:36 |
| *** julim_WFH has quit IRC | 07:38 | |
| *** e0ne has joined #tripleo | 07:40 | |
| *** julim has joined #tripleo | 07:41 | |
| *** ramishra has joined #tripleo | 07:44 | |
| *** eguz has quit IRC | 07:49 | |
| *** jp_at_hp has joined #tripleo | 07:51 | |
| *** stewart has quit IRC | 07:51 | |
| *** e0ne has quit IRC | 07:51 | |
| *** stewart has joined #tripleo | 07:52 | |
| *** e0ne_ has joined #tripleo | 07:53 | |
| xuhaiwei | greghaynes: that error is gone, this time: error: failed to get domain 'seed' error: Domain not found: no domain with matching name 'seed' | 07:56 |
| *** petertoft has joined #tripleo | 07:58 | |
| openstackgerrit | Nicholas Randon proposed a change to openstack/tripleo-incubator: Allow IP ranges used by devtest to be configurable https://review.openstack.org/86308 | 08:00 |
| *** e0ne has joined #tripleo | 08:02 | |
| *** e0ne_ has quit IRC | 08:05 | |
| *** jistr has joined #tripleo | 08:07 | |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: Failure in rabbitmq os-refresh-config on master https://review.openstack.org/86034 | 08:08 |
| *** andreaf has joined #tripleo | 08:12 | |
| *** lucasagomes has joined #tripleo | 08:13 | |
| *** xuhaiwei has quit IRC | 08:20 | |
| *** matsuhashi has quit IRC | 08:21 | |
| *** matsuhashi has joined #tripleo | 08:21 | |
| *** zigo has joined #tripleo | 08:23 | |
| *** xuhaiwei has joined #tripleo | 08:23 | |
| *** matsuhashi has quit IRC | 08:25 | |
| *** killer_prince has quit IRC | 08:26 | |
| *** derekh has joined #tripleo | 08:26 | |
| *** lazy_prince has joined #tripleo | 08:27 | |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: Failure in rabbitmq os-refresh-config on master https://review.openstack.org/86034 | 08:27 |
| *** lazy_prince is now known as killer_prince | 08:27 | |
| *** markmc has joined #tripleo | 08:27 | |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: Failure in rabbitmq os-refresh-config on master https://review.openstack.org/86034 | 08:28 |
| *** lazy_prince has joined #tripleo | 08:29 | |
| *** bogdando has quit IRC | 08:29 | |
| *** bogdando has joined #tripleo | 08:30 | |
| *** killer_prince has quit IRC | 08:32 | |
| *** lazy_prince is now known as killer_prince | 08:32 | |
| *** matsuhashi has joined #tripleo | 08:37 | |
| openstackgerrit | Matthew Gilliard proposed a change to openstack/tripleo-heat-templates: Increase PXE boot timeout for overcloud https://review.openstack.org/86523 | 08:41 |
| *** athomas has joined #tripleo | 08:42 | |
| *** tzumainn has joined #tripleo | 08:44 | |
| *** lsmola has quit IRC | 08:46 | |
| *** tzumainn has quit IRC | 08:47 | |
| openstackgerrit | Matthew Gilliard proposed a change to openstack/tripleo-heat-templates: Increase PXE boot timeout for overcloud https://review.openstack.org/86523 | 08:47 |
| *** tzumainn has joined #tripleo | 08:47 | |
| *** athomas has quit IRC | 08:49 | |
| *** akuznets_ has quit IRC | 08:53 | |
| *** CaptTofu has joined #tripleo | 08:58 | |
| *** lsmola has joined #tripleo | 09:00 | |
| *** untriaged-bot has joined #tripleo | 09:00 | |
| untriaged-bot | Untriaged bugs so far: | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1302881 | 09:00 |
| uvirtbot | Launchpad bug 1302881 in tripleo "incloud CIDR can overlap custom baremetal-network" [Undecided,In progress] | 09:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1290488 | 09:00 |
| *** untriaged-bot has quit IRC | 09:00 | |
| uvirtbot | Launchpad bug 1290488 in tripleo "Baremetal: Invalid credentials" [Undecided,Incomplete] | 09:00 |
| *** akuznetsov has joined #tripleo | 09:00 | |
| *** CaptTofu has quit IRC | 09:03 | |
| *** morganfainberg is now known as morganfainberg_Z | 09:03 | |
| *** athomas has joined #tripleo | 09:04 | |
| *** akrivoka has joined #tripleo | 09:05 | |
| openstackgerrit | Matthew Gilliard proposed a change to openstack/tripleo-heat-templates: Increase PXE deployment timeout for overcloud https://review.openstack.org/86523 | 09:07 |
| *** akuznetsov has quit IRC | 09:10 | |
| *** akuznetsov has joined #tripleo | 09:12 | |
| *** martyntaylor has joined #tripleo | 09:16 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/tripleo-incubator: Register the Tuskar Service in the Undercloud https://review.openstack.org/85050 | 09:17 |
| *** jtomasek has joined #tripleo | 09:18 | |
| *** ilives has quit IRC | 09:18 | |
| *** ilives has joined #tripleo | 09:19 | |
| *** andreaf has quit IRC | 09:22 | |
| *** lazy_prince has joined #tripleo | 09:23 | |
| *** killer_prince has quit IRC | 09:24 | |
| *** lazy_prince is now known as killer_prince | 09:24 | |
| *** akrivoka has quit IRC | 09:24 | |
| *** akrivoka has joined #tripleo | 09:25 | |
| *** matsuhashi has quit IRC | 09:30 | |
| *** matsuhashi has joined #tripleo | 09:34 | |
| howleyt | jprovazn: Can you take a quick look at my response in https://review.openstack.org/#/c/86034/ ? I'll update patch after. Thanks. | 09:34 |
| *** jtomasek has quit IRC | 09:34 | |
| openstackgerrit | Matthew Gilliard proposed a change to openstack/tripleo-incubator: Allows user-specified PXE timeout for overcloud https://review.openstack.org/86539 | 09:36 |
| openstackgerrit | Ricardo Carrillo Cruz proposed a change to openstack/tripleo-incubator: Add 'Tested platforms' section in README.md https://review.openstack.org/84801 | 09:39 |
| openstackgerrit | Cian O'Driscoll proposed a change to openstack/tripleo-image-elements: Enable serial console access to nodes https://review.openstack.org/86298 | 09:39 |
| jprovazn | howleyt: sure, but I'm relocating right now, give me 60m | 09:39 |
| *** jprovazn has quit IRC | 09:40 | |
| *** ramishra has quit IRC | 09:47 | |
| *** jtomasek has joined #tripleo | 09:48 | |
| rcarrillocruz | GheRivero_: hi, uploaded new patchet for https://review.openstack.org/#/c/84801/5 | 09:51 |
| rcarrillocruz | may you please have a look when you have a moment? | 09:52 |
| rcarrillocruz | i removed 'supported' wording | 09:52 |
| *** hashar has quit IRC | 09:54 | |
| GheRivero_ | rcarrillocruz: what about "latest"? | 09:58 |
| GheRivero_ | :P | 09:58 |
| rcarrillocruz | heh | 09:58 |
| rcarrillocruz | also removed it | 09:58 |
| rcarrillocruz | hard to get consensus on this one | 09:58 |
| *** jml has quit IRC | 10:05 | |
| *** xuhaiwei has quit IRC | 10:05 | |
| *** jml has joined #tripleo | 10:10 | |
| *** jprovazn has joined #tripleo | 10:13 | |
| *** ilives has quit IRC | 10:16 | |
| *** ilives has joined #tripleo | 10:17 | |
| openstackgerrit | Matthew Gilliard proposed a change to openstack/tripleo-heat-templates: Increase PXE deployment timeout for overcloud https://review.openstack.org/86523 | 10:17 |
| rcarrillocruz | GheRivero_: i do not put versions as they will be always a moving target. Chances are when a new Ubuntu and Fedora version is released, so will the CI test platforms | 10:18 |
| rcarrillocruz | thanks | 10:19 |
| GheRivero_ | :) let's hope nobody else complains about it | 10:20 |
| *** matsuhashi has quit IRC | 10:20 | |
| rcarrillocruz | yeah, supportability is always a topic open for debate | 10:21 |
| *** matsuhashi has joined #tripleo | 10:25 | |
| openstackgerrit | Derek Higgins proposed a change to openstack-infra/tripleo-ci: Reduce size of log tarball's https://review.openstack.org/86550 | 10:30 |
| *** CaptTofu has joined #tripleo | 10:33 | |
| *** CaptTofu has quit IRC | 10:36 | |
| *** CaptTofu has joined #tripleo | 10:37 | |
| *** andreaf has joined #tripleo | 10:50 | |
| *** e0ne has quit IRC | 10:52 | |
| *** e0ne has joined #tripleo | 10:53 | |
| *** e0ne has quit IRC | 10:57 | |
| *** nosnos has quit IRC | 10:59 | |
| *** nosnos has joined #tripleo | 10:59 | |
| *** nosnos has quit IRC | 11:03 | |
| *** ramishra has joined #tripleo | 11:07 | |
| *** ramishra has quit IRC | 11:11 | |
| *** matsuhashi has quit IRC | 11:14 | |
| jp_at_hp | GheRivero_: Quick - APPROVE IT! All the tests are passed - it has 2 +2's! : https://review.openstack.org/#/c/83250/ | 11:17 |
| *** stewart has quit IRC | 11:18 | |
| jp_at_hp | marios: I think you could approve also! ^^^ | 11:18 |
| marios | jp_at_hp: /me looks | 11:18 |
| *** stewart has joined #tripleo | 11:19 | |
| jp_at_hp | :D - it's been a saga. but I'm hoping.. praying... sacrificing to any gods listening... | 11:19 |
| marios | jp_at_hp: lol @jp. you think _this_ is bad. go take a look at neutron. | 11:20 |
| jp_at_hp | marios: I am also a neutron engineer... | 11:20 |
| * jp_at_hp cries in the corner | 11:20 | |
| marios | jp_at_hp: ah :) i am still new. point in case: minor bug fix I picked up as a 'first thing in neutron' end of november @ https://review.openstack.org/#/c/59212/ | 11:21 |
| jp_at_hp | But - yay for improvements to the patch set. | 11:21 |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Standardise location of environment password/rc files. https://review.openstack.org/83250 | 11:21 |
| jp_at_hp | yes. and neutron has a whole load of 3rd party tests to worry about... | 11:22 |
| jp_at_hp | It helps us learn patience and understanding and grow as people :D | 11:22 |
| *** matsuhashi has joined #tripleo | 11:22 | |
| marios | jp_at_hp: yeah, have no idea why jenkins is foobaring now. it actually earlier +1d set 14 so no idea why failing now | 11:22 |
| marios | maybe i need a rebase | 11:22 |
| jp_at_hp | ah - the git rebase... | 11:23 |
| *** mkerrin has quit IRC | 11:25 | |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/diskimage-builder: Standardise manifest creation and retrieval https://review.openstack.org/83848 | 11:25 |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/tripleo-image-elements: Standardise manifest creation and retrieval https://review.openstack.org/83850 | 11:27 |
| openstackgerrit | Ricardo Carrillo Cruz proposed a change to openstack/tripleo-incubator: Add 'Tested platforms' section in README.md https://review.openstack.org/84801 | 11:27 |
| jp_at_hp | marios: if you have the time - I have 2 changes that needed rebasing onto the patch you just approved - https://review.openstack.org/#/c/83848/ and https://review.openstack.org/#/c/83850/. It would be great if you could take a look. | 11:29 |
| marios | jp_at_hp: will try later, middle of something atm | 11:30 |
| jp_at_hp | np - ta :D | 11:30 |
| *** matsuhashi has quit IRC | 11:31 | |
| *** mkerrin has joined #tripleo | 11:32 | |
| *** matsuhas_ has joined #tripleo | 11:33 | |
| *** CaptTofu has quit IRC | 11:38 | |
| openstackgerrit | Cian O'Driscoll proposed a change to openstack/tripleo-image-elements: Enable serial console access to nodes https://review.openstack.org/86298 | 11:40 |
| *** e0ne has joined #tripleo | 11:46 | |
| *** fandi has quit IRC | 11:48 | |
| *** e0ne has quit IRC | 11:48 | |
| *** e0ne has joined #tripleo | 11:49 | |
| *** e0ne has quit IRC | 11:53 | |
| howleyt | When can we expect to see support for scaling of notcompute/control nodes in the overcloud/undercloud? | 11:54 |
| howleyt | greghaynes: I believe you may be working on this? ^^ | 11:55 |
| *** matsuhas_ has quit IRC | 11:57 | |
| *** e0ne has joined #tripleo | 12:00 | |
| *** jistr has quit IRC | 12:07 | |
| *** jistr has joined #tripleo | 12:07 | |
| *** weshay has joined #tripleo | 12:09 | |
| *** hashar has joined #tripleo | 12:14 | |
| *** CaptTofu has joined #tripleo | 12:15 | |
| *** noslzzp has joined #tripleo | 12:30 | |
| *** jdob has joined #tripleo | 12:32 | |
| *** hewbrocca has joined #tripleo | 12:32 | |
| *** bauzas has joined #tripleo | 12:36 | |
| *** stewart has quit IRC | 12:38 | |
| *** stewart has joined #tripleo | 12:39 | |
| *** CaptTofu has quit IRC | 12:42 | |
| *** rlandy has joined #tripleo | 12:45 | |
| openstackgerrit | A change was merged to openstack/tripleo-incubator: Register the Tuskar Service in the Undercloud https://review.openstack.org/85050 | 12:46 |
| *** dprince has joined #tripleo | 12:50 | |
| *** lblanchard has joined #tripleo | 12:51 | |
| *** yamahata has quit IRC | 12:54 | |
| openstackgerrit | Eamonn O'Toole proposed a change to openstack/tripleo-image-elements: Add support for specifying swift ring directory range https://review.openstack.org/86580 | 12:57 |
| openstackgerrit | Coleman Corrigan proposed a change to openstack/tripleo-image-elements: Update write-pip-manifests git freeze to use -l https://review.openstack.org/86581 | 13:00 |
| openstackgerrit | Eamonn O'Toole proposed a change to openstack/tripleo-image-elements: Place swift-servers in appropriate zones https://review.openstack.org/86375 | 13:04 |
| *** morazi has joined #tripleo | 13:07 | |
| *** matty_dubs|gone is now known as matty_dubs | 13:08 | |
| *** mrunge has quit IRC | 13:11 | |
| *** dtantsur is now known as dtantsur|bbl | 13:11 | |
| openstackgerrit | Eamonn O'Toole proposed a change to openstack/tripleo-image-elements: Place swift-servers in appropriate zones https://review.openstack.org/86375 | 13:14 |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: Failure in rabbitmq os-refresh-config on master https://review.openstack.org/86034 | 13:19 |
| *** giulivo has quit IRC | 13:20 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Update keystone's selinux policies https://review.openstack.org/82981 | 13:21 |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-image-elements: Add keepalive option to rabbitmq config https://review.openstack.org/81920 | 13:23 |
| SpamapS | stewart: the iscsi in that case is actually from the machine -> other machines | 13:25 |
| jp_at_hp | derekh: my reading of the rhel element was that it was common code included by other more specific elements that provided a particular version. Can you confirm that you are asserting the rhel element is a standalone root-fs and operating system for use in image building and that it is in fact rhel7 that just customises this in ways particular to that version? | 13:26 |
| openstackgerrit | Eamonn O'Toole proposed a change to openstack/tripleo-heat-templates: Parametrizing the number of Swift replicas https://review.openstack.org/86076 | 13:27 |
| openstackgerrit | tom-howley proposed a change to openstack/tripleo-heat-templates: Add rabbitmq keepalive and keepalive parameters to overcloud/undercloud template https://review.openstack.org/81934 | 13:28 |
| *** giulivo has joined #tripleo | 13:28 | |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/tripleo-incubator: Doc updates suggested in another review https://review.openstack.org/86594 | 13:29 |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/tripleo-incubator: Make it possible to just build images. https://review.openstack.org/85222 | 13:29 |
| jp_at_hp | oops - I was trying to only upload the new patch, not update the dependecy :( | 13:31 |
| SpamapS | jp_at_hp: git review -R for that | 13:32 |
| SpamapS | jp_at_hp: git review always rebases | 13:32 |
| SpamapS | unless you pass -R | 13:32 |
| jp_at_hp | Hmmm... | 13:32 |
| jp_at_hp | SpamapS: I did git push gerrit HEAD:refs/for/master | 13:33 |
| jp_at_hp | because I had already rebased by cherry-picking onto master | 13:33 |
| *** yamahata has joined #tripleo | 13:33 | |
| jp_at_hp | I guess I shoulda used git review -d <change> to pull down the branch and then edited on top- workflow updated :D | 13:33 |
| SpamapS | git push whaaaa? | 13:33 |
| derekh | jp_at_hp: looking | 13:36 |
| jp_at_hp | Sorry derekh - it is likely to be me just misreading, but confirmation will be great :D | 13:36 |
| openstackgerrit | Ricardo Carrillo Cruz proposed a change to openstack/tripleo-incubator: Add 'Tested platforms' section in README.md https://review.openstack.org/84801 | 13:39 |
| jp_at_hp | derekh: sorry - I took a look. rhel has a root.d and downloads an image, I should have added it to begin with - updating now | 13:41 |
| derekh | jp_at_hp: no prob, ya from I can see both rhel and rhel7 provide a standalone root-fs, | 13:42 |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/diskimage-builder: Use provides to note an element provides an OS https://review.openstack.org/85773 | 13:42 |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/diskimage-builder: Use provides to note an element provides deploy https://review.openstack.org/86333 | 13:42 |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/diskimage-builder: Remove the temporary deploy element after rename https://review.openstack.org/86334 | 13:42 |
| derekh | jp_at_hp: but looks like the rhel7 element requires you to have the qcow on local disk | 13:42 |
| derekh | jp_at_hp: thanks | 13:43 |
| jp_at_hp | np - sorry I missed it to begin with | 13:43 |
| *** lblanchard has quit IRC | 13:44 | |
| derekh | jp_at_hp: no probs at all | 13:44 |
| *** lblanchard has joined #tripleo | 13:45 | |
| *** dricco has joined #tripleo | 13:47 | |
| *** lblanchard has quit IRC | 13:49 | |
| *** lblanchard has joined #tripleo | 13:49 | |
| jp_at_hp | Can I get some core opinioins on this please? https://review.openstack.org/#/c/82807/ - I want to decide whether to carry on with it or abandon it - seems to be split fairly 50/50 over whether we should have it. And if not - I'd love suggestions of something else to achieve the same goal... | 13:50 |
| SpamapS | jp_at_hp: will take a look in a bit.. have to sign off now as plane is landing. :) | 13:51 |
| openstackgerrit | Stuart McLaren proposed a change to openstack/tripleo-heat-templates: ssl: Bring up a virtual IP on the overcloud https://review.openstack.org/85100 | 13:51 |
| lifeless | morning | 13:52 |
| lxsli | SpamapS: a response to config discussion on the mailing list would be amazing | 13:52 |
| SpamapS | lxsli: will also take a loko at that. | 13:52 |
| lxsli | SpamapS: cheers | 13:52 |
| * SpamapS stows laptop | 13:53 | |
| *** CaptTofu has joined #tripleo | 13:54 | |
| *** dtynan has joined #tripleo | 13:55 | |
| openstackgerrit | Eamonn O'Toole proposed a change to openstack/tripleo-image-elements: Add support for specifying swift ring directory range https://review.openstack.org/86580 | 13:58 |
| openstackgerrit | Eamonn O'Toole proposed a change to openstack/tripleo-image-elements: Place swift-servers in appropriate zones https://review.openstack.org/86375 | 13:59 |
| *** hewbrocca has quit IRC | 14:01 | |
| *** jtomasek has quit IRC | 14:02 | |
| *** corrigac has quit IRC | 14:02 | |
| *** noslzzp has quit IRC | 14:02 | |
| *** corrigac1 has joined #tripleo | 14:05 | |
| *** e0ne has quit IRC | 14:07 | |
| *** e0ne has joined #tripleo | 14:08 | |
| *** morazi has quit IRC | 14:11 | |
| *** e0ne has quit IRC | 14:11 | |
| *** e0ne_ has joined #tripleo | 14:11 | |
| openstackgerrit | Stuart McLaren proposed a change to openstack/tripleo-image-elements: stunnel: Allow CA certificate to be specified https://review.openstack.org/85099 | 14:11 |
| openstackgerrit | Matthew Gilliard proposed a change to openstack/tripleo-heat-templates: Increase PXE deployment timeout for overcloud https://review.openstack.org/86523 | 14:13 |
| *** jtomasek has joined #tripleo | 14:14 | |
| openstackgerrit | Nicholas Randon proposed a change to openstack/tripleo-incubator: Allows user-specified PXE timeout for overcloud https://review.openstack.org/86605 | 14:17 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: TEST ONLY: make nova depend on common-venv https://review.openstack.org/79989 | 14:18 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Openstack-clients: don't hard code venv https://review.openstack.org/79988 | 14:18 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Wire in _EXTRA_INSTALL_OPTS... https://review.openstack.org/76966 | 14:18 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Add a new common-venv element https://review.openstack.org/76967 | 14:18 |
| openstackgerrit | Dan Prince proposed a change to openstack/tripleo-image-elements: Horizon: dynamically set config time env vars https://review.openstack.org/82611 | 14:18 |
| *** yamahata__ has joined #tripleo | 14:21 | |
| *** jpeeler has joined #tripleo | 14:21 | |
| *** jpeeler has joined #tripleo | 14:21 | |
| *** sdake has joined #tripleo | 14:21 | |
| *** funzo has quit IRC | 14:25 | |
| *** yamahata__ has quit IRC | 14:26 | |
| *** funzo has joined #tripleo | 14:26 | |
| *** yamahata has quit IRC | 14:26 | |
| derekh | slagle: the reason the CI failed for the stable branches is https://jenkins04.openstack.org/job/check-tripleo-seed-precise/438/console | 14:27 |
| derekh | slagle: 2014-04-10 14:23:00.133 | Caching xtradb-cluster tarball from http://www.percona.com/redir/downloads/TESTING/Percona-XtraDB-Cluster/5.5.33-23.7.5/release-5.5.33/476/binary/linux/x86_64/Percona-XtraDB-Cluster-5.5.33-23.7.5.476-static-openssl-1.0.1e.Linux.x86_64.tar.gz in /home/jenkins/.cache/image-create/source-repositories/xtradb_cluster_b22ec535f83b77540e4030ea83b2d07594b453e8 | 14:27 |
| derekh | 2014-04-10 14:23:01.174 | Server returned an unexpected response code. [404] | 14:27 |
| derekh | slagle: I think there was a fix for this in master recently, we probably need it in stable/icehouse aswell | 14:28 |
| *** jtomasek has quit IRC | 14:29 | |
| slagle | derekh: ok. i need to do a change in incubator as well to make pull-tools checkout the stable branches of everything | 14:31 |
| *** morazi has joined #tripleo | 14:32 | |
| slagle | or, does toci end up using pull-tools actually? | 14:32 |
| derekh | slagle: I think in ci pull-tools is a noop as ZUUL will have retrieved the correct repositories for us | 14:34 |
| derekh | slagle: but you may need to change pull-tools anyways so that people can use it locally | 14:35 |
| slagle | derekh: right. yea i see pull-tools is a noop if ZUUL_REF is defined | 14:36 |
| slagle | we probably need to zuul somehow to use the stable branches for everything. will look into that | 14:37 |
| slagle | infra says the acl updates for gerrit are run on fridays | 14:37 |
| slagle | so we won't be able to +2/+a stuff until then | 14:37 |
| derekh | slagle: it might be already pulling the stable branch of everything | 14:38 |
| derekh | slagle: ok | 14:38 |
| derekh | slagle: nope its pulling in nova master "2014-04-10 14:22:59.690 | HEAD is now at c0d76ef Merge "Remove redundant tests from Qcow2TestCase"" | 14:42 |
| derekh | slagle: this is in the master branch | 14:42 |
| *** jtomasek has joined #tripleo | 14:43 | |
| *** noslzzp has joined #tripleo | 14:44 | |
| slagle | i'm looking how the other stable jobs are setup | 14:46 |
| *** ilives has quit IRC | 14:47 | |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Adding NTP to undercloud template https://review.openstack.org/85445 | 14:47 |
| *** lazy_prince has joined #tripleo | 14:48 | |
| *** ilives has joined #tripleo | 14:48 | |
| *** marun has joined #tripleo | 14:58 | |
| *** untriaged-bot has joined #tripleo | 15:00 | |
| untriaged-bot | Untriaged bugs so far: | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1305859 | 15:00 |
| uvirtbot | Launchpad bug 1305859 in tripleo "write-pip-manifest includes non virtualenv version info" [Undecided,In progress] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1305645 | 15:00 |
| uvirtbot | Launchpad bug 1305645 in tripleo "Cannot clone novnc repository: fatal: https://github.com/kanaka/noVNC.git/info/refs not found: did you run git update-server-info on the server?" [Undecided,New] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1302881 | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1290488 | 15:00 |
| uvirtbot | Launchpad bug 1302881 in tripleo "incloud CIDR can overlap custom baremetal-network" [Undecided,In progress] | 15:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1305785 | 15:00 |
| uvirtbot | Launchpad bug 1290488 in tripleo "Baremetal: Invalid credentials" [Undecided,Incomplete] | 15:00 |
| uvirtbot | Launchpad bug 1305785 in tripleo "ubuntu cloud image: sha256sum: WARNING: 1 computed checksum did NOT match" [Undecided,New] | 15:00 |
| *** untriaged-bot has quit IRC | 15:00 | |
| *** ifarkas has quit IRC | 15:05 | |
| *** nati_ueno has joined #tripleo | 15:10 | |
| *** cody-somerville has quit IRC | 15:15 | |
| *** bauzas has quit IRC | 15:24 | |
| *** jcoufal has quit IRC | 15:25 | |
| *** nati_ueno has quit IRC | 15:26 | |
| bnemec | <3 untriaged-bot | 15:30 |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-incubator: The default in heat-templates is to preserve ephemeral disks https://review.openstack.org/85353 | 15:31 |
| *** cody-somerville has joined #tripleo | 15:31 | |
| *** jtomasek has quit IRC | 15:34 | |
| *** jb11211 has quit IRC | 15:51 | |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Activate venvs in os-*-config elements source install https://review.openstack.org/84810 | 15:51 |
| *** eghobo has joined #tripleo | 15:54 | |
| *** e0ne_ has quit IRC | 15:56 | |
| *** e0ne has joined #tripleo | 15:57 | |
| openstackgerrit | Michael Tupitsyn proposed a change to openstack/tripleo-heat-templates: Configurable Keystone token provider https://review.openstack.org/84807 | 15:58 |
| *** marun has quit IRC | 16:00 | |
| *** marun has joined #tripleo | 16:01 | |
| *** ramishra has joined #tripleo | 16:03 | |
| openstackgerrit | Chris Krelle proposed a change to openstack/tripleo-image-elements: Remove unnessary logging conf files from Ironic elements https://review.openstack.org/86639 | 16:04 |
| openstackgerrit | Coleman Corrigan proposed a change to openstack/tripleo-image-elements: Update write-pip-manifests pip freeze to use -l https://review.openstack.org/86581 | 16:06 |
| *** markmc has quit IRC | 16:06 | |
| *** jcoufal has joined #tripleo | 16:07 | |
| corrigac1 | slagle: - brainfart on my part ,though git freeze might be useful :-) .. | 16:07 |
| slagle | corrigac1: :) | 16:08 |
| slagle | corrigac1: think you got another typo in that commit message..."pirpt freeze" | 16:13 |
| openstackgerrit | Coleman Corrigan proposed a change to openstack/tripleo-image-elements: Update write-pip-manifests pip freeze to use -l https://review.openstack.org/86581 | 16:14 |
| *** dtantsur|bbl is now known as dtantsur | 16:15 | |
| *** hashar has quit IRC | 16:16 | |
| openstackgerrit | Michael Kerrin proposed a change to openstack/diskimage-builder: WIP - Add script to replay elements in an existing image https://review.openstack.org/80267 | 16:16 |
| openstackgerrit | Michael Kerrin proposed a change to openstack/diskimage-builder: Add utility to mount and unmount images https://review.openstack.org/76509 | 16:16 |
| openstackgerrit | Michael Kerrin proposed a change to openstack/diskimage-builder: WIP - enable elements to be re-playable https://review.openstack.org/81494 | 16:16 |
| *** bauzas has joined #tripleo | 16:17 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: set -e all the things https://review.openstack.org/83927 | 16:20 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Make sure all scripts are set -e https://review.openstack.org/81637 | 16:20 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Fix set -eu and pipefail failures https://review.openstack.org/85861 | 16:20 |
| howleyt | jprovazn: would you mind looking at that rabbitmq fix again please? https://review.openstack.org/#/c/86034/ | 16:20 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Fix dib-lint to work with set -e https://review.openstack.org/83930 | 16:20 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Check for set -o pipefail https://review.openstack.org/83929 | 16:20 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: set -u and -o pipefail everywhere https://review.openstack.org/84868 | 16:20 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Ensure scripts are set -u https://review.openstack.org/83928 | 16:20 |
| *** martyntaylor has left #tripleo | 16:23 | |
| bnemec | rbrady: Can you double-check me on the latest rebase of https://review.openstack.org/#/c/78462/ ? | 16:23 |
| bnemec | Just want to make sure I didn't mess up your cinder tgt path fix. | 16:23 |
| *** e0ne has quit IRC | 16:26 | |
| *** e0ne has joined #tripleo | 16:27 | |
| NobodyCam | good morning TripleO, just checking to see if anyone has a free minute to take a quick look at: https://review.openstack.org/86639 (no pressure, hehehe) :) | 16:30 |
| *** e0ne has quit IRC | 16:31 | |
| rbrady | bnemec: checking | 16:35 |
| *** yamahata has joined #tripleo | 16:35 | |
| *** lazy_prince has quit IRC | 16:38 | |
| *** corrigac1 has left #tripleo | 16:42 | |
| SpamapS | are the f20 jobs considered voting? | 16:45 |
| dprince | SpamapS: I've seen them passing. Is there a failure you are concerned with? | 16:46 |
| SpamapS | dprince: https://review.openstack.org/85773 | 16:46 |
| SpamapS | hm wait.. weird | 16:47 |
| SpamapS | thought I saw that fail in an email notification | 16:47 |
| dprince | SpamapS: yeah, it is queued and derekh already pushed an increase to the F20 nodepool size so I'd expect things to speed up once that lands | 16:48 |
| *** andreaf has quit IRC | 16:48 | |
| dprince | SpamapS: I sort of wish we had a federated nodepool setup so we could adjust our max/min sizes much more quickly | 16:49 |
| rbrady | bnemec: lgtm | 16:49 |
| bnemec | rbrady: Thanks! | 16:50 |
| *** akuznetsov has quit IRC | 16:51 | |
| derekh | SpamapS: bnemec also has a patch to make it non voting until we are happy with its stability, things are passing now the instability we saw a few days ago was because of a problem on a compute node https://bugs.launchpad.net/tripleo/+bug/1304085 | 16:51 |
| uvirtbot | Launchpad bug 1304085 in tripleo "check-tripleo-overcloud-f20 fails to get IP on test network" [Critical,In progress] | 16:51 |
| derekh | SpamapS: they look pretty stable today and once the nodepool min-ready change goes in and its happy for a day or 2 we can consider it voting | 16:52 |
| derekh | SpamapS: thats my opinion anyways :-) | 16:52 |
| openstackgerrit | Michael Kerrin proposed a change to openstack/tripleo-image-elements: Start to remoe packages in order to make the image sizes smaller. https://review.openstack.org/86650 | 16:53 |
| SpamapS | derekh: ok cool | 16:54 |
| *** epim has joined #tripleo | 16:56 | |
| *** derekh has quit IRC | 16:57 | |
| *** akuznetsov has joined #tripleo | 16:57 | |
| openstackgerrit | Jordan OMara proposed a change to openstack/tuskar-ui: Fixing django conf url import, such as Bug: #1251248 https://review.openstack.org/86651 | 17:00 |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: WIP: Add unit test for cache-url https://review.openstack.org/84493 | 17:01 |
| *** akuznetsov has quit IRC | 17:07 | |
| *** lucasagomes has quit IRC | 17:09 | |
| *** nati_ueno has joined #tripleo | 17:17 | |
| *** petertoft has quit IRC | 17:19 | |
| *** jistr has quit IRC | 17:20 | |
| *** jistr has joined #tripleo | 17:21 | |
| *** giulivo has quit IRC | 17:22 | |
| *** jistr has quit IRC | 17:28 | |
| *** nati_ueno has quit IRC | 17:35 | |
| *** nati_ueno has joined #tripleo | 17:36 | |
| *** akrivoka has quit IRC | 17:38 | |
| *** jb11211 has joined #tripleo | 17:42 | |
| *** sdake has quit IRC | 17:42 | |
| openstackgerrit | john Brooker proposed a change to openstack/tripleo-incubator: Allow baremetal-networks to be passed to devtest https://review.openstack.org/85446 | 17:44 |
| jprovazn | howleyt, done, thanks for the patch | 17:53 |
| *** jcoufal has quit IRC | 17:53 | |
| *** UtahDave has joined #tripleo | 17:54 | |
| *** UtahDave has left #tripleo | 17:55 | |
| *** morganfainberg_Z is now known as morganfainberg | 17:56 | |
| *** akuznetsov has joined #tripleo | 17:57 | |
| *** jdonalds has joined #tripleo | 18:06 | |
| *** dprince has quit IRC | 18:08 | |
| *** CaptTofu has quit IRC | 18:08 | |
| *** CaptTofu has joined #tripleo | 18:08 | |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-image-elements: Document the required mysql config values https://review.openstack.org/85874 | 18:12 |
| *** cwolferh_ has joined #tripleo | 18:13 | |
| *** hashar has joined #tripleo | 18:18 | |
| *** akrivoka has joined #tripleo | 18:18 | |
| *** jdonalds has quit IRC | 18:21 | |
| vinsh | jprovazn, Ping... | 18:27 |
| jprovazn | vinsh, pong | 18:27 |
| vinsh | jprovazn, Hi, i'm Adam V... I work in fort collins Co. Good to meet you. I wanted to learn more about this work: https://gist.github.com/jprovaznik/9662754 | 18:28 |
| *** jang1 has joined #tripleo | 18:28 | |
| *** jtomasek has joined #tripleo | 18:28 | |
| *** sdake_ is now known as sdake | 18:28 | |
| *** petertoft has joined #tripleo | 18:28 | |
| vinsh | I have this applied to an out of the box devtest run. and generally it does what I expect. but I am unclear about a few things.. such as what is actually applied to haproxy.cfg on the control node. | 18:28 |
| jprovazn | vinsh, hi - that is a temporary template I used so far fo testing multinode (2-node) notcpmpute overcloud setup | 18:29 |
| vinsh | Cool.. ya.. i'm over at HP hacking away on this also | 18:29 |
| vinsh | trying to expand this same template/concept to cover a handful of nodes | 18:29 |
| jprovazn | vinsh, oh, don't expand it - it's really not nice, creating properly autoscaling notcompute template is now in process | 18:30 |
| vinsh | this template.. will yield something like this in the haproxy.cfg "listen swift_proxy_cluster | 18:31 |
| *** jdonalds has joined #tripleo | 18:31 | |
| vinsh | bind *:8090 " | 18:31 |
| vinsh | but.. what IS "swift_proxy_cluster"... | 18:31 |
| vinsh | how does the node resolve that? | 18:31 |
| vinsh | yes.. autoscaling +1 | 18:31 |
| jprovazn | ok, sec, let me find a link | 18:32 |
| *** akuznetsov has quit IRC | 18:33 | |
| jprovazn | vinsh, haproxy's template config is applied when configuring haproxy after boot (or after any metadata change) here: https://github.com/openstack/tripleo-image-elements/blob/master/elements/haproxy/os-config-applier/etc/haproxy/haproxy.cfg | 18:33 |
| vinsh | ya. so I have that running... | 18:34 |
| jprovazn | IOW swift_proxy_cluster is just haproxy.name which is just "label" of section in config file | 18:34 |
| vinsh | this is what was written on disk to the notCompute node: | 18:34 |
| vinsh | listen swift_proxy_cluster | 18:34 |
| vinsh | bind *:8090 | 18:34 |
| vinsh | mode tcp | 18:34 |
| vinsh | balance source | 18:34 |
| vinsh | option tcpka | 18:34 |
| vinsh | option tcplog | 18:34 |
| vinsh | server notCompute0 192.0.2.22:8080 check inter 2000 rise 2 fall | 18:34 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: The MySQL element requires the os-svc-install element https://review.openstack.org/83967 | 18:34 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-heat-templates: Add initial support for galera clustering https://review.openstack.org/83883 | 18:34 |
| vinsh | id have expected heat to have turned that into an IP or something? | 18:34 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-heat-templates: Add initial support for galera clustering https://review.openstack.org/83883 | 18:35 |
| vinsh | jprovazn, so in the template we have this: | 18:36 |
| vinsh | - name: swift_proxy_cluster | 18:36 |
| vinsh | port: 8080 | 18:36 |
| vinsh | proxy_port: 8090 | 18:36 |
| vinsh | and.. its written literally as is.. vs an ip for "swift_proxy_cluster" | 18:36 |
| jprovazn | vinsh, well, in the block you pasted there is an ip: 192.0.2.22 (ip of the node where request is proxied to) | 18:37 |
| vinsh | yeah, but listen is on "swift_proxy_cluster". I was wondering .. how it can listen to that string? | 18:38 |
| vinsh | vs .. a shared IP or something | 18:38 |
| jprovazn | vinsh, it listens on 8090 (row bind *:8090) | 18:38 |
| jprovazn | vinsh, consider "listen swift_proxy_cluster" just as a label - compare with this: http://docs.openstack.org/high-availability-guide/content/ha-aa-haproxy.html | 18:39 |
| vinsh | ah-hah. Thank you. | 18:39 |
| jprovazn | vinsh, np | 18:39 |
| vinsh | so.. what we might want here .. is to set "*" to a shared IP right? | 18:39 |
| vinsh | to allow it to scale across multiple control nodes running haproxy. each using a shared IP? | 18:40 |
| vinsh | vs.. just listening to itself. | 18:40 |
| *** marun has quit IRC | 18:41 | |
| jprovazn | vinsh, I'm not sure I understand - IIRC " bind *:8090" means that it binds to all configured interfaces | 18:42 |
| jprovazn | one of interfaces is configured for VIP | 18:42 |
| vinsh | OK. so that implies that one interface is already bound to that shared IP | 18:42 |
| jprovazn | keepalive element takes care of managing "active" node for VIP | 18:42 |
| vinsh | Do you have keepalive configured for this yet? | 18:43 |
| jprovazn | vinsh, yes | 18:43 |
| jprovazn | vinsh, yep - it worked for me - did you try keepalive element? | 18:43 |
| jang1 | if something is listening on *:9999, then traffic to any interface, port 9999 will reach it. If an interface is added later (such as an HAProxy vip), then traffic coming in on that VIP will also reach the listener. | 18:43 |
| vinsh | I did.. but I didn't have the heat cofnig to configure keepalive | 18:44 |
| vinsh | it just cycled trying to find some config it lacked | 18:44 |
| vinsh | jang1, ACK | 18:44 |
| *** jdonalds has quit IRC | 18:45 | |
| vinsh | basically.. notcompute0 couldn't configure the keepalived element because I probably didn't add the right config for it to overcloud-source | 18:45 |
| jprovazn | vinsh, try it with required heat config - it should work then, or am I missing some bit? | 18:45 |
| jprovazn | ah, ok | 18:45 |
| vinsh | no. you got it.. i'm just new and drinking from the fire hose still :) | 18:45 |
| jprovazn | vinsh, https://gist.github.com/jprovaznik/9662754#file-gistfile1-txt-L750 | 18:45 |
| vinsh | I appreciate your help big time! | 18:46 |
| jprovazn | and alternatively elements/keepalive/README | 18:46 |
| vinsh | cool. will try that. one last thing.. this part " ControllerVirtualIp: | 18:46 |
| vinsh | Default: 192.0.2.254 | 18:46 |
| vinsh | Description: Virtual IP used by haproxy | 18:46 |
| vinsh | NoEcho: true | 18:46 |
| vinsh | Type: String" | 18:46 |
| jprovazn | vinsh, np at all | 18:46 |
| *** jang1 has quit IRC | 18:46 | |
| vinsh | I don't see this ip configured on the notcompute0 node.. what might be missing to enable that? | 18:46 |
| vinsh | as hopefully.. this would then apply to the scaled out set of notcompute nodes.. like you were suggesting | 18:47 |
| *** jtomasek has quit IRC | 18:47 | |
| jprovazn | vinsh, keepalive element should take care of it IIRC | 18:47 |
| vinsh | Got it. This is plenty for me to chew on.. if you want to discuss the scaled out version of this.. or need any help in testing please loop me in! | 18:48 |
| jprovazn | vinsh, cool, help would be definetly needed - there is a trelo card for this: https://trello.com/c/6NuTjWjl/77-run-3-control-plane-nodes and SpamapS sent an initial patch which updates heat template to use SW config - then autosclaing would be done on top of it | 18:51 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Update write-pip-manifests pip freeze to use -l https://review.openstack.org/86581 | 18:51 |
| *** jdonalds has joined #tripleo | 18:51 | |
| jprovazn | vinsh, https://review.openstack.org/#/c/81666/ | 18:52 |
| vinsh | jprovazn, sweet.. I can dial in there and follow a plan vs random crazy ideas I come up with :) | 18:53 |
| jprovazn | vinsh, awesome | 18:53 |
| *** jprovazn is now known as jprovazn_bbl | 18:53 | |
| *** petertoft has quit IRC | 18:54 | |
| *** jcoufal has joined #tripleo | 18:59 | |
| *** yamahata has quit IRC | 19:04 | |
| *** sdake_ has joined #tripleo | 19:04 | |
| *** akuznetsov has joined #tripleo | 19:05 | |
| *** petertoft has joined #tripleo | 19:07 | |
| *** akuznetsov has quit IRC | 19:09 | |
| *** nati_uen_ has joined #tripleo | 19:13 | |
| *** akrivoka has quit IRC | 19:14 | |
| *** nati_ueno has quit IRC | 19:14 | |
| greghaynes | tchaypo: your passwordrc patch merged! | 19:16 |
| *** jprovazn_bbl has quit IRC | 19:17 | |
| *** ilives has quit IRC | 19:24 | |
| *** ramishra has quit IRC | 19:33 | |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/tripleo-incubator: Doc updates suggested in another review https://review.openstack.org/86594 | 19:33 |
| jp_at_hp | tchaypo: greghaynes - that's the miracle for today :D | 19:34 |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Use provides to note an element provides an OS https://review.openstack.org/85773 | 19:35 |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/tripleo-image-elements: Define the contents of images as elements https://review.openstack.org/85776 | 19:36 |
| *** nati_uen_ has quit IRC | 19:36 | |
| *** nati_ueno has joined #tripleo | 19:37 | |
| *** petertoft has quit IRC | 19:40 | |
| openstackgerrit | Jon-Paul Sullivan (jp_at_hp) proposed a change to openstack/diskimage-builder: Use 'element-provides' to denote deploy capability https://review.openstack.org/86333 | 19:41 |
| *** cody-somerville has quit IRC | 19:42 | |
| *** petertoft has joined #tripleo | 19:53 | |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/diskimage-builder: check element list during the start up https://review.openstack.org/85184 | 19:56 |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/diskimage-builder: fails if there is no “operating-system” element https://review.openstack.org/85272 | 19:56 |
| openstackgerrit | Gonéri Le Bouder proposed a change to openstack/diskimage-builder: search for duplicated ”operating-system” elements https://review.openstack.org/85271 | 19:56 |
| *** dividebin has quit IRC | 19:56 | |
| *** jdonalds has quit IRC | 20:01 | |
| *** dividehex has joined #tripleo | 20:03 | |
| *** jp_at_hp1 has joined #tripleo | 20:03 | |
| *** jb11211 has quit IRC | 20:04 | |
| *** jcoufal has quit IRC | 20:04 | |
| *** jp_at_hp has quit IRC | 20:06 | |
| greghaynes | jp_at_hp1: yes, quite the miracle | 20:07 |
| jp_at_hp1 | Reassuring for the rest of us to remember that there is always an end to these things :D | 20:08 |
| greghaynes | when I had a patch like that after I just started I asked clarkb to run a query for the patch with the most review comments | 20:09 |
| greghaynes | still waiting for an answer | 20:09 |
| Goneri | now jp_at_hp1 patch is accpeted (Yearh!) may I ask our beloved cores to take a look on my merge request based on his patch. https://review.openstack.org/#/q/status:open+project:openstack/diskimage-builder+branch:master+topic:virtual-image-elements,n,z | 20:14 |
| Goneri | :) | 20:14 |
| *** e0ne has joined #tripleo | 20:15 | |
| *** NobodyCam is now known as NotNobodyCam | 20:15 | |
| *** NotNobodyCam is now known as NobodyCam | 20:16 | |
| *** jb11211 has joined #tripleo | 20:16 | |
| Lotus907efi | greghaynes: what file was it that pulls down the percona xtradb thing? | 20:18 |
| *** epim has quit IRC | 20:18 | |
| Lotus907efi | we have a user who says he pulled the latest code this morning and he is getting the 404 error on the old percona xtradb file | 20:18 |
| *** lblanchard has quit IRC | 20:19 | |
| Lotus907efi | but I cannot figure out from looking at my logs what script pulls that | 20:19 |
| bnemec | Lotus907efi: https://github.com/openstack/diskimage-builder/blob/master/elements/source-repositories/extra-data.d/98-source-repositories | 20:24 |
| bnemec | Using the settings in https://github.com/openstack/tripleo-image-elements/blob/master/elements/mysql/source-repository-mysql | 20:24 |
| Lotus907efi | ok, cool | 20:25 |
| bnemec | The latter is probably the one that's outdated. | 20:25 |
| Lotus907efi | thanks | 20:25 |
| bnemec | np | 20:25 |
| *** epim has joined #tripleo | 20:27 | |
| *** saju_m has joined #tripleo | 20:29 | |
| *** dprince has joined #tripleo | 20:29 | |
| rwsu | SpamapS: hi, i would like your opinion on whether this change is ok: https://review.openstack.org/#/c/85534/ | 20:30 |
| SpamapS | rwsu: hrm | 20:31 |
| SpamapS | rwsu: I don't really think that is awesome. :-P | 20:31 |
| SpamapS | rwsu: I don't understand why you'd disallow local socket connections really. | 20:32 |
| *** e0ne has quit IRC | 20:32 | |
| *** jdonalds has joined #tripleo | 20:32 | |
| *** e0ne has joined #tripleo | 20:32 | |
| *** ramishra has joined #tripleo | 20:34 | |
| rwsu | SpamapS: haven't been able to work around an selinux issue with clients connecting to mysql using the local socket, I can look at at that some more | 20:35 |
| *** e0ne has quit IRC | 20:36 | |
| rwsu | SpamapS: to help me understand the issues better, do we use localhost because of performance/security? what's the reason we have for not using the actual ip address? | 20:37 |
| *** ramishra has quit IRC | 20:38 | |
| SpamapS | rwsu: because that is the usual pattern. | 20:41 |
| SpamapS | rwsu: but I'd rather see localhost replaced with 127.0.0.1, not the massive IP plumbing | 20:42 |
| rwsu | SpamapS: ok i'll see if 127.0.0.1 works, thanks for your thoughts | 20:43 |
| *** e0ne has joined #tripleo | 20:43 | |
| SpamapS | rwsu: I just don't know why selinux would deny the user access to the socket. | 20:48 |
| saju_m | hi guys | 20:50 |
| bnemec | All of a sudden mysql on my seed is not happy: http://fpaste.org/93325/13971621/ | 20:51 |
| bnemec | It looks like I'm missing .my.cnf in /root | 20:51 |
| bnemec | Anyone else seen anything like this? | 20:52 |
| SpamapS | bnemec: ^^ looks like rwsu is looking at your problem already | 20:52 |
| saju_m | I have couple of servers with 40 processors, 400 GB ram and 1pb disk | 20:53 |
| *** killer_prince has quit IRC | 20:53 | |
| bnemec | SpamapS: I'm not running with selinux though | 20:53 |
| saju_m | I would like to deploy tripleo there for setup openstack development environment | 20:53 |
| bnemec | Or did I miss another discussion? | 20:53 |
| saju_m | spamaos:lifeless: i asked same question to you, so please ignore this | 20:54 |
| saju_m | I went through following online materials http://dpaste.com/1775900/ | 20:54 |
| rwsu | bnemec: I don't think it is related to the heat template discussion, that hasn't been merged to master | 20:54 |
| saju_m | But, i could not find a starting point to install tripleo | 20:54 |
| saju_m | I am looking for a tutorial which explains how to setup seed, undercloud and overcloud in 2 to 3 or more servers | 20:54 |
| bnemec | saju_m: Have you looked at http://docs.openstack.org/developer/tripleo-incubator/ ? | 20:55 |
| slagle | saju_m: this is the starting point: http://docs.openstack.org/developer/tripleo-incubator/devtest.html | 20:55 |
| slagle | yea, what he said :) | 20:55 |
| saju_m | <slagle>: can i install seed, undercloud and overcloud in one server machine? | 20:58 |
| geerdest | saju_m: yes one machine | 20:59 |
| *** untriaged-bot has joined #tripleo | 21:00 | |
| untriaged-bot | Untriaged bugs so far: | 21:00 |
| untriaged-bot | https://bugs.launchpad.net/tripleo/+bug/1302881 | 21:00 |
| uvirtbot | Launchpad bug 1302881 in tripleo "incloud CIDR can overlap custom baremetal-network" [Undecided,Incomplete] | 21:00 |
| *** untriaged-bot has quit IRC | 21:00 | |
| saju_m | this document is complicated http://docs.openstack.org/developer/tripleo-incubator/devtest.html | 21:00 |
| saju_m | it should give a simple introduction in the beginning | 21:01 |
| *** matty_dubs is now known as matty_dubs|gone | 21:01 | |
| saju_m | for example, In overview: | 21:02 |
| saju_m | it saying, Define a VM that is your seed node | 21:02 |
| saju_m | where should i Define a VM that is your seed node ? | 21:02 |
| greghaynes | mmm, very rebase, much conflict with tripleo-heat-templates | 21:02 |
| bnemec | saju_m: Did you continue into the devtest_seed instructions? They explain how to set up a seed node. | 21:04 |
| saju_m | Please assume that, I have a plane Server with ubuntu 12.04 installed in it. What is the next and first step i have to do ? | 21:04 |
| *** petertoft has quit IRC | 21:06 | |
| saju_m | bnemec: let me check | 21:07 |
| saju_m | bnemec: i am there http://docs.openstack.org/developer/tripleo-incubator/devtest_seed.html | 21:07 |
| *** jdonalds has quit IRC | 21:08 | |
| mordred | rwsu: are you sure it's selinux and not just default perms on the mysql database? | 21:09 |
| mordred | there is a difference between allow over 127.0.0.1 and allow over socket in the mysql permissions table, and if there is an anonymous user entry in the db, it can screw you on trying to make socket conenctions in ways you may not expect | 21:10 |
| *** e0ne has quit IRC | 21:10 | |
| *** e0ne has joined #tripleo | 21:11 | |
| *** saurabhs has joined #tripleo | 21:11 | |
| *** e0ne has quit IRC | 21:15 | |
| saurabhs | Hye guys trying to see why tests are failing on my reviews. http://logs.openstack.org/05/82605/3/check-tripleo/check-tripleo-undercloud-precise/8f72d89/devstack-gate-setup-workspace-new.txt.gz | 21:16 |
| saurabhs | http://logs.openstack.org/05/82605/3/check-tripleo/check-tripleo-undercloud-precise/8f72d89/devstack-gate-setup-workspace-new.txt.gz#_2014-04-09_09_37_03_056 | + git fetch http://zm01.openstack.org/p/openstack-dev/devstack refs/zuul/master/Z18bac1e322b14a16977d3c68b13a6756 | 21:16 |
| saurabhs | http://logs.openstack.org/05/82605/3/check-tripleo/check-tripleo-undercloud-precise/8f72d89/devstack-gate-setup-workspace-new.txt.gz#_2014-04-09_09_37_03_527 | fatal: Couldn't find remote ref refs/zuul/master/Z18bac1e322b14a16977d3c68b13a6756 | 21:16 |
| saurabhs | http://logs.openstack.org/05/82605/3/check-tripleo/check-tripleo-undercloud-precise/8f72d89/devstack-gate-setup-workspace-new.txt.gz#_2014-04-09_09_37_03_534 | Unexpected end of command stream | 21:16 |
| saurabhs | + git fetch http://zm01.openstack.org/p/openstack-dev/devstack refs/zuul/master/Z18bac1e322b14a16977d3c68b13a6756 | 21:16 |
| saurabhs | fatal: Couldn't find remote ref refs/zuul/master/Z18bac1e322b14a16977d3c68b13a6756 | 21:17 |
| saurabhs | Unexpected end of command stream | 21:17 |
| greghaynes | clarkb: ^ | 21:17 |
| greghaynes | looks infraey | 21:17 |
| saurabhs | this is the review: https://review.openstack.org/#/c/82605/ | 21:18 |
| saju_m | i cloned stable/icehouse branch of tripleo-incubator | 21:19 |
| *** CaptTofu has quit IRC | 21:20 | |
| saju_m | i can see the script devtest.sh in tripleo-incubator/scripts/, Is that what i have to run first ? | 21:20 |
| saju_m | If i run, tripleo-incubator/scripts/devtest.sh, that will install seed, under nad overcloud in one machine, right ? | 21:21 |
| *** rpodolyaka has quit IRC | 21:21 | |
| saju_m | that is what i understand. Please correct me , if i am wrong. | 21:22 |
| *** dprince has quit IRC | 21:24 | |
| bnemec | saju_m: Yes, although you have to pass --trash-my-machine to devtest.sh to let it actually run. | 21:25 |
| saju_m | bnemec: like, ./scripts/devtest.sh --trash-my-machin , right ? | 21:26 |
| *** rpodolyaka has joined #tripleo | 21:26 | |
| bnemec | saju_m: Yes | 21:28 |
| *** morazi is now known as morazi-biab | 21:28 | |
| rwsu | mordred: good to know, will check | 21:30 |
| *** ramishra has joined #tripleo | 21:34 | |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-heat-templates: Add initial support for galera clustering https://review.openstack.org/83883 | 21:39 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-heat-templates: WIP: Switch overcloud to software-config https://review.openstack.org/81666 | 21:39 |
| *** ramishra has quit IRC | 21:39 | |
| greghaynes | ruh roh | 21:40 |
| greghaynes | think I went a little overboard with the rebasing | 21:40 |
| clarkb | greghaynes: saurabhs there are networking issues between hp land and rackspace | 21:41 |
| *** hashar has quit IRC | 21:44 | |
| NobodyCam | clarkb: did the string between the two cans come losea | 21:44 |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-heat-templates: WIP: Switch overcloud to software-config https://review.openstack.org/81666 | 21:44 |
| clarkb | noslzzp: no idea | 21:45 |
| clarkb | NobodyCam: whhoops ^ | 21:45 |
| greghaynes | ok, we can just act like that rebase derp never happened ;) | 21:45 |
| NobodyCam | hehe | 21:47 |
| tchaypo | saju_m: http://docs.openstack.org/developer/tripleo-incubator/devtest.html should help guide you through the process | 21:51 |
| tchaypo | saju_m: Under the heading "stability warning" there's a series of commands that will lead you to building the version of that doc that corresponds with the exect checkout you have on the stable branch, so once you hit that point you'll want to switch to following your locally-generated copy | 21:52 |
| saju_m | <tchaypo>: yes, thanks, I am trying that | 21:52 |
| saju_m | <tchaypo>: let you know, if a get any issue | 21:53 |
| *** ramishra has joined #tripleo | 22:03 | |
| *** nijaba has quit IRC | 22:03 | |
| *** ramishra has quit IRC | 22:07 | |
| *** jp_at_hp1 has quit IRC | 22:10 | |
| *** yamahata_ has quit IRC | 22:11 | |
| *** yamahata_ has joined #tripleo | 22:11 | |
| *** saju_m has quit IRC | 22:20 | |
| *** noslzzp has quit IRC | 22:23 | |
| vinsh | greghaynes, giving the ha tripleO stuff a whirl still.. have you seen this out of it? 'ERROR: The Parameter (NeutronBridgeMappings) was not defined in template.' | 22:41 |
| vinsh | this was using the over-cloud source from Jan | 22:42 |
| greghaynes | Nope, but sounds pretty straightforward - that parameter isnt defined in the template and youre passing it in ;) | 22:42 |
| greghaynes | id diff that template against whats in master for triple-heat-templates | 22:42 |
| vinsh | cool. the only places I currently see that paramater are here | 22:43 |
| vinsh | ./nova-compute-instance.yaml: NeutronBridgeMappings: | 22:43 |
| vinsh | ./nova-compute-instance.yaml: bridge_mappings: {Ref: NeutronBridgeMappings} | 22:43 |
| greghaynes | looks like there should be a paremeter in overcloud-source for that too | 22:44 |
| vinsh | bingo :) | 22:47 |
| *** CaptTofu has joined #tripleo | 22:58 | |
| *** jpeeler has quit IRC | 23:00 | |
| stewart | SpamapS: huh... okay, iscsi interesting | 23:01 |
| stewart | lifeless, SpamapS: was thinking, might be god to run PMP on mysqld just to see where things are stalling for all that time. help track down if it is in fact disk latency or it's actually doing work | 23:02 |
| *** yamahata has joined #tripleo | 23:03 | |
| *** ramishra has joined #tripleo | 23:03 | |
| *** ramishra has quit IRC | 23:08 | |
| *** mikal is now known as stillbert | 23:19 | |
| *** tchaypo has quit IRC | 23:20 | |
| *** tchaypo has joined #tripleo | 23:21 | |
| *** eguz has joined #tripleo | 23:21 | |
| *** eguz has quit IRC | 23:21 | |
| *** CaptTofu has quit IRC | 23:23 | |
| *** tchaypo has quit IRC | 23:24 | |
| *** eghobo has quit IRC | 23:25 | |
| *** tchaypo has joined #tripleo | 23:25 | |
| *** tchaypo has quit IRC | 23:25 | |
| *** tchaypo has joined #tripleo | 23:26 | |
| openstackgerrit | Gregory Haynes proposed a change to openstack/tripleo-heat-templates: Add initial support for galera clustering https://review.openstack.org/83883 | 23:26 |
| *** xuhaiwei has joined #tripleo | 23:32 | |
| xuhaiwei | morning | 23:33 |
| NobodyCam | morning xuhaiwei :) | 23:36 |
| xuhaiwei | :) | 23:36 |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-incubator: Add some clarity to the first-time user experience https://review.openstack.org/83294 | 23:38 |
| tchaypo | morning xuhaiwei | 23:38 |
| xuhaiwei | morning, tchaypo | 23:39 |
| xuhaiwei | are you guys in UTC+8? | 23:39 |
| tchaypo | Australian Eastern | 23:39 |
| tchaypo | I think we're back to +10 now | 23:40 |
| xuhaiwei | oh | 23:40 |
| tchaypo | well, I am anyway, no idea about NobodyCam | 23:40 |
| tchaypo | greghaynes is physically in US West but seems to work Aus East | 23:40 |
| xuhaiwei | it seems many people work in Aus here | 23:41 |
| greghaynes | Its funny - ive always ended up working on open source projects with people in AUS East, didnt realize until last year it likely want just coincidence | 23:41 |
| StevenK | greghaynes is on GST -- Greg Standard Time | 23:42 |
| greghaynes | actually, isnt there a google standard time | 23:43 |
| greghaynes | but yes, GST | 23:44 |
| lifeless | stewart: oh hai | 23:45 |
| stewart | lifeless: goood morning | 23:46 |
| lifeless | stewart: yeah eatmydata 'fixed' it | 23:46 |
| lifeless | stewart: I think there's likely still huge fat and contention there but .... | 23:46 |
| stewart | lifeless: eatmydata ftw :) | 23:46 |
| stewart | lifeless: probably not production ready though :) | 23:46 |
| lifeless | hmm | 23:47 |
| lifeless | I may need an O_DIRECT LD_PRELOADer | 23:48 |
| stewart | lifeless: you can make mysqld do O_DIRECT | 23:48 |
| lifeless | no, qemu-img | 23:48 |
| stewart | lifeless: innodb_flush_mode IIRC | 23:48 |
| lifeless | stewart: so to deploy 41 machines | 23:49 |
| stewart | lifeless: ahh.. believe it or not, been discussing how qemu flushes things recently | 23:49 |
| lifeless | stewart: we have 41 1.8G images. | 23:49 |
| stewart | lifeless: i wonder if tuning io scheduler to have mysql have real high prio would help at all... there are things | 23:49 |
| *** yamahata has quit IRC | 23:50 | |
| lifeless | stewart: right now we pull them from glance, use qemu-img convert -f raw to make a raw file then dd it | 23:51 |
| lifeless | stewart: I'm thinking we can qemu-img it directly over iscsi | 23:51 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!