Thursday, 2015-04-09

xgermanthe way it logs is that it logs to rsyslogd00:00
ptoohillwell yea00:00
xgermanso to get to a file we need to change the rsyslogd configuration00:00
xgermanand johnsom just told me that my plan of making it local<listener-id> diesn't work since we only have local0-700:01
xgermanbut I think being able to configure that is probably good00:01
xgermanso I am not sure if the requirements ever asked for breaking out logs by listener00:03
ptoohillI dont remember seeing such requirement. But, maybe thats a thing we want? Otherwise i assume well want a parser of sorts to easily identify problems00:05
ptoohillwere you originally concerned with what the local* was or that they were going to one file00:05
xgermanmore about going to one file00:07
xgermanI though I could use a different identifier to break it up00:07
ptoohillto break up per listener?00:07
*** Aish has quit IRC00:07
xgermanwell, the plan asks for an haproxy process per listener -- so I just like to not comingle logs from different haproxies00:08
ptoohillso i need to learn more about the logging also :/00:09
xgermanbut at that point I am pretty confused anyway and I was only looking at it from the angle that I like to remove log files when a listener gets deleted00:09
xgermanand maybe we don't need to do that ;-)00:09
ptoohilli see00:09
xgermanok, will sleep over it ;-)00:10
ptoohilli got some reading to do ;)00:10
ptoohillwell have a config for each listener right?00:11
*** vivek-ebay has joined #openstack-lbaas00:12
ptoohillif so, and im understanding it right (which im probably not) we can just pass in the path/file we want for each listener for the log_http log_server. This of course assumes syslog is set up00:13
*** xgerman has quit IRC00:15
*** chlong has joined #openstack-lbaas00:25
ptoohillI suppose this will work, but we will need to add a way for the driver to handle rsyslog updates? I assume well discuss more tomorrow ;)00:27
*** bharath has quit IRC00:28
*** bharath has joined #openstack-lbaas00:34
*** sbfox has quit IRC00:34
*** bharath has quit IRC00:38
*** vivek-ebay has quit IRC00:44
*** vivek-ebay has joined #openstack-lbaas00:45
*** amotoki has joined #openstack-lbaas00:49
openstackgerritMadhusudhan Kandadai proposed openstack/neutron-lbaas: Introduced admin_api tempest tests
*** crc32 has quit IRC01:14
*** madhu_ak has quit IRC01:26
*** madhu_ak has joined #openstack-lbaas01:28
*** madhu_ak has quit IRC01:28
*** SumitNaiksatam has joined #openstack-lbaas01:31
*** Varun_Lodaya has quit IRC01:33
*** Varun_Lodaya has joined #openstack-lbaas01:35
*** Tiancheng has joined #openstack-lbaas01:44
*** Varun_Lodaya has quit IRC01:52
*** vivek-ebay has quit IRC02:02
*** vivek-ebay has joined #openstack-lbaas02:03
*** vivek-ebay has quit IRC02:03
*** sbalukoff has quit IRC02:11
*** Varun_Lodaya has joined #openstack-lbaas02:15
*** Aish has joined #openstack-lbaas02:30
*** fnaval has quit IRC02:34
*** Aish has left #openstack-lbaas02:39
*** vivek-ebay has joined #openstack-lbaas02:56
*** mwang2 has quit IRC03:06
*** rm_you| has quit IRC03:11
*** raginbajin has quit IRC04:32
*** Tiancheng has quit IRC04:34
*** raginbajin has joined #openstack-lbaas04:35
*** vivek-ebay has quit IRC04:41
*** madhu_ak has joined #openstack-lbaas04:46
*** chlong has quit IRC05:09
*** Aish has joined #openstack-lbaas05:17
*** Aish has quit IRC05:17
*** kiran has joined #openstack-lbaas05:38
openstackgerritTrevor Vardeman proposed stackforge/octavia: Amphora SSH Driver
*** Varun_Lodaya has quit IRC05:49
*** Varun_Lodaya has joined #openstack-lbaas05:50
*** Varun_Lodaya has quit IRC05:50
*** madhu_ak has quit IRC05:59
*** kiran is now known as kiran-r06:06
*** sbfox has joined #openstack-lbaas06:26
*** sbfox has quit IRC06:35
*** sbalukoff has joined #openstack-lbaas06:46
*** vivek-ebay has joined #openstack-lbaas07:42
*** vivek-ebay has quit IRC07:46
*** woodster_ has quit IRC08:10
*** apuimedo has joined #openstack-lbaas08:17
*** Tiancheng has joined #openstack-lbaas08:18
*** mwang2 has joined #openstack-lbaas09:03
*** mwang2 has quit IRC09:07
*** pcaruana has joined #openstack-lbaas09:17
*** haigang has joined #openstack-lbaas09:39
*** f13o has quit IRC09:54
*** Tiancheng has quit IRC10:22
*** amotoki has quit IRC11:19
*** haigang has quit IRC11:24
*** woodster_ has joined #openstack-lbaas11:49
*** ganeshna1 has joined #openstack-lbaas11:59
*** ganeshna has joined #openstack-lbaas12:05
*** ganeshna1 has quit IRC12:14
*** ganeshna has quit IRC12:43
*** rohara has quit IRC13:23
*** kiran-r has quit IRC13:23
*** kiran_ has joined #openstack-lbaas13:23
*** kiran_ has quit IRC13:29
*** ganeshna has joined #openstack-lbaas13:30
mesteryblogan: I re-proposed the pecan switch spec ( and added you as well as kevinbenton as primarry assignees, FYI.13:34
*** ganeshna has quit IRC13:46
*** rohara has joined #openstack-lbaas13:48
*** openstackgerrit has quit IRC13:53
*** openstackgerrit has joined #openstack-lbaas13:53
*** ganeshna has joined #openstack-lbaas13:57
*** pcaruana has quit IRC14:17
*** vivek-ebay has joined #openstack-lbaas14:24
*** ganeshna has quit IRC14:25
*** xgerman has joined #openstack-lbaas14:25
*** xgerman has quit IRC14:26
*** xgerman has joined #openstack-lbaas14:28
*** Varun_Lodaya has joined #openstack-lbaas14:31
*** vivek-ebay has quit IRC14:55
*** amotoki has joined #openstack-lbaas14:57
*** jschwarz has joined #openstack-lbaas14:58
*** martin001 has joined #openstack-lbaas15:39
martin001Have a question about healthmonitors with lbaas. If you have multiple health monitors associated with a pool, is the member status determined by the OR or the AND of the responses to the individual monitors. Any help much appreciated15:42
*** busterswt has joined #openstack-lbaas15:45
martin001Or is it something althogether different - e.g. the memeber status is the result  of the last monitor heartbeat, whatever it is, and regadless of the number of monitors15:45
*** jschwarz has quit IRC15:52
*** rm_mobile has joined #openstack-lbaas15:52
rm_mobilejohnsom, you there?15:52
johnsomWhat is up rm_mobile?15:54
rm_mobilejohnsom: in sprint planning right now, looking like your CR is still blocking us pretty hard. Are you back to actively working on it? Or is it actually ready yet?15:54
rm_mobileHaven't looked in the last day or two15:54
rm_mobileThe one above the queue consumer15:54
johnsomCR?  Controller Worker?  I have been working on it.  You can grab it and integrate, it just isn't fully functional.  Yesterday I started the stevedore to load the driver plugins15:55
openstackgerritMerged openstack/neutron-lbaas: Open Liberty development
johnsomShould have that done today.  Then it's updating for the database stuff we talked about yesterday and any fine tuning.15:55
johnsomThe API is all there though15:55
rm_mobileBut you're actively working on it?15:56
rm_mobileWe'd really like to see that done ASAP15:57
johnsomYeah, I am putting up patchsets for review weekly15:57
johnsomIf not more15:57
johnsomI want to see it done ASAP too15:58
rm_mobileMight want to move to like... Daily or more :P I'm available to review any time, and I'm sure blogan would too15:58
johnsomYou are welcome to take a pass on the WIP15:58
rm_mobileI'll review it five times a day if you have that maybe patches to push :)15:59
rm_mobileWill review that when we get out of our planning meeting this afternoon then15:59
johnsomAreas I know need work are amphora driver tasks needs updating and stevedore, the database work for updates, the some workflow stuff around the network driver15:59
johnsomOk, thanks16:00
*** mlavalle has joined #openstack-lbaas16:07
*** vivek-ebay has joined #openstack-lbaas16:15
*** Varun_Lodaya has quit IRC16:17
*** madhu_ak has joined #openstack-lbaas16:19
*** apuimedo has quit IRC16:24
xgermanrm_work once you guys get out we probbaly shoudl chat about logging, too -17:02
*** mwang2 has joined #openstack-lbaas17:02
*** Youcef has joined #openstack-lbaas17:02
xgermanthat still leaves me confused17:02
*** openstack has quit IRC17:13
*** openstack has joined #openstack-lbaas17:13
*** amotoki has quit IRC17:21
*** sbfox has joined #openstack-lbaas17:23
*** SumitNaiksatam has quit IRC17:29
*** bharath has joined #openstack-lbaas17:35
*** sai has joined #openstack-lbaas17:39
*** martin001 has quit IRC17:40
*** SumitNaiksatam has joined #openstack-lbaas17:51
*** mestery has quit IRC18:07
*** mestery has joined #openstack-lbaas18:10
*** sbfox has quit IRC18:11
*** sbfox has joined #openstack-lbaas18:32
rm_workjohnsom: jeebus crisp! somehow I hadn't noticed this was 4700+ lines of code until I got a few files in and was like "wait, how far through this am I?" T_T18:38
bloganmestery: thanks! just got a chance to read the backlogs18:39
johnsomrm_work That is why I have been doing regular WIPs18:39
mesteryblogan: no worries, I just signed you up for more work :)18:39
johnsomHalf of it is tests if that makes you feel better18:39
rm_workjohnsom: yeah, I guess there isn't really a clean way to split it up?18:39
rm_workwell, we're going to hack away at it today, want to get this merged ASAP18:40
johnsomYeah, pretty intertwined18:40
johnsomAre you committing changes or reviews?18:40
johnsomChanges I would hope we can coordinate.  Ok, cool18:40
johnsomWelcome them.  There is still work to do, but good to get comments early18:41
bloganmestery: lol you will pay for this one day!18:41
bloganjohnsom: i'm reviewing as well, i'll be extra pedantic18:42
rm_worki don't know a whole lot about taskflow, so I'm mostly just looking to make sure things make some logical sense right now18:48
rm_workwill have to test it later... does it "run"?18:48
*** sbfox has quit IRC18:48
johnsomYes, it runs.  It will create an amphora vm and LB18:50
johnsomThe rest, no so much yet.  I.e. the amphora driver stuff is borked18:50
*** sbfox has joined #openstack-lbaas19:04
*** mwang2 has quit IRC19:08
*** madhu_ak has quit IRC19:12
*** mwang2 has joined #openstack-lbaas19:21
*** sai has quit IRC19:35
ptoohillso, logging?19:37
ptoohillxgerman, want to discuss?19:41
xgermanI talked with johnsom and he thinks it might be best to just comingle all the listener logs in one log file19:41
ptoohillwhich is fine for me(us) because i believe well most likely have a parser that sends else where for customers and isnt part of scope till later19:42
ptoohillThough, that review is probably not needed anymore19:42
xgermannot so fast ;-)19:42
ptoohillyou want to change the tag for http logs?19:42
xgermanit seems haproxy has a global log file for errors and a way to have http logs send elswehere19:43
xgermanyeah, I was wondering if we should send that in different files19:43
*** madhu_ak has joined #openstack-lbaas19:43
ptoohillwell the identifier is the log tag19:43
ptoohillwe can totally send different err logs else where, but we would probably do that via rsyslog config yea?19:44
ptoohillso, if we want to do those type of things we would need either another service or the driver to configure/restart rsyslog19:44
xgermanyep, true19:45
ptoohillatleast thats what ive gathered from my readings lastnight. please correct if wrong. Im not 100% (not even 50%) confident in the syslog logging stuffs19:45
ptoohillbut, if thats the case i dont think thats a priority till 1.0?19:46
ptoohilland should be ok for now19:46
xgermansame here. I have seen some haproxy configs using more than one local -- but I think we can just leave local0 and put that all together for a leter point in time19:46
ptoohillisnt the local* the syslog levels?19:46
ptoohillso those identify the logs with a tag right?19:46
xgermanmmh, I thought it's local0.level...19:47
ptoohilli was seeing where local0 meant 'info' and local1 meant 'notice' or something similar19:48
xgermanyeah, you needed to make it so19:48
xgermanin haproxy19:48
ptoohill1.5 does it a bit differently also19:48
xgermanhence my confusion19:48
ptoohilland alot of these blogs show the 'old' way that actually gets ignored19:48
ptoohillindeed :/19:48
rm_workjohnsom: posted some comments, nothing major I don't think, but like I said I am new to taskflow so I have to just assume a lot of that code will work :P19:50
xgermanwell, so probably vest is to kill my review and keep local019:50
ptoohillfor now atleast untill we need something more19:50
xgermanok, sounds good, will abandon19:50
rm_workhave to run home to take care of some stuff, taking a half day :/19:50
xgermanI thought getting the logs separated by listerner would be easy... what a foolish thought...19:53
ptoohillsame here :/19:54
ptoohillwas trying to get hold of bedis to see if there was any other way to handle this. But unless we get some kind of feature request in i dont think there is19:55
*** rm_mobile has quit IRC20:11
*** mwang2_ has joined #openstack-lbaas20:21
*** mwang2 has quit IRC20:24
openstackgerritPhillip Toohill proposed stackforge/octavia: Preparing for tempest testing
*** goncalo has joined #openstack-lbaas20:48
dougwigblogan: you'll never guess where i am right now.20:53
dougwighow did you ever guess?20:53
bloganyour mailing address is an airport20:54
dougwigheh, then why am i #17 on the upgrade list?  hmm?20:58
bloganjust bc you live there doesn't make you important20:58
dougwigdamnit, openstack follows me everywhere.21:00
bloganyou mean you logging into openstack irc channels everyday?21:00
*** crc32 has joined #openstack-lbaas21:01
xgermanopenstack is stalking you?21:03
dougwigno, my lack of importance is infectious.21:04
xgermannot that it will help but you can change yiur name to King and make Doug your middle name - and the inist thye call you King Doug?21:05
dougwigthe board with the list publicly lists me as "WIE, D".  what can we do with four letters?21:06
dougwigKarl Fuchs ?21:06
xgermanLord (
dougwigalaska's policy is pretty easy to game, but only if i have advance notice of the trip.   and if i had advance notice of trips to san jose, i wouldn't have a connecting flight anyway, so it'd be a turboprop without a first class cabin.  so, i'm just bitching for no reason.21:12
johnsomOh I have memories of Boise and turboprops....21:16
*** rm_you has joined #openstack-lbaas21:16
*** rm_you has joined #openstack-lbaas21:16
*** TrevorV_ has joined #openstack-lbaas21:18
*** bharath has quit IRC21:38
*** bharath has joined #openstack-lbaas21:38
*** jorgem has joined #openstack-lbaas21:44
openstackgerritGerman Eichberger proposed stackforge/octavia: Implements the haproxy amphora agent api server
*** sbfox has quit IRC21:58
xgermanrm_work can you have another look at
xgermanI really like to have that out of my hair22:04
*** sbfox has joined #openstack-lbaas22:06
johnsomTrevorV Ping.  Your SSH driver isn't taking log in on the class init but the noop and Susanne's driver are.22:18
johnsomCan you guys discuss and align?22:19
*** xgerman has quit IRC22:28
*** TrevorV_ has quit IRC22:41
*** sbfox has quit IRC22:45
*** Varun_Lodaya has joined #openstack-lbaas22:47
*** jorgem has quit IRC22:48
*** bharath has quit IRC22:55
*** bharath has joined #openstack-lbaas22:56
*** mlavalle has quit IRC23:01
TrevorVjohnsom xgerman told me to remove it.23:11
*** xgerman has joined #openstack-lbaas23:11
johnsomxgerman did you ask Susanne to remove it too?23:11
TrevorV:S sorry johnsom :P23:15
TrevorVI do have a few comments for your review johnsom23:16
johnsomNot a problem, just need to get it aligned so the driver loading works23:16
johnsomCool, bring em on....23:16
openstackgerritMichael Johnson proposed stackforge/octavia: Removing a log parameter from the noop driver
openstackgerritMichael Johnson proposed stackforge/octavia: Removing a log parameter from the noop driver
xgermanyeah, ping rm_work or blogan for another +2 or are they alreday home because of Texas time?23:27
johnsomIt would be great if we can get that merged so I can rebase on it and have a noop driver to test with.23:28
*** crc32 has quit IRC23:38
*** ajmiller has quit IRC23:43
*** xgerman has quit IRC23:48
*** SumitNaiksatam has quit IRC23:53
rm_youi am around23:54
rm_youthis nickname is me at home23:54
rm_youjohnsom: i am confused, commented23:56

Generated by 2.14.0 by Marius Gedminas - find it at!