Friday, 2014-03-28

*** reed has quit IRC00:18
*** amitgandhi has quit IRC00:30
*** megan_w|afk is now known as megan_w00:32
*** megan_w is now known as megan_w|afk00:42
*** nosnos has joined #openstack-marconi01:37
*** flwang has quit IRC03:06
*** nosnos has quit IRC03:17
*** flwang has joined #openstack-marconi03:39
*** megan_w|afk is now known as megan_w03:52
*** nosnos has joined #openstack-marconi04:06
*** megan_w is now known as megan_w|afk04:23
*** chandankumar_ has joined #openstack-marconi04:23
*** chandankumar_ has quit IRC04:23
*** flwang has quit IRC04:30
*** flwang has joined #openstack-marconi04:31
openstackgerritFei Long Wang proposed a change to openstack/marconi: Improve cleanup of test_insert_queue_header_asterisk
openstackgerritFei Long Wang proposed a change to openstack/marconi: Improve cleanup of test_insert_queue_header_asterisk
chandan_kumarflaper87|afk, heya!06:04
*** flwang has quit IRC06:35
*** flwang has joined #openstack-marconi06:37
*** saju_m has joined #openstack-marconi06:38
*** flwang has quit IRC06:41
*** flwang has joined #openstack-marconi06:43
*** flwang has quit IRC07:02
*** AAzza has joined #openstack-marconi07:30
*** haomaiw__ has quit IRC07:33
*** jeremyh28 has joined #openstack-marconi08:01
*** flaper87|afk is now known as flaper8708:05
*** jeremyh281 has joined #openstack-marconi08:05
*** jeremyh281 has left #openstack-marconi08:06
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Add a help text for auth_strategy
flaper87chandan_kumar: yo yo08:24
chandan_kumarflaper87, how are you?08:26
flaper87chandan_kumar: I'm doing fine, what about you?08:27
chandan_kumarflaper87, i am also good. .08:28
chandan_kumarflaper87, have a look this hilarious ad .08:28
*** jeremyh28 has quit IRC08:32
*** saju_m has quit IRC09:05
*** saju_m has joined #openstack-marconi09:25
*** jamie_h_ has joined #openstack-marconi09:47
*** nosnos has quit IRC10:16
*** flaper87 is now known as flaper87|afk10:18
*** flaper87|afk is now known as flaper8710:42
*** AAzza has left #openstack-marconi10:54
*** tedross has joined #openstack-marconi11:48
*** saju_m has quit IRC11:50
*** oz_akan_ has joined #openstack-marconi12:24
*** mwagner_bbl has quit IRC12:35
*** sriram has joined #openstack-marconi12:36
*** AAzza has joined #openstack-marconi12:50
*** chandan_kumar has quit IRC12:51
*** cyrichardson has joined #openstack-marconi13:11
*** cpallares has joined #openstack-marconi13:14
*** saju_m has joined #openstack-marconi13:15
*** amitgandhi has joined #openstack-marconi13:21
*** oz_akan_ has quit IRC13:22
*** amitgandhi has quit IRC13:22
*** oz_akan_ has joined #openstack-marconi13:22
*** amitgandhi has joined #openstack-marconi13:23
*** mpanetta has joined #openstack-marconi13:27
*** vkmc has joined #openstack-marconi13:33
*** vkmc has quit IRC13:33
*** vkmc has joined #openstack-marconi13:33
*** mpanetta_ has joined #openstack-marconi13:38
*** mpanetta has quit IRC13:38
*** flwang has joined #openstack-marconi13:42
*** malini_afk is now known as malini13:46
*** mpanetta has joined #openstack-marconi13:49
*** balajiiyer has joined #openstack-marconi13:51
*** Kuwagata has joined #openstack-marconi13:51
*** mpanetta_ has quit IRC13:52
*** mwagner_bbl has joined #openstack-marconi13:53
maliniHappy Friday!!14:00
maliniI just noticed this in the ML
*** jmckind has joined #openstack-marconi14:00
maliniMight be a great idea for Marconi to be represented14:07
maliniWe 'll get validation from users14:07
alcabrera|zzgood morning, all. :)14:08
*** alcabrera|zz is now known as alcabrera14:08
*** russellb is now known as rustlebee14:08
malinihello alcabrera14:11
alcabreramalini: +1 about that14:11
alcabreraops at every session makes sense, since a system like OpenStack is pretty much deployer-driven14:12
alcabreraor should be14:12
cpallaresmalini, alcabrera: o/14:13
malinigood morning cpallares!14:13
alcabreracpallares: hey! :D14:14
openstackgerritOz Akan proposed a change to openstack/marconi: keystone middleware version change due to a bug
*** openstack has joined #openstack-marconi14:21
alcabreraoz_akan_: thanks14:24
alcabreracpallares: how go things?14:24
cpallaresalcabrera, malini: Good morning!14:24
cpallaresalcabrera: Pretty good and you? I slept so well last night. I was so tired yesterday.14:25
alcabreracpallares: I'm great! Glad to hear you slept well. What got you so tired in the first place? Too much work/travel? :)14:26
cpallaresalcabrera: Well I didn't sleep very well in Boston since I would get up really early (plus the hour difference) and I would go back really late since there were a lot of events happening at night. That plus the 4-5+ hours walking everyday exploring the city was exhausting.14:28
alcabreraah, makes perfect sense.14:28
cpallaresalcabrera: These past couple of days I haven't been sleeping well because I guess I'm still on that schedule :P14:28
alcabreratravel usually kills me, since I get my sleep all out of whack and then it's even harder to sleep because of the excitement of the travel itself14:28
cpallaresalcabrera: yeah! I love being in new cities :)14:29
cpallaresalcabrera: Sleep can wait till I get back to boring old home.14:29
alcabreraI'm roughly like that. :P14:30
alcabreraToo many people to see, too many foods to eat, too many Tweets to Tweet14:30
alcabrerano time for sleep14:30
maliniNothing can stop me from sleeping :D14:30
maliniThat makes me a good traveller14:31
cpallareshaha malini. I'm that way at home :P14:31
alcabreraand thusly, the world was balanced. malini will sleep for all of us, and in microincrements, we shall all sleep enough to offset the sleep that flaper87 doesn not get14:31
alcabrera*does not14:31
* cpallares wonders if flaper87 even sleeps14:31
flaper87alcabrera: :D :D :D14:31
* flaper87 doesn't sleep14:31
maliniSomebody needs to bear the brunt for the team ..I volunteer to do that ;)14:31
alcabreraso generous, malini. :D14:32
maliniI am always generous like that ;)14:33
maliniflaper87: did you see my post above from the ML ?14:33
alcabreracpallares: I've got an awesome math article for you: (math refactored: set theory -> type theory)14:35
flaper87malini: nope, what happened ?14:37
flaper87oh mmh14:37
cpallaresalcabrera: :D14:38
*** rwsu has quit IRC14:38
maliniflaper87: I think it'll be a great opportunity for Marconi14:38
maliniWe can get feedback from users + convince the skeptics that we are solving problems with real use cases14:38
*** rwsu has joined #openstack-marconi14:41
*** megan_w|afk is now known as megan_w14:43
openstackgerritA change was merged to openstack/marconi: keystone middleware version change due to a bug
flaper87people, we should really add new design session proposals asap14:57
flaper87alcabrera: oz_akan_ amitgandhi ^^^^^14:57
flaper87I know you guys had some ideas14:57
alcabreralink, flaper87?14:58
alcabreraI don't remember having any recent ideas, but perhaps scanning the current proposals will inspire me. ;)14:58
flaper87there are 4 proposed sessions, we'll likely merge some of those15:00
flaper87we've space for just 4 sessions15:00
amitgandhii think my ideas were around showcasing customers, and production stats - it looks like its covered by malini's proposal
*** jcru has joined #openstack-marconi15:01
amitgandhiare design sessions 1 hr?15:02
amitgandhii wonder if its too much info for one session and should be broken up15:02
malinifeel free to edit mine (I dont know if you can)15:02
alcabrera173 sessions in all proposed on that page so far15:04
*** zyuan has joined #openstack-marconi15:04
alcabrerafmap (length . lines) $ readFile "sessions.txt"15:04
alcabrerataking a look, I'm pretty happy with the proposed sessions for Marconi15:07
alcabreraI'm thinking... "What gap have we left uncovered...?"15:07
alcabreraI suppose an obvious one would be storage backends15:07
alcabreraand what the plans are for supporting partially-compatible backends15:08
*** zyuan has quit IRC15:08
alcabrerae.g., amqp being able to implement most of the API, but not all15:08
alcabreraOther gaps...15:08
alcabreraAPI Evolution15:08
*** reed has joined #openstack-marconi15:08
alcabreraWhat's coming for 2.0 and 1.1?15:08
*** zyuan has joined #openstack-marconi15:09
*** zyuan has quit IRC15:09
alcabreraI think devs/ops/users would have something to say on that, and we'd be able to cover a lot of feedback quickly /cc kgriffs|afk15:09
*** zyuan has joined #openstack-marconi15:09
*** zyuan has quit IRC15:10
alcabreraBenchmarking for Marconi might be another interesting one -- what would people consider a worthwhile benchmark of a queuing system?15:10
alcabrerae.g., 3DMark : grapgics engines :: QueueBench : Queue Systems15:10
*** zyuan has joined #openstack-marconi15:10
*** zyuan has quit IRC15:12
amitgandhialcabrera: rather than asking what might be worthwhile benchmarking, i think a better topic would be what kind of performance do users expect and why15:13
amitgandhias that is why ppl benchmark - to see if it meets their needs15:14
malinitj*** already asked for the benchmarks in ML15:14
maliniWe shpuld just run a benchmark of whatever we have now & start the conversation on what else ppl want to see15:15
amitgandhiyeh its something that should happen in the ML15:15
alcabreraamitgandhi: noted. However, given that these are design sessions, how would founding a design session on the question "what performance levels do you expect of your queuing system?" influence concrete design proposals?15:16
mpanettaOne thing that may help, is to split the benchmarks up by operation... Queue Creation/Deletion for instance does not need to be as performant as posts/claims/deletes15:16
mpanettaJust a suggestion...15:16
amitgandhihow does the method of benchmarking help design of marconi?15:16
maliniWe can also get answers from the operators during this session
maliniI added Marconi as an interested project in etherpad15:17
malinimpanetta: +115:17
mpanettaI keep seeing people talking about benchmarks, but not talking about what we are benchmarking...  Saying that we are 'benchmarking marconi' is kind of nebulous...15:17
amitgandhii almost feel like we need to present our benchmark findings (on the ML) and the design session should be "is this acceptab;e15:18
maliniWe should just point them to our deployment & benchmark configs along with the results we have15:18
amitgandhiwhat kind of creation time are you looking for15:18
maliniSo ppl can play with it & reach their own conclusions15:18
amitgandhiwhat kind of read perf are u looking for15:18
amitgandhiwhy do you need that perf15:18
amitgandhihow can we meet that need?15:18
*** zyuan has joined #openstack-marconi15:19
*** zyuan has joined #openstack-marconi15:19
amitgandhiand that will drive the importance of things like redis15:20
amitgandhiwe can discuss redis with them15:20
amitgandhiwill that meet their need15:21
amitgandhiare there other issues with that15:21
amitgandhithat we didnt think of15:21
amitgandhimaybe cust perf needs vary a lot15:21
amitgandhis/vary/varie ?15:21
alcabrerahow do we frame that as something actionable? The problem with asking "How much performance do you need for these operations?" is that we'll either get "as fast as you can go" or "as fast as my existing system"15:23
amitgandhithe question then becomes what is your existing system, and does marconi meet that need15:23
amitgandhiwhy do you need that performace15:23
amitgandhiis it realistic for marconi to reach that performance15:23
amitgandhiwhat can marconi achieve15:23
amitgandhithe actionable items become what storage drivers do we need to address15:24
amitgandhiis amqp really important15:24
*** ayoung-afk is now known as ayoung15:24
amitgandhior is redis important15:24
amitgandhior something else15:24
amitgandhiwhat perf do our users expect us to achieve15:24
amitgandhiit gives us something to aim for15:25
amitgandhiit tells us what our users want15:25
amitgandhi(that marconi aims to solve for)15:25
flaper87sorry, I keep getting pulled into other things but I wanted to say that we shouldn't think that we are "fast enough"15:25
flaper87we should always try to be faster15:25
amitgandhiflaper87: +115:25
flaper87regardless we've met the user's need or not15:25
amitgandhibut we should know if we are even in the ballpark15:25
alcabrerathat's part of the reason I feel that the performance discussion won't be fruitful15:25
amitgandhiie user expects usec, and we are msec15:26
alcabreradesigning a means to measure it *is*15:26
flaper87We can know that by comparing marconi's performance with other queuing services /technologies15:26
malinialcabrera: & We should make it easy for anybody who wants to benchmark15:26
flaper87(note the difference services AND technologies)15:26
flaper87SQS and qpid / rabbitmq, that is15:26
alcabreradefinitely, flaper8715:27
flaper87whether Marconi will ever be as fast as rabbit or not is not under discussion15:27
maliniGive them the deployment/benchmrk scripts15:27
flaper87but it should *always* try to get there15:27
flaper87So the closer we are to qpid / rabbit / zmq / SQS / SNS / IronMQ the better15:27
alcabreramaybe I'll write queue bench -- asynchronous/distributed, overload a typeclass to handle new service/techs, give it a config file, and take down a queueing system15:28
flaper87It really depends on what storage Marconi is using and that makes it *really* important to have bench tables w/ different storage drivers and transprots15:28
alcabreraqueuebench marconi_bench.conf --(html, txt, json) > report.(html, txt, json)15:30
alcabreraqueuebench sqs_bench.conf15:30
alcabreraqueuebench ironio_bench.conf15:30
flaper87alcabrera: that sounds really interesting15:30
alcabreracabal install queuebench15:31
flaper87alcabrera: TBH, we could start by just benching marconi and coparing it to whatever bench exists out there for those services15:31
flaper87alcabrera: LOL15:31
alcabreraI'll see what I can do on the next rax:hackday15:31
alcabreraflaper87: definitely. I know we've used tsung extensively for determining how rax:marconi behaves under load. I wonder what other benchmarking efforts have been done with queue systems? :)15:32
*** zyuan has left #openstack-marconi15:34
*** saju_m has quit IRC15:35
flaper87alcabrera: that's a really good question. TBFH, I'm not sure if there are "standard" benchmarks for queuing systems15:36
* flaper87 googles15:36
alcabrerathat'd be pretty cool if there were15:36
alcabreraas far as I've seen, it's a lot of ad hoc measurements15:37
alcabreraat various levels of system load15:37
alcabrerawith latency expressed as averages with stddevs at various load points15:37
*** reed has quit IRC15:54
*** AAzza has quit IRC16:18
*** reed has joined #openstack-marconi16:32
oz_akan_about benchmarking, marconi can be really fast if you give enough hardware and fast network16:33
oz_akan_I think the challenge is reducing the cost16:33
oz_akan_and "fast" is so arguable... fast for what type of application16:34
oz_akan_fast for writes or reads..16:34
oz_akan_anyone in operations knows not to rely on any published benchmarks. people would like to do their own benchmarks16:34
alcabreraoz_akan_: +116:34
oz_akan_so we may try to provide what affects performance how16:35
oz_akan_more shards, number of replicas, web servers etc..16:35
oz_akan_number of uwsgi instances16:35
oz_akan_kernel parameters16:35
oz_akan_then anyone can run his benchmark with his workload16:35
flaper87oz_akan_: that's exactly the issue and you detailed it very well16:39
flaper87normally, queuing systems are benched under different scenarios16:39
flaper87the common scenarios are (at least the ones I'm familiar with)16:39
flaper87Tons of posts with small / big messages (no reads)16:39
flaper87Tons of reads with small / big messages (no writeS)16:40
flaper87and concurrent writes / reads of small / big messages16:40
flaper87obviously, The hardware, network and environment has to be the same16:40
flaper87We should always try to make it as fast as possible under the worst environment16:41
flaper87That is not much ram nor CPU16:41
oz_akan_that is for pub-sub test case I guess16:41
flaper87well, no, producer and consumer too.16:41
flaper87why would it be just for pub-sub ?16:42
oz_akan_you wrote no writes for reads test16:42
oz_akan_with claims you have to read and write16:42
flaper87but you can also read without claiming16:42
oz_akan_that is pub - sub16:42
flaper87you're still consuming the message16:43
oz_akan_claim is for producer - consumer16:43
oz_akan_so there is only one consumer16:43
oz_akan_pub - sub, there may be many subscribers16:44
*** rossk has joined #openstack-marconi16:46
flaper87I see what you mean, however that's not exactly what makes it pub-sub. We could have prod-consumer without using claims. What claims bring is the guarantee that a message will be delirvered once.16:49
flaper87btw, I didn't mean to be pedantic, sorry if that sounded a bit jerkish16:49
flaper87but yeah, we'd likely use claims for the prod / consumer test16:50
flaper87and well, by writes  I meant message post16:50
*** rossk has quit IRC16:51
*** rossk has joined #openstack-marconi16:51
oz_akan_what makes it pub-sub?16:59
oz_akan_since marconi doesn't really have subscription?17:00
*** jmckind has quit IRC17:03
oz_akan_flaper87: ^17:05
*** megan_w is now known as megan_w|afk17:09
oz_akan_can someone tell me what is about ?17:13
alcabreraoz_akan_: it's the design summit. My understanding is that marconi will get four 1-hour slots17:14
alcabreraoz_akan_: used for the purpose of discussing future design decisions with the rest of the OS community17:14
amitgandhisubscriber semantics for marconi is i have access to the queue and will just subscribe (read) messages, not claim/delete them17:15
amitgandhisubscribers keep track of the last seen id, so they dont read ones they already have17:15
amitgandhiand eventually those messages expire and disappear17:15
oz_akan_amitgandhi: check flaper87 s response17:16
oz_akan_how producer - consumer work without claiming messages, I couldn't get tat17:16
amitgandhii always think of producer-consumer as claiming17:17
oz_akan_alcabrera: so we have 4 proposals, should we have just one, or is it good to have many?17:17
amitgandhiif a consumer isnt claiming, arent they just a subscriber?17:17
oz_akan_amitgandhi: I would say so17:18
amitgandhiflaper87: ^^17:18
malinioz_akan_: we have 4 slots & would want to use all of those17:18
alcabreraamitgandhi: producer-consumer is not dependent on claiming. Claiming is only necessary for once-and-only-once delivery.17:18
amitgandhiflaper87: can you please elaborate on why a consumer who doesnt claim is still a consumer and not a subscriber?17:18
oz_akan_alcabrera: ok so you don't claim what do you do17:18
* flaper87 back17:18
amitgandhithen arent they just a subscriber if they dont claim17:19
flaper87so, yeah, it can get very confusing in Marconi because there's basically no theoretical separation of those patterns in the way marocni does things17:19
*** saju_m has joined #openstack-marconi17:19
amitgandhito me, publisher = producer.  consumer is one who claims, subscriber is one who just reads17:20
alcabrerayou read, possibly repeating tasks off the front of the producer queue, oz_akan_ - you may have duplicate processing, which might be okay/necessary if your producer-consumer model utilizes redundant processing17:20
flaper87however, theoretically speaking,  a pub-sub is works in a way that messages are produced without worrying who'll consume them and then there are several subscribers streaming those messages out17:20
oz_akan_alcabrera: that is pub - sub, you may read more than once, a message might be for many workers17:20
flaper87the big difference is that in a pub-sub fashiong the subscriber needs to keep track of the messages it already processed and move forward17:20
flaper87the consumer doesn't have to do that17:21
flaper87a consumer would get the message, do something, delete the message and get another one17:21
amitgandhicorrect.  consumer doesnt have to do that since they are claiming messages, hence messages read go away17:21
flaper87the subscriber gets a set of messages, keeps track of the last processed and get another set starting from last+117:21
amitgandhiso explain to me a consumer who doesnt claim17:21
oz_akan_flaper87: could you tell me how producer-consumer pattern happens without claiming17:21
flaper87but the consumer doesn't need claims, claims are just a way to guarantee once and only once17:22
amitgandhiso how is that diff to a subscriber17:22
flaper87just like I explained, without claims, though, the once and only once delivery is not guaranteed17:22
flaper87the difference is that the subscriber doesn't delete the message from the queue17:22
amitgandhiso a consumer who doesnt claim and keeps track of id === a subscriber17:22
flaper87it just process the message and moves on17:22
amitgandhia consumer who doesnt claim, and doesnt track id != subscriber ??17:22
*** megan_w|afk is now known as megan_w17:23
flaper87the consumer doesn't need to keep track of the id17:23
flaper87the consumer gets the message, deletes it and moves on17:23
flaper87a subscriber gets a message and moves on17:23
flaper87(by keeping track of the id and getting a new one with id+1)17:24
alcabrerathe semantics of messaging patterns are pretty cool. They're subtle, too. :)17:24
alcabrerathere's so many that aren't encoded in Marconi17:24
amitgandhiso if a consumer gets a message, deletes it and moves on (without claiming), then another consumer can also work on it, until consumer1 deletes it17:24
amitgandhii cant imagine why anyone would do that17:25
flaper87amitgandhi: I'm not saying it's the right thing to do17:25
flaper87I'm saying that *claims* don't define what prod-consumer is17:25
flaper87What defines producer and consumer is the way you produce and consume messages17:25
oz_akan_flaper87: do you mean it for marconi?17:25
oz_akan_are there different ways to produce messages?17:26
oz_akan_I think I just understand these as much as they are implemented in marconi17:26
oz_akan_that might be the problem17:26
flaper87oz_akan_: no differences in how messages are produced. There are some that think that in a prod-consumer pattern the producer *may* know who will consume the message but that's more a req-resp pattern -which we don't support)17:27
oz_akan_if there are multiple costumers of a message, it sounds like they are subscribers17:27
oz_akan_anyway, let me read 2 books on messaging patterns before I say anything further :)17:27
amitgandhiive always thought of consumer as claiming.  if amit consumes a coke, then no one else can have that coke17:29
amitgandhii guess you can have amit consumes a coke while everyone else watches?17:29
amitgandhibut then im still the only consumer and everyone else is a subscriber17:29
* amitgandhi takes another sip of his coke17:30
* alcabrera duplicates the coke and passes it on the amitgandhi-217:30
*** saju_m has quit IRC17:35
*** saju_m has joined #openstack-marconi17:36
*** cpallares has quit IRC17:41
*** Kuwagata has quit IRC18:24
*** Kuwagata has joined #openstack-marconi18:31
*** jamie_h_ has quit IRC18:35
*** openstackgerrit has quit IRC18:48
*** openstackgerrit has joined #openstack-marconi18:48
*** jamie_h_ has joined #openstack-marconi19:06
*** malini has left #openstack-marconi19:11
*** prashanthr_ has quit IRC19:24
*** Kuwagata has quit IRC19:27
*** prashanthr_ has joined #openstack-marconi19:31
*** prashanthr_ has quit IRC19:39
*** megan_w is now known as megan_w|afk20:00
*** alcabrera is now known as alcabrera|zz20:08
*** jamie_h_ has quit IRC20:09
*** vkmc has quit IRC20:12
*** prashanthr_ has joined #openstack-marconi20:13
*** metral has quit IRC20:15
*** metral has joined #openstack-marconi20:16
*** metral has quit IRC20:22
*** metral has joined #openstack-marconi20:22
*** metral has quit IRC20:22
*** metral has joined #openstack-marconi20:23
*** oz_akan_ has quit IRC20:24
*** metral has quit IRC20:26
*** metral has joined #openstack-marconi20:26
*** metral has quit IRC20:28
*** prashanthr_ has quit IRC20:28
*** metral has joined #openstack-marconi20:28
*** sriram has quit IRC20:30
*** metral has quit IRC20:33
*** metral has joined #openstack-marconi20:34
*** prashanthr_ has joined #openstack-marconi20:35
*** metral has quit IRC20:37
*** mwagner_bbl has quit IRC20:37
*** prashanthr_ has quit IRC20:41
*** prashanthr_ has joined #openstack-marconi20:46
*** metral has joined #openstack-marconi20:48
*** balajiiyer has quit IRC20:49
*** metral has quit IRC20:54
*** metral has joined #openstack-marconi20:58
*** mpanetta has quit IRC20:59
*** amitgandhi has quit IRC21:01
*** metral has quit IRC21:02
*** metral has joined #openstack-marconi21:02
*** metral has quit IRC21:06
*** metral has joined #openstack-marconi21:07
*** jergerber has joined #openstack-marconi21:08
*** metral has quit IRC21:18
*** prashanthr_ has quit IRC21:18
*** cyrichardson has quit IRC21:20
*** tedross has quit IRC21:23
*** prashanthr_ has joined #openstack-marconi21:24
*** prashanthr_ has quit IRC21:32
*** prashanthr_ has joined #openstack-marconi21:37
*** prashanthr_ has quit IRC21:40
*** prashanthr_ has joined #openstack-marconi21:42
*** mwagner_bbl has joined #openstack-marconi22:11
*** vkmc has joined #openstack-marconi22:55
*** jergerber has quit IRC22:57
*** jcru has quit IRC23:13
*** rossk has quit IRC23:40
*** vkmc has quit IRC23:52

Generated by 2.14.0 by Marius Gedminas - find it at!