Friday, 2015-08-07

*** breitz has quit IRC00:05
notmynameho: are you coming to austin next week?00:06
notmynameho: also, I saw that there was a patch from someone else at fujitsu. that's great!00:07
notmynameok, I'm headed home00:09
*** sanchitmalhotra has joined #openstack-swift00:13
*** sanchitmalhotra1 has quit IRC00:15
openstackgerritMinwoo Bae proposed openstack/swift: Error handling of DiskFileNoSpace()  https://review.openstack.org/19890900:18
*** chlong has joined #openstack-swift00:27
honotmyname: i'll be there :-)00:33
honotmyname: yeah, we try to increase contributors and I'm a teacher for it.00:35
mattoliverauho: nice ho-sensei :)00:37
homattoliverau: :-)00:37
*** minwoob has quit IRC00:39
*** sanchitmalhotra1 has joined #openstack-swift00:49
*** zhill has quit IRC00:49
*** jasondotstar has quit IRC00:50
*** sanchitmalhotra has quit IRC00:51
*** jasondotstar has joined #openstack-swift00:52
*** jasondotstar has quit IRC00:56
*** sanchitmalhotra has joined #openstack-swift00:59
*** sanchitmalhotra1 has quit IRC01:01
*** sanchitmalhotra1 has joined #openstack-swift01:05
*** sanchitmalhotra has quit IRC01:07
*** bill_az has quit IRC01:10
*** kota_ has joined #openstack-swift01:16
*** ChanServ sets mode: +v kota_01:16
kota_good morning01:17
*** tongli has joined #openstack-swift01:18
notmynameho: that's great to hear (on both counts)01:21
*** mfalatic has quit IRC01:21
*** mfalatic_ has joined #openstack-swift01:21
*** jasondotstar has joined #openstack-swift01:27
*** sanchitmalhotra has joined #openstack-swift01:36
*** blmartin has joined #openstack-swift01:38
*** sanchitmalhotra1 has quit IRC01:39
*** wbhuber has joined #openstack-swift01:43
*** hrou has joined #openstack-swift01:50
*** robefran_ has joined #openstack-swift02:06
*** hezhiqiang has joined #openstack-swift02:07
*** swat30 has quit IRC02:08
*** robefran has quit IRC02:09
honotmyname: thanks! as for the contributors, it looks good progress and i think it really good things for us we can be open software engineers.02:10
hokota_: morning!02:10
*** zul has joined #openstack-swift02:11
*** swat30 has joined #openstack-swift02:12
kota_ho: nice to hear that more contributors will be in Japan :-)02:16
*** tongli has quit IRC02:16
*** tsg has quit IRC02:25
*** yuan has joined #openstack-swift02:35
*** jrichli has joined #openstack-swift02:38
*** lcurtis has joined #openstack-swift02:38
*** wbhuber has quit IRC02:46
hokota_: yeah, tokyo summit push our activity. it's great!02:51
*** breitz has joined #openstack-swift03:04
*** silor has joined #openstack-swift03:05
openstackgerritJanie Richling proposed openstack/swift: Trivial Key Master for encryption  https://review.openstack.org/19374903:10
*** sakaYK has joined #openstack-swift03:14
*** sanchitmalhotra1 has joined #openstack-swift03:19
*** silor has quit IRC03:20
*** silor has joined #openstack-swift03:20
*** sanchitmalhotra has quit IRC03:22
openstackgerritMerged openstack/swift: pep8 fix: assertEquals -> assertEqual  https://review.openstack.org/20997603:23
*** kota_ has quit IRC03:40
*** blmartin has quit IRC03:46
*** jasondotstar has quit IRC04:09
homattoliverau: is it possible have your re-review for patch 138342 before hack-a-thon? (starred by PTL patch)04:11
patchbotho: https://review.openstack.org/#/c/138342/04:11
mattoliverauho: sure, will do my best to, I've got gertty setup so I can review on the plane :)04:12
*** gyee has quit IRC04:13
mattoliverauif I don't get to it today.04:13
mattoliverauI'm just in the middle of packing, so long as I don't have to run out and buy missing things, I'll get a few reviews done beofre the end of the day :)04:14
* mattoliverau leaves for mid-cycle first thing in the morning. 04:14
jrichlimattoliverau: have a good flight!04:16
mattoliveraujrichli: thanks :) At least it's only 2 flights.04:16
jrichlimattoliverau: that's good, but sounds like long flights.  bring snacks!04:17
lcurtiscould someone answer a swift question for me? why would someone actually need to obtain a container listing?04:18
lcurtisif requests are specifically /account/container/object04:19
lcurtisis container listing necessary?04:19
mattoliveraujrichli: 14.5 hours to LAX then 3 to Austin, so only 17.5 hours in flight 20+ in transit.. easy :P04:19
jrichliwow.  and you do that a lot ... i have never had a flight over 9 hours, i think04:20
mattoliveraulcurtis: if you want to see what objects are inside a container.04:20
mattoliveraulcurtis: if your using dlo, then they are important04:20
notmynamejrichli: it's not as fun as it sounds (long flights) ;-)04:21
mattoliveraujrichli: 9 hours will be my flight to tokyo :).. so yeah except for NZ it's always long :P04:21
lcurtissorry, what does dlo stand for?04:21
mattoliveraulcurtis: Dynamic Large Objects04:22
jrichlinotmyname: have you experienced several long flights?04:22
lcurtisah..thank you04:22
mattoliveraujrichli: yeah he comes visit Oz in summer :)04:22
jrichlinice.04:22
mattoliverauI first heard about swift from notmyname in one of his LCA swift talks a few years back04:23
notmynameaustralia/nz, europe, hong kong, brazil, and morocco have all been long flights. and I've been to many of those more than once ;-)04:23
jrichlidang04:23
jrichliyou travel alot!04:23
jrichlibtw mattoliverau, I really got a kick out of that slip slap slop or whatever it was04:24
notmynamethis month (6 weeks?) I'll have been in portland, seattle, austin, and day trips to the peninsula04:24
mattoliveraulol, yeah, the 80s ad campain.. I remember when that was on TV :)04:24
mattoliverauSlip Slop Slap is still a common saying in Oz, so the campain definitely worked04:25
jrichlinotmyname: you could do a travel show.04:26
notmynamewhat is that? mattoliverau are you holding out on me for aussie slang?04:26
mattoliveraunotmyname: I'll go find the link again04:26
jrichlihttps://www.youtube.com/watch?v=b7nocIenCYg04:27
jrichliyou would be amazed at what i save off in my notes04:27
notmynamecompared to the rest of my family, I travel a lot. compared to a lot of people in opnestack, I don't. mattoliverau takes 2 flights and does 3 months of my travel ;-)04:27
jrichlido you make use of the freq flyer miles?04:29
mattoliveraujrichli: thanks thats is. notmyname ^^04:29
notmynameyeah, some.04:29
lcurtisso it is necessary to do a container list to retrieve a DLO?04:30
mattoliveraulol, only in distance, I'm not away for too long. Only 2 weeks this trip.04:30
notmynamelcurtis: yes04:30
notmynamewhich is both the best and worst thing about DLOs04:31
lcurtisThanks notmyname04:31
lcurtisI will understand this stuff yet04:31
notmynamelcurtis: also, container listing are used for accounting and for container reconciling if you have more than one policy04:31
notmynameand container sync to keep 2 clusters in sync04:32
notmynameand, internally, expiring objects04:32
lcurtisah...makes more sense then04:32
notmynameand, from the end user perspective, the first important thing is to store the data. the next thing is to know what you've stored. listings are a simple way to do that04:32
lcurtisin terms of writes to sqlitedb...could you recommend where i might find the sql for reads and writes to the sqlite db?04:34
lcurtisim curious as to why so many actual writes to disk based on x a mount of throughput to server04:34
lcurtiscontainer server that is04:35
notmynameI saw your question earlier, but I don't have a specific answer04:35
notmynamethe code is in container/backend.py and db.py04:35
*** darrenc is now known as darrenc_afk04:35
lcurtissure...i would imagine that is pretty vague question04:35
lcurtisi am more on syseng side compared to dba/dev04:36
notmynamefirst thing to check is if you were actually just doing contianer IO at the time. no replication or auditing or anything else04:36
lcurtisokay...guessing would be mostly container io04:38
*** _hrou_ has joined #openstack-swift04:38
lcurtisis there a document that shows workflow of client to proxy to account/container logic?04:39
*** nakagawamsa has joined #openstack-swift04:39
*** hrou has quit IRC04:40
*** sakaYK has quit IRC04:41
notmynamelcurtis: doesn't look like it :-(04:41
*** sakaYK has joined #openstack-swift04:41
notmynamelcurtis: I've drawn that on a whiteboard dozens of times for people, but it doesn't look like it's ever been written down or recorded04:41
notmynamehmm...let me check some videos04:41
lcurtisive done the ring one a few times04:42
notmynamenope04:42
mattoliveraulcurtis: as notmyname mentioned background services will also be doing things inside the dbs. The replicator will merge items, and relaim (clear out listings marked as deleted) so a good  to take them out of the equation too.04:43
lcurtisokay..no worries...i think ive reached my quota for pestering questions for the day04:43
lcurtismattoliverau - thats an excellent point04:44
mattoliveraulol04:44
lcurtisi really appreciate the help guys04:44
lcurtisthanks much04:44
mattoliveraunps anytime. But if were quiet next week it cause we're all locking in a room04:45
*** hezhiqiang has quit IRC04:46
jrichliI even got a chain and padlock from security to lock up the room :-)04:47
lcurtishehehe04:52
lcurtissorry..one more q...what are some of the largest known swift clusters in world?04:54
notmynamerackspace, hp, ibm, and ntt have all been public about theirs04:55
mattoliveraulcurtis: you should read notmyname's latest blog post, he mentions some :)04:55
mattoliveraubut yeah rackspace (I think I'm suppose to say that first) :p04:56
notmynameon http://www.rackspace.com/blog/happy-birthday-openstack/ Rackspace say 115+PB in Swift04:57
notmynamelargest single clusters are in the neighborhood of 50-80PB, from what I've been told04:58
lcurtisphew...okay04:58
notmyname(not necessarily Rackspace, but including them)04:58
lcurtisgreat info to know04:58
lcurtisthanks again04:59
*** darrenc_afk is now known as darrenc05:01
jrichlig'night all.  safe travels!05:03
*** jrichli has quit IRC05:04
*** SkyRocknRoll has joined #openstack-swift05:08
lcurtiserg..sorry...another one...checking this page, it notes container database is updated asynchronously to reflect new object in it https://swiftstack.com/openstack-swift/05:10
lcurtisi thought proxy service actually updated container?05:10
*** haomaiwang has quit IRC05:11
*** haomaiwa_ has joined #openstack-swift05:11
*** hezhiqiang has joined #openstack-swift05:11
notmynamethe proxy sends info to the object servers which attempt to update the container server synchronously at the time of the request. however, failure to update the container server will queue it for later (the request still succedes05:14
*** ppai has joined #openstack-swift05:15
*** jasondotstar has joined #openstack-swift05:16
lcurtisokay...got it...05:18
lcurtisawesome..ty05:19
*** breitz has quit IRC05:21
*** jasondotstar has quit IRC05:21
*** jamespage_ has joined #openstack-swift05:29
*** mahatic has joined #openstack-swift05:32
*** hezhiqiang has quit IRC05:40
*** sakaYK has quit IRC05:41
homattoliverau: thanks! you have a lot of time to austin. i hope it's enogh time for my review :-)05:44
mattoliverauho: lol, me too. When are you arriving in Ausin?05:45
*** zaitcev has quit IRC05:46
*** ppai has quit IRC05:58
*** _hrou_ has quit IRC05:59
openstackgerritMatthew Oliver proposed openstack/swift-specs: Update sharding spec, notably CountingTrie and shrinking  https://review.openstack.org/20100006:02
*** lcurtis has quit IRC06:03
*** zul has quit IRC06:06
*** ttrumm has joined #openstack-swift06:10
*** ppai has joined #openstack-swift06:12
*** sakaYK has joined #openstack-swift06:22
homattoliverau: tomorrow (means saturday)06:31
mattoliverauho: me too, might see you during the weekend then :)06:32
*** saltsa has quit IRC06:34
homattoliverau: yeah, breakfast in the hotel :-)06:35
mattoliverauSo your staying at home2 suites too, yay!06:36
*** ppai has quit IRC06:42
hoyay!06:46
*** ppai has joined #openstack-swift06:55
*** saltsa has joined #openstack-swift06:56
*** jamespage_ has quit IRC06:58
*** jasondotstar has joined #openstack-swift07:05
*** ppai has quit IRC07:09
*** jasondotstar has quit IRC07:09
*** nexusz99 has joined #openstack-swift07:15
*** ttrumm_ has joined #openstack-swift07:19
*** rledisez has joined #openstack-swift07:20
*** ttrumm has quit IRC07:20
*** ppai has joined #openstack-swift07:23
*** ppai has quit IRC07:33
*** ppai has joined #openstack-swift07:47
*** chlong has quit IRC08:04
*** jistr has joined #openstack-swift08:06
*** ttrumm_ has quit IRC08:14
*** acoles_ is now known as acoles08:14
acolesgood morning08:17
mahaticgood morning!08:18
*** joeljwright has joined #openstack-swift08:20
*** ChanServ sets mode: +v joeljwright08:20
cschwedehello!08:21
*** jordanP has joined #openstack-swift08:24
jordanPhey guys, did you know that twitch.tv (the streaming platform) was using swift ? That's cool08:24
*** ttrumm_ has joined #openstack-swift08:25
nexusz99jordanP, wow! that's cool !!! where can i found that information?08:34
jordanPnexusz99,  whatch a replay, then use wireshark...08:35
nexusz99jordanP, OK I will try thanks08:36
jordanPnexusz99, here the URL from the stream: http://vod.ak.hls.ttvnw.net/v1/AUTH_system/vods_17a1/dota2ti_15669404032_283207604/high/index-0000000779-It5H.ts?start_offset=19671944&end_offset=2047977908:36
jordanPit looks like swift...08:36
nexusz99jordanP,  wow..08:38
nexusz99jordanP,  yes.. i checked twitch.tv's response, they have the swift's response header (X-Trans-Id: tx0c89b1049635473bb6d04-0055c46e4c)08:40
*** ttrumm has joined #openstack-swift08:41
*** ttrumm_ has quit IRC08:45
*** ppai has quit IRC08:46
*** jasondotstar has joined #openstack-swift09:00
*** ppai has joined #openstack-swift09:00
*** eandersson has joined #openstack-swift09:07
mattoliverauMorning cschwede, acoles, mahatic, and anyone else I missed09:09
acoleshey mattoliverau09:09
mattoliverauOh and acoles, don't mention the ashes.. Tho I expect to here about it all next week :p09:10
onovyhi, can someone look to https://review.openstack.org/#/c/192562/ please?09:10
acolesoh mattoliverau i was biting my tongue :D09:10
mattoliverauLol09:10
acolesmattoliverau: on bbc news "11 aussies are seeking political asylum in uk for fear of persecution if they return home" :)09:11
mattoliverauacoles: they better be, there not welcome here anymore :P09:12
acolesheh. it ain't over yet.09:12
* acoles is overly cautious09:12
*** SkyRocknRoll has quit IRC09:13
mahaticMorning mattoliverau !09:13
acolesmattoliverau: so have your sports channels replaced ashes coverage with back re-runs of neighbours?09:14
*** ppai has quit IRC09:14
*** km has quit IRC09:15
*** ppai has joined #openstack-swift09:28
*** rledisez has quit IRC09:29
*** jasondotstar has quit IRC09:29
*** rledisez has joined #openstack-swift09:30
mattoliverauacoles: lol, yup so much neighbours :p09:37
*** jasondotstar has joined #openstack-swift09:42
*** sakaYK has quit IRC09:54
*** ttrumm has quit IRC09:59
*** marzif has joined #openstack-swift09:59
*** ho has quit IRC10:10
*** ppai has quit IRC10:12
*** jasondotstar has quit IRC10:13
*** mlanner has quit IRC10:17
*** joearnold has quit IRC10:18
*** dosaboy has quit IRC10:18
*** occup4nt has joined #openstack-swift10:19
*** acorwin has quit IRC10:19
*** occupant has quit IRC10:19
*** klrmn has quit IRC10:19
*** redbo has quit IRC10:19
*** redbo has joined #openstack-swift10:19
*** ChanServ sets mode: +v redbo10:19
*** dosaboy has joined #openstack-swift10:19
*** jistr has quit IRC10:20
*** klrmn has joined #openstack-swift10:20
*** ttrumm has joined #openstack-swift10:20
*** jistr has joined #openstack-swift10:20
*** jeblair has quit IRC10:21
*** joearnold has joined #openstack-swift10:22
*** mlanner has joined #openstack-swift10:22
*** jeblair has joined #openstack-swift10:22
*** acorwin has joined #openstack-swift10:23
*** silor has quit IRC10:23
*** ppai has joined #openstack-swift10:26
*** mahatic has quit IRC10:32
*** mahatic has joined #openstack-swift10:32
*** robefran has joined #openstack-swift10:33
*** ttrumm_ has joined #openstack-swift10:34
*** robefran_ has quit IRC10:35
*** SkyRocknRoll has joined #openstack-swift10:35
*** ttrumm has quit IRC10:36
*** jordanP has quit IRC10:45
*** mtreinish has quit IRC10:45
*** openstackgerrit has quit IRC10:45
*** jamespage has quit IRC10:45
*** HenryG has quit IRC10:45
*** dfg has quit IRC10:45
*** pgbridge has quit IRC10:45
*** wer has quit IRC10:45
*** cschwede has quit IRC10:45
*** raginbajin has quit IRC10:45
*** dhellmann has quit IRC10:45
*** cebruns has quit IRC10:45
*** ctennis has quit IRC10:45
*** d0ugal has quit IRC10:45
*** torgomatic has quit IRC10:45
*** ttrumm_ has quit IRC10:46
*** jordanP has joined #openstack-swift10:49
*** mtreinish has joined #openstack-swift10:49
*** jamespage has joined #openstack-swift10:49
*** openstackgerrit has joined #openstack-swift10:49
*** HenryG has joined #openstack-swift10:49
*** dfg has joined #openstack-swift10:49
*** pgbridge has joined #openstack-swift10:49
*** wer has joined #openstack-swift10:49
*** cschwede has joined #openstack-swift10:49
*** raginbajin has joined #openstack-swift10:49
*** dhellmann has joined #openstack-swift10:49
*** cebruns has joined #openstack-swift10:49
*** ctennis has joined #openstack-swift10:49
*** d0ugal has joined #openstack-swift10:49
*** torgomatic has joined #openstack-swift10:49
*** cameron.freenode.net sets mode: +vv cschwede torgomatic10:49
*** occup4nt has quit IRC10:57
*** occupant has joined #openstack-swift10:57
*** SkyRocknRoll has quit IRC11:12
*** bill_az has joined #openstack-swift11:16
*** sanchitmalhotra1 has quit IRC11:36
*** haomaiwa_ has quit IRC11:37
*** haomaiwa_ has joined #openstack-swift11:38
*** ppai has quit IRC11:42
*** lcurtis has joined #openstack-swift11:44
*** ttrumm has joined #openstack-swift11:47
*** ppai has joined #openstack-swift11:54
*** ttrumm has quit IRC11:54
*** robefran has quit IRC12:04
*** ttrumm has joined #openstack-swift12:04
*** lcurtis has quit IRC12:06
*** MVenesio has joined #openstack-swift12:09
*** ttrumm_ has joined #openstack-swift12:12
*** ttrumm has quit IRC12:15
*** ppai has quit IRC12:16
*** chlong has joined #openstack-swift12:16
*** ppai has joined #openstack-swift12:30
*** robefran has joined #openstack-swift12:34
*** tongli has joined #openstack-swift12:38
*** bapalm has joined #openstack-swift12:39
*** bill_az has quit IRC12:47
*** nakagawamsa has quit IRC12:49
*** ppai has quit IRC12:59
*** dustins has joined #openstack-swift13:05
*** joeljwright has quit IRC13:07
*** logan2 has quit IRC13:11
*** NM has joined #openstack-swift13:11
*** logan2 has joined #openstack-swift13:12
*** mgriffin has joined #openstack-swift13:18
mgriffinI am trying to use /usr/bin/swift 2.0.3, can I store -U (username) and -K (api key) in environment variables instead of args?13:19
*** hrou has joined #openstack-swift13:20
mgriffinOS_USERNAME didn't seem to work13:20
*** dustins has quit IRC13:21
*** joeljwright has joined #openstack-swift13:24
*** ChanServ sets mode: +v joeljwright13:24
*** petertr7_away is now known as petertr713:26
*** NM has quit IRC13:31
*** proteusguy has quit IRC13:41
*** dustins has joined #openstack-swift13:42
*** tongli has quit IRC13:45
mahaticacoles: wrt to yesterday's, removing crypto mock will pretty much error out all the tests since all them are asserting based on fake encryption13:46
mahaticall of them*13:46
*** cdelatte has joined #openstack-swift13:46
*** delattec has joined #openstack-swift13:46
acolesmgriffin: if you do 'swift --os-help' you will see details of the env vars to use13:47
acolesmgriffin: I think OS_USERNAME should be used in conjunction with OS_PROJECT_NAME13:47
acolesmgriffin: that's assuming you are using keystone as auth service13:48
mgriffinswift: error: no such option: --os-help13:48
mgriffinmaybe i have too old a version in ubuntu 14.0413:49
acolesmgriffin: hmm, yeah maybe that came i later than your version13:49
*** hw_ has joined #openstack-swift13:49
mgriffini suppose i can use -U and -K but wanted avoid this being visible in "ps"13:50
mgriffinacoles: thanks13:50
acolesmgriffin: ok, so OS_USERNAME=tester, OS_PROJECT_NAME=test, OS_PASSWORD=testing13:50
hw_Hi. Anybody here who can tell me, how to pass values through the FormPost Middleware? I know this is possible with Headers, but I only have a plain form, so this is no solution. I have already tested it with a query parameter. But I have looked into the source code of FormPost, the environment seems to be cleaned before the subrequest is made....13:51
hw_Mayby anybody has an idea?13:51
*** wbhuber has joined #openstack-swift13:51
acolesmahatic: oh :(13:52
acolesmahatic: so do they make assertions about the ciphertext value?13:52
hroumahatic, what assertions ? I wouldn't think the UT would care about the encryption algo used.13:53
acolesmgriffin: wait, on older versions you may need to use OS_TENANT_NAME, sorry, not sure when it changed13:53
acolesmgriffin: that is instead of OS_PROJECT_NAME13:53
hroumgriffin, acoles  - for V2 keystone, OS_TENANT_NAME + OS_AUTH_URL13:53
mahaticacoles: nope, assertions based on Fake_body13:53
*** proteusguy has joined #openstack-swift13:53
acolesmgriffin: keystone rebranded tenants as projects at some point13:54
mahaticacoles: ciphertext does not come into the picture in the code. We're only asserting the header values and fake body13:54
mahaticheaders that we set (as if the client would)13:56
acolesmgriffin: yes that should do it, and iirc OS_TENANT_NAME will still work with V3 but OS_PROJECT_NAME was also added13:56
mgriffinOS_PASSWORD is supposed to be the api key?13:56
*** tongli has joined #openstack-swift13:57
acolesmahatic: yes, by ciphertext i meant the body output from encrypter - sounds like the tests are making assertions about that base on mock cipher?13:57
acolesmgriffin: yes13:57
mahaticacoles: yes13:58
mgriffinacoles: strangely i get an auth error13:58
acolesmgriffin: :(13:59
openstackgerritZhao Lei proposed openstack/swift: Fix a spelling typo in comment  https://review.openstack.org/21044813:59
acolesmahatic: oic :(13:59
openstackgerritjanonymous proposed openstack/swift: pep8 fix: assertEquals -> assertEqual  https://review.openstack.org/20964713:59
*** jrichli has joined #openstack-swift14:01
mahaticacoles: so when I remove the crypto (I also had to change the key length for crypto to accept), the encrypted value doesn't match upto the fake_body obviously. So it needs refactoring I believe. I guess that's the way forward?14:01
acolesmgriffin: i just rolled back to swiftclient v2.0 and checked with OS_USERNAME, OS_TENANT_NAME, OS_PASSWORD, OS_AUTH_URL in env14:03
mahatics/crypto/crypto mock14:03
*** silor has joined #openstack-swift14:03
acolesmahatic: well its either that or start on a separate set of tests. i'd hate for you to change them all and then get voted down for not isolating the unit tests enough!14:06
acolesmahatic: as a start, how about duplicating one of the tests (is there a test of a PUT for example, replacing the mock crypto with the real thing and get that passing14:06
mahaticacoles: hehe, definitely sounds good14:06
acolesjust to see if there are any gotchas14:07
mahaticacoles: yes, I could do that. that is how I started testing too (removing fake crypto stuff)14:07
acolesmahatic: i'm just trying to avoid unnecessary test duplication, but also want to avoid wasted time on your part :)14:07
jrichlimahatic acoles: the tests do not assert any particular mock: it is just uses an encrypt/decrypt in the mock right now to assert14:08
jrichliso it will decrypt what the encrypter spits out to match it against plaintext14:09
mahaticacoles: right, makes sense :)14:09
mahaticjrichli: correct14:09
openstackgerritZhao Lei proposed openstack/swift: Fix a spelling typo in object-expirer.conf  https://review.openstack.org/21045314:09
jrichliso, just change the assert to not be that.  just add a new assertion like we talked about the other day in IRC ...14:09
jrichliill post in a minute14:09
acolesjrichli: i should stop hand-waving :)14:10
mahaticjrichli: replace the fake_encrypt part you mean?14:10
jrichliwell, you may as well remove the use of the fake crypto altogether.  basically, we are going to assert that the ciphertext is not cipherable14:11
*** ttrumm_ has quit IRC14:11
mahaticjrichli: right, I think for that acoles is suggesting to start off in a new thing because if I remove fake_crypto and someone thinks the unit tests are not isolated enough, that'll be a problem :)14:12
mgriffinacoles: alright ST_AUTH ST_USER and ST_KEY work in my environment14:16
acolesmgriffin: just for anyone listening in, you need ST_USER ST_KEY ST_AUTH for cloud files14:16
acoles:)14:16
mgriffinacoles: you think i should just spam ST_AUTH OS_USERNAME and OS_TENANT_NAME to avoid caring about this much?14:16
jrichlimahatic: the new assertion could be something like this (from notmyname): assertFalse(any(word in word_list for words in cyphertext.split()))14:17
jrichliwhere word_list are the words in your plaintext14:18
acolesmgriffin: err, not sure if the OS_ options being present in your env would cause swift client to assume you are using a keystone auth system, i can't remember the precedence.14:18
acolesmgriffin: might be best to keep it simple14:19
acolesnow it works14:19
*** hw_ has quit IRC14:19
hroujrichli, are you worried about leading to indeterminate results (assuming we use a different key / iv each UT invocation) could very well lead to the same word (though very unlikely)  ?14:20
mgriffinacoles: which auth system is using ST_AUTH?14:20
acolesmgriffin: whatever cloud files use i guess. (plus we have a dumb tempauth system for dev testing that uses it)14:21
jrichlihrou: seems like a safe gamble, but I am open to opinions14:21
*** jlhinson has joined #openstack-swift14:22
mahaticjrichli: acoles should I go ahead and make modifications on the existing tests?14:22
hroujrichli, I like the idea : )  I just mean if we use the same key / iv each time, then we're guaranteed if it passes once it will always pass.14:22
*** lcurtis has joined #openstack-swift14:23
acolesjrichli: you see any problem with what hrou just said cos i like the sound of that14:23
mgriffinacoles: cool i found this https://community.rackspace.com/products/f/25/t/64414:23
acolesie use a fixed key/iv for each unit test14:23
jrichlihrou acoles: sounds good.14:24
openstackgerritZhao Lei proposed openstack/swift: Fix some spelling typo in comments  https://review.openstack.org/21045314:24
acolesjrichli: mahatic so that swhat i had been assuming - unit test knows the key/iv so can calculate the expected ciphertext and therefore assert that object body is as it should be14:26
acolesmgriffin: cool, hurricanerix can probably help you better with rax specific client questions but hopefully you are sorted now14:27
acolesjrichli: should i bring a raincoat?14:28
jrichlihrou acoles mahatic: looking back at it, we do use the same key/iv each time already14:28
mahaticjroll: yeah, in crypto.py tesst right?14:28
jrollhi :)14:28
jrichliacoles: I will check.  we haven't gotten rain since june, i think14:28
mahaticjrichli:14:28
mahaticjroll: sorry :)14:29
mgriffinacoles: thanks again.14:29
jroll:P no worries14:29
mahaticjrichli: no, in test_Encrypter as well. sorry14:29
acolesjrichli: actually that was tongue in cheek - i assumed hot and dry ;)14:29
acolesmgriffin: np14:30
wbhuberacoles: ;-)  bring a sweater.14:30
jrichlihrou acoles mahatic: oh, nm.  i meant right now in the tests, but we all knew that.  yes, using the real crypto there will need to be a way to control the key/iv form the test14:31
jrichliwithout using the current mock14:31
hroujrichli, yea sorry that's what I meant ; ) i.e. when using the real crypto, that's what advantage of mocking it entirely.14:31
acoleswbhuber: :)14:32
hrouacoles, we had someone from Austin here in Toronto and I was mentioning it was one of the hottest days in the entire summer and they just laughed and walked away, so yea bring a sweater : )14:33
jrichliacoles: asking if its gonna rain is a valid question, actually.  we do sometimes get rain in the summer.  but I dont see it in the forecast.14:33
acolesjrichli: thx14:36
acolesjrichli: mahatic so looks like the tests do control the key via FakeAppThatSetsKeys14:37
mahaticacoles: yup14:38
acolesbut the IV is set by crypto14:38
mahaticright14:38
* acoles keeps forgetting the IV!14:38
acolesso is there something in crypto.py we can mock to return a fixed IV maybe14:38
acoles?14:38
mahaticI think we can mock "create_iv" in crypto14:39
hrouyep, all it does is return os.urandom(16)14:40
*** mgriffin has left #openstack-swift14:42
*** breitz has joined #openstack-swift14:42
mahaticacoles: ^14:45
acolesmahatic: agree14:47
mahaticI still have a question though, this goes into the existing tests or not? :)14:48
mahaticacoles: jrichli ^14:50
*** ttrumm has joined #openstack-swift14:50
notmynamegood morning14:57
*** geaaru has joined #openstack-swift14:59
*** minwoob has joined #openstack-swift15:01
notmynamejordanP: that's really cool about twitch.tv using swift!15:06
notmynameactually, that's extra cool since they are owned by amazon ;-)15:07
jordanPnotmyname, lol I forgot about that !15:07
*** ttrumm has quit IRC15:10
notmynamejordanP: so I'm trying to find that myself15:12
jordanPnotmyname, I've checked only replay/vods, I don"t know about live streams15:13
notmynameah, ok. live stream doesn't seem to15:14
jrichlimahatic: Start with whatever makes the most sense to you now.  I dont have a strong opinion at this point.  As you get more into implementation, we might want to reassess that decision15:14
jrichlifor example, if you start with a new file, you might feel like there is a lot of copy/paste or duplicated code15:15
jrichliif that happens and you feel like you could integrate with the existing one at that point, then great15:15
notmynamejordanP: yup. found one15:16
jrichlimahatic: basically, its hard for me to make that call without being the one in all the details. :-)15:17
jordanPi wonder which feature of Swift they are using15:17
mahaticjrichli: yes :) got that, lemme take a back up of test_encrypter and start off on that one (maybe on a PUT initially and go on from there), will see what works out15:21
jrichlimahatic: sounds good.  thanks!15:22
notmynamelooks like they are using varnish (which makes sense, too) and the /info enpoint is returning an error from varnish. looks like they are hiding that endpoint15:26
notmynamewhich also makes sense for their use case15:26
notmynamejordanP: I'm guessing the swift feature they are using is "large-scale, cheap storage that can handle a *lot* of concurrent connections" ;-)15:28
*** ttrumm has joined #openstack-swift15:28
*** ttrumm has quit IRC15:29
jordanPnotmyname, possibly :)15:31
*** wbhuber_ has joined #openstack-swift15:36
*** wbhuber has quit IRC15:36
*** chsc has joined #openstack-swift15:38
*** logan2 has quit IRC15:38
*** logan2 has joined #openstack-swift15:41
openstackgerritAlistair Coles proposed openstack/swift: Replace assertTrue(not <condition>) with assertFalse(<condition>)  https://review.openstack.org/20645415:42
*** gyee has joined #openstack-swift15:42
*** nadeem has joined #openstack-swift15:51
*** nadeem has quit IRC15:53
*** nadeem has joined #openstack-swift15:53
*** mahatic has quit IRC15:57
*** cazino has left #openstack-swift16:01
*** jordanP has quit IRC16:08
*** wbhuber_ is now known as wbhuber16:09
*** rledisez has quit IRC16:16
*** petertr7 is now known as petertr7_away16:21
openstackgerritHamdi Roumani proposed openstack/swift: Doc instructions for post rebase steps  https://review.openstack.org/20866916:26
*** jistr has quit IRC16:28
*** janonymous_ has joined #openstack-swift16:29
janonymous_Hi Guys, I think it would be helpful if someone could review these patches, so that new one could be submitted once all are done: https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:master+topic:py3,n,z16:30
janonymous_Thanks in advance... :)16:30
*** bapalm has quit IRC16:31
*** lyrrad has joined #openstack-swift16:37
*** nadeem has quit IRC16:40
*** swat30 has quit IRC16:43
*** thurloat_isgone has quit IRC16:43
*** geaaru has quit IRC16:45
*** marzif has quit IRC16:45
openstackgerritjanonymous proposed openstack/swift: Use six.moves.map/filter/zip inplace of itertools.izip/imap/ifilter  https://review.openstack.org/20341416:45
*** thurloat_isgone has joined #openstack-swift16:46
*** thurloat_isgone is now known as thurloat16:46
*** swat30 has joined #openstack-swift16:47
*** joeljwright has quit IRC16:59
*** nadeem has joined #openstack-swift17:02
janonymous_So it's friday night in most of the places ... time to chill :)17:12
*** geaaru has joined #openstack-swift17:12
*** mfalatic_ has quit IRC17:16
*** nadeem has quit IRC17:18
*** lyrrad has quit IRC17:22
*** lyrrad has joined #openstack-swift17:26
lcurtisgood day swiftstackians...noticing that rsync throughput seems to be slow in general...is there some kind of throttling or other xinetd setting i should be checking?17:27
*** nadeem has joined #openstack-swift17:27
lcurtislike 40-80mbps on 1gbps link, known wire speed possible on link17:27
openstackgerritMerged openstack/swift: Fix a spelling typo in comment  https://review.openstack.org/21044817:28
openstackgerritEran Rom proposed openstack/swift: Container-Sync to iterate only over synced containers  https://review.openstack.org/20580317:29
openstackgerritMerged openstack/swift: Fix some spelling typo in comments  https://review.openstack.org/21045317:30
*** petertr7_away is now known as petertr717:39
*** geaaru has quit IRC18:00
*** nadeem has quit IRC18:03
*** gyee has quit IRC18:05
openstackgerritMinwoo Bae proposed openstack/swift: Error handling of DiskFileNoSpace()  https://review.openstack.org/19890918:05
*** nadeem has joined #openstack-swift18:09
*** nadeem has quit IRC18:18
*** lyrrad has quit IRC18:20
*** lyrrad has joined #openstack-swift18:21
*** breitz has quit IRC18:23
*** lyrrad has quit IRC18:23
openstackgerritThiago da Silva proposed openstack/swift: versioned writes middleware  https://review.openstack.org/13434718:23
*** eandersson has quit IRC18:24
*** lyrrad has joined #openstack-swift18:24
*** bapalm has joined #openstack-swift18:26
*** lyrrad has quit IRC18:27
*** acoles is now known as acoles_18:28
*** lyrrad has joined #openstack-swift18:29
*** hrou has quit IRC18:36
*** bapalm has quit IRC18:38
*** bapalm has joined #openstack-swift18:48
*** bapalm has quit IRC18:48
*** delattec has quit IRC18:50
*** cdelatte has quit IRC18:50
*** gyee has joined #openstack-swift18:52
*** hrou has joined #openstack-swift18:54
*** ttrumm has joined #openstack-swift18:54
*** robefran has quit IRC18:56
*** jbonjean has joined #openstack-swift19:04
*** ttrumm has quit IRC19:13
*** mfalatic has joined #openstack-swift19:19
*** tongli has quit IRC19:28
claygsomewhere right around here I'm feeling very related to the conversation -> https://youtu.be/wf-BqAjZb8M?t=61119:28
portanteclayg: speaking with emacs, yes!19:32
openstackgerritMerged openstack/swift: Replace assertTrue(not <condition>) with assertFalse(<condition>)  https://review.openstack.org/20645419:32
claygportante: I have some vms building in the background and kicked this off for the background19:32
claygportante: when he said "you step into this code I've been maintaing for 15 years and pep8ify it for me" I thought a few others might also get some smiles on Friday19:33
portanteyes!19:33
portanteclayg: first time I have heard him speak, he is pretty good (raymond hettinger)19:35
*** amoturi has quit IRC19:40
*** amoturi has joined #openstack-swift19:40
*** MVenesio has quit IRC19:45
*** chsc has quit IRC19:47
tdasilvaJSÓN19:50
tdasilvajay-son19:50
tdasilvalol19:50
portantewax on wax off19:50
*** kevinc_ has joined #openstack-swift19:59
notmyname"pep8 unto thyself, not unto others" :-)19:59
*** petertr7 is now known as petertr7_away20:02
*** nadeem has joined #openstack-swift20:05
*** chlong has quit IRC20:06
portantenotmyname: yes!20:07
portantegreat line20:07
*** petertr7_away is now known as petertr720:07
kevinc_Anyone have experience with Seagate Kinetic drives? I setup our test cluster to read and write to the Kinetic drives, but the data doesn't get replicated when I simulate a failure20:09
MooingLemurtdasilva: I like that. JSÓN.20:12
tdasilvaMooingLemur: did I get the accent right?20:15
*** silor has quit IRC20:15
minwoobtdasilva: In https://review.openstack.org/#/c/198909/ were you actually looking for the variable to be prepended with '_' ? In that case, all the other msg variables might need that change.20:20
minwoobtdasilva: Otherwise, feel free to take a look at the most updated version.20:21
*** jlhinson_ has joined #openstack-swift20:28
*** jlhinson has quit IRC20:28
tdasilvaminwoob: I think that looks good, wasn't looking for you to change every single one20:28
*** jlhinson_ has quit IRC20:32
minwoobtdasilva: The other ones match up to that one, just making sure that was the change that was needed though, to eliminate some alternative modifications that could have been made.20:33
*** zul has joined #openstack-swift20:36
*** dustins has quit IRC20:41
*** jbonjean has left #openstack-swift20:45
claygkevinc_: you have to run the kinetic-swift-replciator daemon - it's seperate from the swift-object-replicator (swift-init stuff)20:47
*** sweeper has quit IRC20:52
kevinc_clayg: ok, great! That was what I was looking for. I assume i need to do the same for the auditor and the updater.20:53
kevinc_looks like the objects are replicating now20:53
kevinc_awesome20:53
claygneat!20:54
*** logan2 has quit IRC20:56
*** ccavanna has quit IRC21:07
*** ccavanna has joined #openstack-swift21:08
*** ccavanna has quit IRC21:09
*** ccavanna has joined #openstack-swift21:09
*** gyee has quit IRC21:18
openstackgerritEran Rom proposed openstack/swift: Container-Sync to iterate only over synced containers  https://review.openstack.org/20580321:19
openstackgerritEran Rom proposed openstack/swift: Add container sync probe test to SAIO default set  https://review.openstack.org/20580421:19
*** _hrou_ has joined #openstack-swift21:19
*** petertr7 is now known as petertr7_away21:20
*** hrou has quit IRC21:20
*** tsg has joined #openstack-swift21:21
*** minwoob has quit IRC21:24
*** _hrou_ has quit IRC21:24
*** jrichli has quit IRC21:27
*** kevinc_ has quit IRC21:28
openstackgerritBill Huber proposed openstack/swift: Add unit test for container_update for unmounted device  https://review.openstack.org/21061021:35
*** nadeem has quit IRC21:38
*** nadeem has joined #openstack-swift21:39
*** logan2 has joined #openstack-swift21:39
*** zaitcev has joined #openstack-swift21:47
*** ChanServ sets mode: +v zaitcev21:47
*** wbhuber has quit IRC21:52
mattoliverauMorning all, at airport, the waiting begins22:10
*** jamespage_ has joined #openstack-swift22:19
*** otoolee has quit IRC22:20
*** otoolee has joined #openstack-swift22:23
*** portante has quit IRC22:32
*** portante has joined #openstack-swift22:34
*** lyrrad_ has joined #openstack-swift22:39
*** lcurtis has quit IRC22:46
*** lyrrad has quit IRC22:49
*** zul has quit IRC22:49
*** nexusz99 has quit IRC22:49
*** changbl has quit IRC22:49
*** lyrrad_ is now known as lyrrad22:49
*** nexusz99 has joined #openstack-swift22:52
*** zul has joined #openstack-swift22:54
*** changbl has joined #openstack-swift22:56
*** cazino has joined #openstack-swift22:57
*** wbhuber has joined #openstack-swift23:11
*** logan2 has quit IRC23:15
*** nexusz99 has quit IRC23:35
*** nadeem has quit IRC23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!