Wednesday, 2014-02-26

*** cpallares has quit IRC00:12
*** cpallares has joined #openstack-marconi00:16
*** amitgandhi has joined #openstack-marconi00:35
*** kgriffs is now known as kgriffs_afk00:38
*** amitgandhi has quit IRC00:50
*** cpallares has quit IRC00:58
*** amitgandhi has joined #openstack-marconi01:15
*** vkmc has quit IRC01:17
*** reed has quit IRC01:19
*** fifieldt has joined #openstack-marconi01:21
*** jergerber has quit IRC01:35
*** amitgandhi has quit IRC01:40
*** nosnos has joined #openstack-marconi01:41
*** rongze has joined #openstack-marconi02:00
*** amitgandhi has joined #openstack-marconi02:05
*** ayoung has quit IRC02:12
*** amitgandhi has quit IRC02:26
*** flwang has quit IRC02:46
*** amitgandhi has joined #openstack-marconi02:53
*** ayoung has joined #openstack-marconi02:56
*** flwang has joined #openstack-marconi03:17
*** amitgandhi has quit IRC03:24
*** kgriffs_afk is now known as kgriffs03:27
*** amitgandhi has joined #openstack-marconi03:49
*** amitgandhi has quit IRC03:59
openstackgerritKurt Griffiths proposed a change to openstack/marconi: test(wsgi): Refactor tests in preparation for API v1.1  https://review.openstack.org/7604004:36
openstackgerritKurt Griffiths proposed a change to openstack/marconi: feat(api): API v1.1 bulk delete change  https://review.openstack.org/7609304:36
*** rongze has quit IRC04:44
*** rongze has joined #openstack-marconi04:44
openstackgerritKurt Griffiths proposed a change to openstack/marconi: feat(api): API v1.1 bulk delete change  https://review.openstack.org/7609304:47
*** rongze has quit IRC04:49
*** ayoung is now known as ayoung_zZzZZzz_04:55
*** amitgandhi has joined #openstack-marconi04:55
*** amitgandhi has quit IRC05:00
*** chandan_kumar has joined #openstack-marconi05:07
*** rongze has joined #openstack-marconi05:15
*** rongze_ has joined #openstack-marconi05:16
*** rongze has quit IRC05:17
*** rongze_ has quit IRC05:21
*** amitgandhi has joined #openstack-marconi05:56
*** amitgandhi has quit IRC06:00
*** rongze has joined #openstack-marconi06:15
*** rongze has quit IRC06:19
*** rongze has joined #openstack-marconi06:21
*** rwsu has quit IRC06:27
*** kgriffs is now known as kgriffs_afk06:28
*** kgriffs_afk is now known as kgriffs06:30
*** flaper87|afk is now known as flaper8706:48
*** saju_m has joined #openstack-marconi06:50
*** amitgandhi has joined #openstack-marconi06:57
*** amitgandhi has quit IRC07:01
*** haomaiwang has quit IRC07:16
*** haomaiwang has joined #openstack-marconi07:16
*** haomaiw__ has joined #openstack-marconi07:30
*** haomaiwang has quit IRC07:33
*** kgriffs is now known as kgriffs_afk07:35
*** amitgandhi has joined #openstack-marconi07:58
*** amitgandhi has quit IRC08:03
flwangflaper87: ping08:03
flaper87flwang: pong08:03
flwangflaper87: may I ask some questions about storage backend?08:04
flaper87flwang: sure you can08:04
flwangflaper87: what's the difference between sqlite driver and the sqlalchemy driver? or in other words, will the sqlite driver be deprecate given the sqlalchemy is merged?08:05
flaper87flwang: I'm working on that as we speak08:05
flaper87flwang: I mean, removing sqlite08:05
flwangflaper87: got it, make more sense for me08:06
flwangflaper87: and another question I asked several days ago, but you didn't answer me08:08
flaper87flwang: didn't I?08:08
flaper87waa?08:08
* flaper87 is a good guy08:08
flwangflaper87: AFAIK, Mongodb supports auto sharding, so why we add the 'manual' sharding? would you mind telling more?08:09
flwangflaper87: maybe you just missed it, no worries :)08:09
flaper87oh wait, don't let the mongodb 'auto-sharding' thing fool you. MongoDB does support auto-sharding but it's not that it'll do everything for you08:09
flaper87you still have to tell it what keys you want to use for sharding08:10
flaper87besides that, we still don't recommend using sharding for marconi because we simply haven't figured a good key to use08:10
flaper87nor we have an environment where this can be tested at a high scale08:10
flaper87flwang: do you happen to have a lab where this can be tested?08:11
flaper87that would be awesome08:11
flwangflaper87: nope :( I'm the ooooooooooooooooly guy working on Marconi from IBM as you saw08:11
flaper87flwang: yeah, let me put this in other words. Do you happen to have a lab that we can steal ? :D08:12
flwangflaper87: and I didn't see anything shown IBM will deliver Marconi with its product yet, at least until now08:12
flaper87:P08:12
flwangflaper87: haha, internally, I'm proposing to prepare  a big scale test environment, but I'm not sure when it will be done.08:13
*** kgriffs_afk is now known as kgriffs08:26
*** rongze has quit IRC08:34
*** kgriffs is now known as kgriffs_afk08:35
*** flwang has quit IRC08:50
*** amitgandhi has joined #openstack-marconi08:58
*** ykaplan has joined #openstack-marconi09:03
*** amitgandhi has quit IRC09:04
*** kgriffs_afk is now known as kgriffs09:26
*** jamieh has joined #openstack-marconi09:34
*** flwang has joined #openstack-marconi09:57
*** amitgandhi has joined #openstack-marconi09:59
*** amitgandhi has quit IRC10:04
*** kgriffs is now known as kgriffs_afk10:04
*** davidhadas has quit IRC10:06
flwangflaper87: ping10:09
flaper87flwang: pong10:10
flwangflaper87: seems we're stilling using sqlite driver as the test storage backend for some cases, though we have sqlalchemy driver now, right?10:11
flwangflaper87: any plan to do the transition?10:12
flaper87FIRE IN THE HOLE10:24
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Enable pragma foreing key for sqlite  https://review.openstack.org/7648010:24
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Fix message's get and deletion  https://review.openstack.org/7648110:24
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Register sqlalchemy endpoints  https://review.openstack.org/7648210:24
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Don't raise `DoesNotExist` on claims deletion  https://review.openstack.org/7648310:24
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Deprecate sqlite in favor of sqlalchemy  https://review.openstack.org/7648410:24
flaper87flwang: sorry, back! I was headsdown ^10:24
flaper87flwang: and ^ is the answer to your question :D10:25
flwangflaper87: wow, so many patches :) I will eat them as my dinner10:25
flaper87flwang: THANKS!10:26
flwangflaper87: sure :)10:31
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Register sqlalchemy endpoints  https://review.openstack.org/7648210:35
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Don't raise `DoesNotExist` on claims deletion  https://review.openstack.org/7648310:35
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Enable pragma foreing key for sqlite  https://review.openstack.org/7648010:35
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Fix message's get and deletion  https://review.openstack.org/7648110:35
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Deprecate sqlite in favor of sqlalchemy  https://review.openstack.org/7648410:35
*** davidhadas has joined #openstack-marconi10:41
*** kgriffs_afk is now known as kgriffs10:56
*** amitgandhi has joined #openstack-marconi11:00
*** amitgandhi has quit IRC11:04
*** kgriffs is now known as kgriffs_afk11:05
*** ykaplan has quit IRC11:22
*** malini_afk is now known as malini11:29
maliniflaper87: PING11:29
maliniDon't hide..I know you are here :D11:30
*** kgriffs_afk is now known as kgriffs11:56
*** amitgandhi has joined #openstack-marconi12:01
*** ykaplan has joined #openstack-marconi12:04
*** amitgandhi has quit IRC12:05
*** kgriffs is now known as kgriffs_afk12:06
*** ykaplan has quit IRC12:10
*** ykaplan has joined #openstack-marconi12:16
flaper87malini: PONG12:20
flaper87I'M HERE!!!12:20
maliniflaper87: wooooooot !!!!!!!12:20
flaper87sooooooooooooooooooooooooooo12:20
flaper87good morning12:20
flaper87how are you doing?12:20
maliniI have been hunting for you long time :D12:21
maliniI am good..How are you?12:21
flaper87I know, I'm so sorry, I had to step away yday12:21
flaper87:P12:21
flaper87I'm doing fine, I just had lunch so I'm definitely better now12:21
malinino..you chose the wrong day to sleep ;-)12:21
flaper87:D12:21
malininp*12:21
maliniwas kidding..12:21
flaper87hahah12:21
flaper87:D12:21
flaper87so, the issue12:21
maliniBut ur IRC message left me in suspense12:21
malinidid you ever get to the screen issue?12:22
maliniI have nightmares abt being ready to graduate except for tempest :D12:22
flaper87hahah hahaha12:22
flaper87weeeeeeeeell, tempest is the only bit missing12:23
flaper87huahuahuahuahua12:23
* flaper87 is messing with malini12:23
flaper87:D12:23
flaper87malini: https://github.com/openstack-dev/devstack/blob/master/lib/marconi#L3712:23
malinitht was not very kind12:23
flaper87that should be /var/log/marconi12:23
maliniflaper87: why wud that matter?12:23
flaper87because otherwise marconi will try to write to a log_file that doesn't exist12:23
flaper87and it will fail12:24
maliniflaper87: I think I tried manually creating one with tht name..but didnt help12:24
flaper87I mean, Marconi's default log file is /var/log/marconi/queues.log12:24
flaper87That dir needs to exist and be writable12:24
maliniflaper87: cud u get it running with tht change?12:25
flaper87yup12:25
* flaper87 tries again12:25
maliniI will try tht as well12:28
maliniI really really really hope you are right..12:28
*** vkmc has joined #openstack-marconi12:32
*** vkmc has quit IRC12:32
*** vkmc has joined #openstack-marconi12:32
openstackgerritFlavio Percoco proposed a change to openstack/marconi: Deprecate sqlite in favor of sqlalchemy  https://review.openstack.org/7648412:39
maliniflaper87: tht didnt work for me :(12:39
maliniBut I see some weirdness in the mongodb log12:39
maliniflaper87: http://paste.openstack.org/show/69834/12:40
maliniwe are cleaning the db & stuff after the restart12:40
malinimaybe marconi is shutting down for some reason?12:40
flaper87mmhh12:42
flaper87malini: I'm setting up a new vm12:42
flaper87I'll try again12:42
malinithanks flaper87 !12:44
maliniits weird why it wud be different with the screen12:44
*** tedross_ has joined #openstack-marconi12:49
*** kgriffs_afk is now known as kgriffs12:50
flaper87malini: erm, mind sharing your localrc again?12:57
flaper87malini: found it13:01
*** amitgandhi has joined #openstack-marconi13:01
flaper87malini: running13:02
maliniflaper87: sure13:06
*** amitgandhi has quit IRC13:06
maliniflaper87: http://paste.openstack.org/show/69840/13:07
maliniflaper87: thts weird..I still have no luck in ubuntu with the log destination update..13:08
flaper87malini: don't waste much time there, I'll try again and let you know13:10
flaper87https://wiki.openstack.org/wiki/Marconi/DevTips13:12
maliniI need to go offline now to drive to the office..13:13
maliniwill be back in 2 hours13:13
*** kgriffs is now known as kgriffs_afk13:30
*** ykaplan has quit IRC13:37
*** jamieh has quit IRC13:37
*** jamieh has joined #openstack-marconi13:38
*** nosnos has quit IRC13:41
*** sriram has joined #openstack-marconi13:50
*** malini is now known as malini_afk13:54
*** sriram has quit IRC13:54
*** sriram has joined #openstack-marconi13:55
*** jamieh has quit IRC14:14
*** amitgandhi has joined #openstack-marconi14:15
*** amitgandhi has quit IRC14:16
*** amitgandhi has joined #openstack-marconi14:16
*** rossk has joined #openstack-marconi14:16
*** mpanetta has joined #openstack-marconi14:19
*** jamieh has joined #openstack-marconi14:20
*** ayoung_zZzZZzz_ is now known as ayoung14:20
openstackgerritShawnyu proposed a change to openstack/marconi: Convention Fixes. doc(queues.storage.mongodb.catalogue): Add docstrings  https://review.openstack.org/7480714:21
*** kgriffs_afk is now known as kgriffs14:22
*** mpanetta has quit IRC14:26
*** mpanetta has joined #openstack-marconi14:26
*** jamieh has quit IRC14:28
*** jamieh has joined #openstack-marconi14:30
*** jcru has joined #openstack-marconi14:32
*** kgriffs is now known as kgriffs_afk14:32
*** ykaplan has joined #openstack-marconi14:34
*** jcru has quit IRC14:34
*** jcru has joined #openstack-marconi14:35
*** cpallares has joined #openstack-marconi14:38
*** ametts has joined #openstack-marconi14:43
*** kgriffs_afk is now known as kgriffs14:44
*** rossk has quit IRC14:46
*** kgriffs is now known as kgriffs_afk14:53
*** rossk has joined #openstack-marconi14:57
*** malini_afk is now known as malini14:57
*** saju_m has quit IRC15:02
*** sballe_ has joined #openstack-marconi15:07
*** sballe has quit IRC15:10
*** tedross_ is now known as tedross15:12
openstackgerritFei Long Wang proposed a change to openstack/marconi: feat(v1.1): Implement new /health endpoint for v1.1  https://review.openstack.org/7655515:16
*** rwsu has joined #openstack-marconi15:18
*** jcru has quit IRC15:18
*** jcru has joined #openstack-marconi15:18
*** jamieh has quit IRC15:19
flwangflaper87: ping15:20
*** kgriffs_afk is now known as kgriffs15:23
*** jcru has quit IRC15:23
flwangkgriffs: ping15:25
*** jamieh has joined #openstack-marconi15:25
*** jcru has joined #openstack-marconi15:47
*** jcru has quit IRC15:49
*** jcru has joined #openstack-marconi15:50
*** chandan_kumar has quit IRC15:52
*** ayoung is now known as ayoung-mtg15:55
*** chandan_kumar has joined #openstack-marconi15:56
maliniflaper87: I am still messing with the logs in devstack16:06
malinimight need ur help again, if I cant nail it down16:06
*** ayoung-mtg is now known as ayoung16:11
mpanettaflwang: ping16:14
flwangmpanetta: pong :)16:14
mpanettaHey, I just saw your note.16:14
flwangmpanetta: you mean the patch?16:14
flaper87malini: sounds good, let me know16:14
flaper87flwang: pong16:14
mpanettaI did try that patch yesterday, and it didn't seem to help.16:14
mpanettaYeah16:14
mpanettaThe one the kurt put in16:15
mpanettaI commented on it too.16:15
flwangmpanetta: ah, got16:15
mpanettaThat was why I opened a new bug, I thought maybe that patch was for something else.16:15
flwangflaper87: may i get your attention on this https://review.openstack.org/#/c/76555/ ? just wanna get some comments about the overall shape16:15
flwangmpanetta: i see. I think you should let kgriffs know that. maybe it can be fixed by the way16:18
mpanettaOk, thank you flwang16:19
flwangmpanetta: sure, I wish i can help more :)16:19
mpanettakgriffs: Ping!!16:19
kgriffsmpanetta: pong16:19
mpanettaDid you see the above?16:19
mpanettaARe you watching the ZeroVM talk? heh16:19
kgriffsyeah, 30% watching16:21
kgriffsmpanetta: I saw the above, but not sure I follow16:22
flwangkgriffs: ping16:22
mpanettakgriffs: Ahh, the patch you supplied here: https://review.openstack.org/#/c/75165/ Does not keep this: https://bugs.launchpad.net/marconi/+bug/1284817 from happening.16:23
flwangkgriffs: may i get your attention on this https://review.openstack.org/#/c/76555/ ? just wanna get some comments about the overall shape16:23
kgriffsmpanetta: ah, so I'm not suprised16:23
kgriffsthe autoreconnect is when you have a sort of graceful step down of the master or when the master crashes, but secondaries are still available16:23
mpanettaAh ok16:24
mpanettaSo then I ws correct in creating a new bug...16:24
kgriffsso that means the bug you have there must be related to network partitioning or when the entire cluster crashes (latter is highly unlikely tho, heh)16:24
kgriffsmpanetta: yes16:24
kgriffsgood bug16:24
mpanettaIn our case we lost network connectivity, the cluster was ok, but we could not connect to it for awhile.16:25
mpanettaAnd when connectivity came back, marconi did not.16:25
openstackgerritKurt Griffiths proposed a change to openstack/marconi: test(wsgi): Refactor tests in preparation for API v1.1  https://review.openstack.org/7604016:29
openstackgerritKurt Griffiths proposed a change to openstack/marconi: feat(api): API v1.1 bulk delete change  https://review.openstack.org/7609316:29
kgriffsflaper87: hey, you've been busy!16:30
kgriffsNice to see those patches16:30
kgriffsflwang: I addressed your comments in the latest patchset...16:30
kgriffsthanks for catching those!16:30
maliniflaper87: I updated the log files path in devstack & can see marconi start up in the log o/p16:31
maliniBut something is still going wrong16:31
maliniI dont see a good example in any of the other projects16:31
maliniI am trying this out on a brand new server16:31
flwangkgriffs: it would be appreciated if you can take a look at my patch against /health v1.116:32
kgriffsflwang: sure. I'm at a conference, but am trying to sneak in some marconi love when I can during the day16:33
flwangkgriffs: cool, thanks16:34
*** ykaplan has quit IRC16:37
cpallareskgriffs: I'm doing a code review? :D16:39
* cpallares has never done a code review before16:39
flaper87cpallares: you have to review people's code16:40
flaper87expecially kgriffs16:41
flaper87I've heard that d00d likes to break things16:41
flaper87mpanetta: ^^^ right ?16:41
cpallareshaha16:41
amettsflaper87:  Oh, the stories I could tell...16:41
mpanettakgriffs breaks all the things flaper8716:41
maliniflaper87: ping me when you have a free minute16:42
flaper87kgriffs: yeah, that was simpler than I expected... I swear I didn't use `sed` (chuckles)16:42
flaper87ametts: oh, do tell! :D :D :D16:42
*** jcru has quit IRC16:42
flaper87malini: mmh, I still pay for minutes, are they being given away for free now?16:43
flaper87:(16:43
flaper87malini: KIDDIIIIIIIIIIIING16:43
flaper87malini: tell me16:43
maliniflaper87: The other projects dont specify a log file16:43
maliniSo I am trying thta route & all logs by default go to /home/stack/logs16:44
*** kgriffs is now known as kgriffs_afk16:44
maliniBut tht still doesnt fix our issue16:44
maliniwhy in the world do we log ^o)16:45
malini& why is it different with screen on & off :-S16:46
malini<end of whining/>16:46
cpallareslol malini16:46
*** kgriffs_afk is now known as kgriffs16:47
jay-atlHey Guys... Ok, I am pulling the base url from the service catalog and I've noticed a problem...16:56
jay-atlThe URL in the service catalog does not include the version number /v1/16:56
jay-atlSo I would have to append that in my calls... Which is not what the other products do...16:56
jay-atlThey include the version in the url in the service catalog.16:56
flwangflaper87: do you think we should let Marconi be monitored by Ceilometer?16:57
amettsjay-atl: This issue seems really familiar.16:58
amettsamitgandhi ^16:58
flwangflaper87: i'm going to create a bp, so want to get some insights about this16:58
jay-atlExample URL for cinder from the service catalog: "publicURL": "http://XXX.XX.XXX.XXX:8774/v3"17:01
maliniametts: we had the same question pop-up in RAX service catalog17:02
jay-atlMarconi URL from service catalog: "publicURL": "http://XXX.XX.XXX.XXX:8888"17:02
jay-atlI use a standard call in Horizon: base.url_for(request, 'queuing') which returns this url, and it won't work because the version is missing.17:03
amettsmalini: I knew this sounded familiar.  What was the resolution on the RAX side?  Do we need to submit a pull request to keystone?17:03
kgriffsjay-atl: this is for the stock Keystone catalog?17:03
jay-atlYes17:03
ametts(and by "we", I mean jay-atl) :)17:04
maliniits easy to update this in devstackā€¦but tht will screw up the entire base_url + href17:04
amettsjay-atl is always causing trouble.17:04
jay-atl:-)17:05
kgriffsThe issue we had at Rackspace was he catalog was including the project (AKA tenant) ID in the URI17:05
kgriffsit didn't have anything to do with the version in the URI iirc17:05
amettsoic17:06
*** saju_m has joined #openstack-marconi17:06
kgriffsflaper87, flwang: do you know of any OS projects that allow you to change the logging level without restarting the app17:07
mpanettakgriffs: Marconi needs this. ;)17:08
kgriffsmpanetta: yes!17:08
kgriffsmpanetta: can you submit a blueprint?17:08
* mpanetta knows what talk kgriffs is at.17:08
kgriffsLOL17:09
* mpanetta is watching it remotely :)17:09
mpanettakgriffs: I donno.  What is involved?17:09
mpanettaI've never done a blueprint :)17:09
kgriffsnot mutch, it's easy17:09
flwangkgriffs: http://stackoverflow.com/questions/19617355/dynamically-changing-log-level-in-python-without-restarting-the-application17:09
kgriffsflwang: gtk17:10
kgriffsmpanetta: you can reference that link in the bp17:10
flwangkgriffs: i'm not aware if there is any OS project using this, but I do love it because deployer/administrator can be benefited by this dramatically17:10
kgriffs+117:10
kgriffsmpanetta: just go here17:10
kgriffsand on the top-right, just click "register bp"17:11
kgriffsdon't worry about a link "to the full spec" for now, just write a short description and possibly include that SO link17:11
mpanettawhere is here?17:12
flwangkgriffs: glad to see if we can implement it in Marconi17:12
kgriffsmpanetta: also, feel free to propose any other operational maturity bps that are on your mind17:12
kgriffs;)17:12
kgriffsmpanetta: oops17:12
kgriffsforgot to paste17:12
kgriffshttps://blueprints.launchpad.net/marconi17:12
mpanettahaha17:12
flwangmpanetta: you know remember this link ;)17:12
flwangknow/should17:12
kgriffs+117:12
mpanettaOk :)17:13
mpanettaI will bookmark it17:13
flwangmpanetta: i'd like to see more valuable/operational level bp from you :)17:13
mpanettaOk :)17:14
flwangmpanetta: and I'd happy to work on those17:14
mpanettakgriffs: What do you mean by SO link?17:14
mpanettaOh the stackoverflow link17:15
jay-atlSo.... Who needs to do what to fix the service catalogs URL...17:17
flwangI'm going to drop, have a nice day, guys17:17
mpanettaHave a good day flwang17:17
flwangZzzz...17:17
mpanettaThanks for the help!17:17
flwangmy time is 1:17AM now :0)17:18
mpanettaWow ok, so have a good night :)17:18
flwangmpanetta: anytime17:18
flwangmpanetta: later :)17:18
kgriffs<random>https://blueprints.launchpad.net/marconi/+spec/structured-logging</random>17:21
kgriffsI just updated that with a link to the solum tracing patch that just landed for reference17:21
kgriffsI think this would be great thing to explore during Juno.17:21
kgriffsperhaps cinder would also find value in this17:22
kgriffsI'd like to get a few projects playing with this17:22
kgriffsand if it has widespread value, work on getting it into oslo17:22
mpanettaI wonder if there is any value to enforcing not running marconi as root...17:22
mpanettakgriffs: Better lgging is very good,  I do detest random stacktraces in logs... :P17:23
kgriffsmpanetta: mmm, not sure about that. We've always said that the operator brings their own WSGI server17:23
kgriffsI think the operator should just know to not run their server as root. :p17:23
mpanettaStack traces to me mean that you forgot to handle an exception somewhere...17:23
kgriffslol17:23
mpanettakgriffs: I agree :)17:23
kgriffsmpanetta: I also have a bp for error codes and stuff, but I think that is going to slip to Juno17:24
mpanettaOk17:24
mpanettaAt least it is there!17:24
kgriffsyep. I think operational maturity would be a great theme for Juno17:24
kgriffsflwang, flaper87: ^^^17:24
kgriffsbbl17:26
mpanettakgriffs: https://blueprints.launchpad.net/marconi/+spec/dynamic-log-level17:31
flaper87kgriffs: our set of patches will conflict for sure. Who of us should suffer the rebase process this time?17:33
flaper87I think you suffered it last time so, I'm happy to get it17:33
*** kgriffs is now known as kgriffs_afk17:36
*** jcru has joined #openstack-marconi17:59
*** reed has joined #openstack-marconi18:22
*** reed has quit IRC18:23
maliniflaper87: mpanetta found the issue!!!!!!18:25
mpanettaYeah, I remembered some oddness with marconi-server from when I first started work here.18:25
mpanettamarconi-server CLI process requires that it be run in a terminal.18:26
mpanettaWhich is why it works with screen and doe not work without it.18:26
flaper87malini: what is it?18:28
mpanettamarconi-server hangs when run without a screen18:30
mpanettaWell, crashes, or hangs, depending...18:31
mpanettaflaper87: ^^18:31
mpanettaWe got it to work with gunicorn, but from what I understand adding requirements to devstack is a bit of a chore18:32
*** saju_m has quit IRC18:35
*** reed has joined #openstack-marconi18:35
*** ykaplan has joined #openstack-marconi18:36
*** davidhadas has quit IRC18:44
*** chandan_kumar has quit IRC18:46
*** kgriffs_afk is now known as kgriffs19:07
kgriffsmpanetta: d00000d19:09
kgriffslet's fix that!19:09
kgriffsanyone already on it?19:09
mpanettaMe19:09
mpanettaI fixed it by commenting out one line of code...19:09
kgriffshehe19:09
kgriffswhich line?19:09
* kgriffs has a suspicioun19:09
mpanettahttps://github.com/openstack/marconi/blob/master/marconi/cmd/server.py#L2219:10
mpanettaThat basically gets rid of all the odd terminal handling things that are not really needed...19:10
mpanettaProbably not the best fix, but t is the simplest :P19:11
mpanettaAs far as I can tell, all that decorator does is wrap the run function with some stuff to handle termios19:12
mpanettaWhich we don't want to do, when there is no terminal...19:12
kgriffsheh19:13
kgriffsthat was my suspicion19:13
kgriffsso...19:13
mpanettaSo...?19:14
mpanettaheh19:14
mpanettaWhy was that stuff added?19:14
mpanettaDo you know?19:14
mpanettakgriffs: ?  You disappeared! :P19:16
kgriffsso let's break it down19:17
*** ayoung has quit IRC19:17
kgriffs_...19:18
mpanettaMorse code? :P19:18
kgriffsheh19:18
kgriffssorry, lots of distractions here19:18
kgriffsso19:18
mpanettaI can imagine19:18
kgriffsthe wrapper does the following19:18
kgriffs1. echo toggle so when you type it doesn't put junk on the screen - nice to have, not necessary19:19
kgriffsbut19:19
kgriffsit is already checking os.isatty19:19
kgriffsand skipping that19:19
kgriffsso, it shouldn't be the problem19:19
kgriffsyou can test that by going into marconi/common/cli.py19:19
*** ykaplan has quit IRC19:20
kgriffsand just inserting a return statement at the top of _enable_echo19:20
kgriffs(so it never does anything19:20
kgriffs2. logging setup19:20
*** davidhadas has joined #openstack-marconi19:21
kgriffs3. unhandled exception logging and "graceful" exit19:21
kgriffsso, I suspect #119:21
kgriffsperhaps L48 isn't working as expected?19:21
mpanettaHmm, possibly.19:22
mpanettaI can test19:22
mpanettaI will try your suggestions and see what I can do.  Then figure out how to submit a PR heh19:22
maliniI hope we get this fixed soooob19:22
malinisoooooon*19:22
mpanettaYes!19:23
maliniWith tht my tempest patch shud be good to go19:23
maliniI have been pulling out my hair for two weeks coz of this19:23
malinimpanetta gets a BIG box of dar chocolate pop-tarts19:23
malinidark19:23
mpanettawoot!19:23
srirammpanetta: AWESOME :)19:23
mpanettabrb19:23
malinisriram: I dont know if he likes to share ;)19:24
srirammalini: hah19:24
*** jamieh has quit IRC19:37
*** kgriffs is now known as kgriffs_afk19:41
*** ametts_ has joined #openstack-marconi19:42
*** reed has quit IRC19:47
*** ametts has quit IRC19:47
*** reed has joined #openstack-marconi19:53
*** sballe has joined #openstack-marconi19:56
*** jraim has quit IRC19:57
*** jraim has joined #openstack-marconi19:58
*** jraim has quit IRC19:59
*** sballe_ has quit IRC20:01
*** tedross has quit IRC20:01
*** tedross has joined #openstack-marconi20:05
*** jraim has joined #openstack-marconi20:05
*** jraim has quit IRC20:05
*** jraim has joined #openstack-marconi20:05
*** tedross has quit IRC20:05
*** tedross has joined #openstack-marconi20:05
mpanettakgriffs_afk: Yes, the isatty call is not doing what it should20:10
mpanettaYeah, shite be broke...20:12
*** jamieh has joined #openstack-marconi20:13
*** flaper87 has quit IRC20:20
*** flaper87 has joined #openstack-marconi20:20
*** orwell.freenode.net sets mode: +o flaper8720:20
*** jcru has quit IRC20:26
*** jcru has joined #openstack-marconi20:26
*** malini is now known as malini_afk20:42
mpanettakgriffs_afk: Ok, fixed the backgrounding issue in marconi cli.py.21:11
*** openstackgerrit has quit IRC21:20
*** openstackgerrit has joined #openstack-marconi21:21
openstackgerritMike Panetta proposed a change to openstack/marconi: Fix marconi-server so it does not stop when backgrounded.  https://review.openstack.org/7665921:22
flaper87mpanetta: is that how other projects do that ?21:23
*** jcru_ has joined #openstack-marconi21:23
*** davidhadas_ has joined #openstack-marconi21:24
mpanettaflaper87: They don't use termios at all from what I can tell.21:24
*** ayoung_ has joined #openstack-marconi21:25
mpanettawe looked at keystone today and they just set up a wsgiref app.21:25
mpanettaI didn't notice any termios stuff in it at all.21:25
mpanettaMy personal choice would be to get rid of it entirely...21:25
mpanettaIt looks like the only reason we use it is to prevent typing on the terminal from mucking up the screen.21:26
mpanettaflaper87: What do you think?21:28
flaper87mpanetta: +1 for getting rid of it21:28
flaper87if that's what is causing this issue, I don't see a good reason to keep it21:29
mpanettaYeah, when you start mucking with the terminal all sorts of things start to come in to play.21:29
flaper87agreed21:30
flaper87and then you start writing php within C# code to make process forks work21:30
flaper87wait, did I just say that?21:30
mpanettahaha21:30
flaper87:P21:30
mpanettalol21:30
mpanettaI wonder what kgriffs_afk thinks.21:31
*** sriram has quit IRC21:31
flaper87mpanetta: get rid of it, he'll comment on the issue21:32
mpanettaOk21:32
flaper87I'll make sure it doesn't get approved until he sees it21:32
mpanettaSweet :)21:32
mpanettaTy21:32
flaper87mpanetta: thank YOU!21:32
flaper87mpanetta: and if something bad happens, it'll be cpallares's fault21:32
mpanettaLOL21:32
cpallaresflaper87, mpanetta: :P21:34
*** jcru has quit IRC21:34
*** davidhadas has quit IRC21:35
*** kgriffs_afk is now known as kgriffs21:37
kgriffsm21:38
flaper87there he is21:39
flaper87kgriffs: wanna review some patches? please, please please21:39
flaper87kgriffs: https://review.openstack.org/#/q/status:open+project:openstack/marconi+branch:master+topic:bp/sql-storage-driver,n,z21:39
kgriffsI was just reviewing the irc log21:39
kgriffsmpanetta: did you find that the  os.isatty(sys.stdin.fileno()) wasn't working or something?21:40
*** jamieh has quit IRC21:40
flaper87kgriffs: is there an IRC log patch on gerrit?21:40
flaper87O.O21:40
* flaper87 chuckles21:40
kgriffsheh21:40
mpanettakgriffs: It works, but with unexpected results.21:41
kgriffsmpanetta: ok21:41
mpanettaYou cant do marconi-server &21:41
kgriffsso is the proposal to just remove the _enable_echo from cli.py ?21:42
mpanettaBecause isatty returns true (problem 1) and using termios causes that to stop the process because it lost control of stdin.21:42
kgriffs(along with L72-73)21:42
kgriffsheh21:42
mpanettaYes21:42
kgriffsok, it's not a big deal21:42
kgriffsby which I mean21:42
kgriffsI'm fine with dropping _enable_echo21:42
kgriffsflaper87: I'll get to those patches ASAP - may not be until this evening (US CST)21:44
mpanettaHow do I squash everything in to one commit?21:45
mpanettahttps://gist.github.com/anonymous/cdfb58f354c8d2545f7121:46
mpanettaSorry if that is a dumb question, I am not a git expert, yet :P21:46
mpanettaOr should I just say yes?21:46
mpanettaflaper87, kgriffs ^^21:46
flaper87no no, not yes21:47
mpanettaok21:47
flaper87you may actually want to get rid of the previous21:48
flaper87do this21:48
flaper87$ git log21:48
mpanettaok21:48
flaper87pick the sha of the parent of Fix marconi-server so it does not stop when backgrounded.21:48
flaper87then do21:48
flaper87$ git rebase -i $the_sha_you_copied21:48
flaper87then remove the line of the commit: "Fix marconi-server so it does not stop when backgrounded."21:48
flaper87save and exit21:48
mpanettahrm, it gives an error21:52
mpanettahttps://gist.github.com/anonymous/9616368da38670ccece921:52
mpanettaProbably something I did wrong...21:52
flaper87you'll have to resolve the conflict21:53
mpanettaOh, yeah I just found it21:53
flaper87mpanetta: lesson (1) Don't work on master ;)21:53
flaper87Says the guy that always works on master21:53
mpanettalol21:54
mpanettaYeah I realize I should have brabched21:54
mpanettaer branched21:54
mpanettaOk rebase successfull21:55
openstackgerritMike Panetta proposed a change to openstack/marconi: Remove termios support (and associated issues).  https://review.openstack.org/7666821:55
mpanettawee21:55
mpanettaOk, now the waiting starts? :)21:57
mpanettaNeed someone to drop that other review.22:00
flaper87mpanetta: you can drop it22:01
flaper87clic on the abandon button22:01
mpanettaI just figured it out when you sent that ;)22:01
mpanettaThank you!22:01
*** ayoung_ is now known as ayoung22:01
mpanettaI reffed the new review link in the abandon message.22:01
flaper87mpanetta: cool22:02
mpanettaSometimes I ask questions before thinking I guess :P22:03
mpanettaflaper87: Thank you for all the help!22:05
*** tedross has quit IRC22:11
kgriffsI know it hasn't gated yet, but I am going to preemptively +2 since I happen to be online right now. :p22:21
openstackgerritKurt Griffiths proposed a change to openstack/marconi: test(wsgi): Refactor tests in preparation for API v1.1  https://review.openstack.org/7604022:24
openstackgerritKurt Griffiths proposed a change to openstack/marconi: feat(api): API v1.1 bulk delete change  https://review.openstack.org/7609322:24
*** jcru_ has quit IRC22:25
*** jcru has joined #openstack-marconi22:26
*** jcru has quit IRC22:30
flaper87mpanetta: my pleasure22:33
flaper87mpanetta: your patch will fail because of pep822:34
flaper87there are too many blank lines22:34
flaper87mpanetta: run tox -epep822:34
openstackgerritKurt Griffiths proposed a change to openstack/marconi: fix(mongodb): Limit kwarg must be an int  https://review.openstack.org/7667822:35
*** jcru has joined #openstack-marconi22:36
mpanettaflaper87: ok22:37
*** mpanetta has quit IRC22:41
kgriffsjudging by how long the tempest test is taking, I suspect it is working now23:01
kgriffs:p23:01
kgriffsin other news, we need to make the mongo tests run a bit faster.23:02
kgriffs;)23:02
openstackgerritJenkins proposed a change to openstack/marconi: Updated from global requirements  https://review.openstack.org/7669223:28
*** kgriffs is now known as kgriffs_afk23:31
openstackgerritJenkins proposed a change to openstack/python-marconiclient: Updated from global requirements  https://review.openstack.org/7669923:34
*** amitgandhi has quit IRC23:38
*** sballe has quit IRC23:50
*** sballe has joined #openstack-marconi23:50
*** vkmc has quit IRC23:56
*** ametts_ has quit IRC23:56
*** sballe has quit IRC23:58

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