Wednesday, 2015-11-18

*** akshai_ has quit IRC00:05
*** rhagarty has joined #openstack-manila00:10
*** rushil has quit IRC00:19
openstackgerritOpenStack Proposal Bot proposed openstack/python-manilaclient: Updated from global requirements  https://review.openstack.org/24423500:23
*** gouthamr has quit IRC00:42
*** jasonsb_ has quit IRC00:42
*** jasonsb has joined #openstack-manila00:42
*** jasonsb has quit IRC00:47
*** rushil has joined #openstack-manila00:48
*** rushil has quit IRC01:01
*** cknight1 has joined #openstack-manila01:10
*** cknight has quit IRC01:13
bswartzcsaba: pong01:13
*** leeantho has quit IRC01:21
*** eharney has quit IRC01:38
*** gouthamr has joined #openstack-manila02:01
*** gouthamr_ has joined #openstack-manila02:02
*** rushil has joined #openstack-manila02:03
*** gouthamr has quit IRC02:05
*** rushil has quit IRC02:10
*** jasonsb has joined #openstack-manila02:20
openstackgerritMerged openstack/manila: Refactor HP 3PAR share driver to now be HPE  https://review.openstack.org/24035302:29
*** 7JTAA216T has joined #openstack-manila02:30
*** 7JTAA216T has quit IRC02:30
*** 21WAACTNP has joined #openstack-manila02:31
*** 21WAACTNP has quit IRC03:01
*** haomaiwang has joined #openstack-manila03:01
*** lpabon has quit IRC03:05
*** jasonsb has quit IRC03:33
*** gouthamr_ has quit IRC03:37
*** akshai has joined #openstack-manila03:44
*** haomaiwang has quit IRC04:01
*** haomaiwang has joined #openstack-manila04:01
*** jasonsb has joined #openstack-manila04:04
*** akerr_ is now known as akerr_away04:05
*** dcain has quit IRC04:09
*** jasonsb has quit IRC04:09
openstackgerritNidhiMittalHada proposed openstack/python-manilaclient: Allows type-create to return dict instead of list  https://review.openstack.org/24573104:23
*** cknight1 has quit IRC04:36
*** jasonsb has joined #openstack-manila04:52
*** haomaiwang has quit IRC05:01
*** haomaiwang has joined #openstack-manila05:01
*** akshai has quit IRC05:05
*** mtanino has quit IRC05:09
openstackgerritMerged openstack/manila: Prevent Share operations during share migration  https://review.openstack.org/23305205:22
*** deepakcs has joined #openstack-manila05:52
*** yangyapeng has joined #openstack-manila05:57
*** haomaiwang has quit IRC06:01
*** haomaiwang has joined #openstack-manila06:01
*** nidhimittalhada has joined #openstack-manila06:21
nidhimittalhada@vponomaryov106:22
nidhimittalhadamanilaclient/common/constants.py" 8206:23
nidhimittalhadain this file we used to keep max_api_version06:23
nidhimittalhadai kept it this way06:23
nidhimittalhadaEXPERIMENTAL_HTTP_HEADER = 'X-OpenStack-Manila-API-Experimental'06:23
nidhimittalhadaMAX_API_VERSION = '2.7'06:23
nidhimittalhadaV1_API_VERSION = '1.0'06:23
nidhimittalhadaV1_SERVICE_TYPE = 'share'06:23
nidhimittalhadaV2_SERVICE_TYPE = 'sharev2'06:23
nidhimittalhadabut now latest version of this file doesnt contain MAX_API_VERSION at all06:23
nidhimittalhadawhere can we keep this version now?06:23
*** lpetrut has joined #openstack-manila06:33
*** nidhimittalhada has quit IRC06:40
*** haomaiwang has quit IRC06:42
*** nidhimittalhada has joined #openstack-manila06:48
*** haomaiwang has joined #openstack-manila06:49
*** haomaiwang has quit IRC07:01
*** haomaiwang has joined #openstack-manila07:01
*** nkrinner has joined #openstack-manila07:14
openstackgerritNidhiMittalHada proposed openstack/python-manilaclient: API 'manage' can set share attr 'is_public'  https://review.openstack.org/23096307:16
*** nidhimittalhada has quit IRC07:35
*** akerr has joined #openstack-manila07:40
*** netapp has quit IRC07:42
*** timcl1 has joined #openstack-manila07:42
*** akerr_away has quit IRC07:43
*** timcl has quit IRC07:43
*** netapp has joined #openstack-manila07:48
*** sgotliv_ has quit IRC07:56
*** breitz has quit IRC07:58
*** breitz1 has joined #openstack-manila07:58
*** haomaiwang has quit IRC08:01
*** haomaiwang has joined #openstack-manila08:01
*** lpetrut has quit IRC08:07
*** lpetrut has joined #openstack-manila08:07
*** zhongjun has quit IRC08:07
*** zhongjun has joined #openstack-manila08:08
*** jvarlamova has joined #openstack-manila08:13
*** mkoderer has quit IRC08:15
*** mkoderer has joined #openstack-manila08:18
*** lpetrut has quit IRC08:45
*** akerr_ has joined #openstack-manila08:57
*** timcl has joined #openstack-manila08:58
*** timcl1 has quit IRC08:59
*** akerr has quit IRC08:59
*** netapp has quit IRC09:00
*** haomaiwang has quit IRC09:01
*** haomaiwang has joined #openstack-manila09:01
*** netapp has joined #openstack-manila09:03
*** sgotliv_ has joined #openstack-manila09:13
*** akerr has joined #openstack-manila09:24
*** akerr_ has quit IRC09:26
*** rraja_ has quit IRC09:45
*** lpetrut has joined #openstack-manila09:52
*** ociuhandu has quit IRC10:00
*** haomaiwang has quit IRC10:01
*** haomaiwang has joined #openstack-manila10:01
*** martyturner has joined #openstack-manila10:02
*** martyturner has quit IRC10:06
*** dzamboni has joined #openstack-manila10:18
openstackgerritSilvan Kaiser proposed openstack/manila: Implements ensure_share() in Quobyte driver  https://review.openstack.org/23968610:24
*** tpsilva has joined #openstack-manila10:36
*** eduardo_ has joined #openstack-manila10:42
openstackgerritIgor Malinovskiy proposed openstack/manila: Add update_access() method to driver interface  https://review.openstack.org/24512610:46
*** deepakcs has quit IRC10:50
*** haomaiwang has quit IRC11:01
*** haomaiwang has joined #openstack-manila11:01
*** yangyapeng has quit IRC11:13
*** shinobu_ has joined #openstack-manila11:13
openstackgerritSilvan Kaiser proposed openstack/manila: Add share id to Quobyte backend volume name  https://review.openstack.org/24167111:13
*** shinobu_ is now known as Guest1384811:13
*** Guest13848 has quit IRC11:14
*** shinobu___ has joined #openstack-manila11:15
*** haomaiwang has quit IRC11:51
*** ociuhandu has joined #openstack-manila12:19
*** ociuhandu_ has joined #openstack-manila12:22
*** ociuhandu has quit IRC12:23
*** ociuhandu_ is now known as ociuhandu12:23
*** jsheeren has joined #openstack-manila12:29
*** yangyapeng has joined #openstack-manila12:32
*** shinobu___ has quit IRC12:33
*** sgotliv_ has quit IRC12:43
*** sgotliv has joined #openstack-manila12:51
*** gouthamr has joined #openstack-manila12:52
*** akerr has quit IRC12:59
*** gouthamr has quit IRC13:31
*** akerr has joined #openstack-manila13:34
*** gouthamr has joined #openstack-manila13:37
*** bswartz has quit IRC13:42
gouthamrvponomaryov: Would appreciate your input on this: https://review.openstack.org/#/c/246555/13:44
*** tpsilva has quit IRC13:49
*** merooney has joined #openstack-manila13:50
*** jsheeren has quit IRC13:51
*** absubram has joined #openstack-manila13:52
openstackgerritJulia Varlamova proposed openstack/manila: Add LVM driver  https://review.openstack.org/23297013:52
*** dcain has joined #openstack-manila13:53
*** bill_az has joined #openstack-manila13:53
*** eharney has joined #openstack-manila13:53
*** dcain1 has joined #openstack-manila13:54
*** dcain has quit IRC13:57
*** porrua has joined #openstack-manila13:58
*** akshai has joined #openstack-manila14:04
openstackgerritSilvan Kaiser proposed openstack/manila: Implements ensure_share() in Quobyte driver  https://review.openstack.org/23968614:09
*** bswartz has joined #openstack-manila14:09
*** absubram has quit IRC14:27
*** cknight has joined #openstack-manila14:29
*** tpsilva has joined #openstack-manila14:30
*** dustins has joined #openstack-manila14:30
*** dcain has joined #openstack-manila14:31
*** openstackgerrit has quit IRC14:31
*** openstackgerrit has joined #openstack-manila14:32
*** dcain1 has quit IRC14:33
*** lpabon has joined #openstack-manila14:46
*** dustins has quit IRC14:57
*** dustins has joined #openstack-manila15:00
vponomaryovgouthamr: done15:04
*** a_ta has joined #openstack-manila15:04
gouthamrthanks vponomaryov15:08
*** sgotliv has quit IRC15:15
*** adrianofr has quit IRC15:16
vponomaryovbswartz, cknight, markstur, toabctl: If you have time, please look at https://review.openstack.org/#/c/239689/15:17
*** adrianofr has joined #openstack-manila15:18
cknightvponomaryov: OK15:18
*** rushil has joined #openstack-manila15:22
bswartzvponomaryov: I already looked at it15:22
gouthamrvponomaryov: how strongly do you feel against using function names as the action names for policy checking? wsgi does its routing based on function names :)15:23
gouthamrvponomaryov: reference: https://review.openstack.org/#/c/246555/1/manila/api/openstack/wsgi.py #114515:24
toabctlvponomaryov: looks good. merging...15:24
vponomaryovgouthamr: routing can be set up to any variant in router.py module15:24
toabctlah. bswartz started already the workflow :)15:25
vponomaryovcknight, bswartz, toabctl: thanks!15:25
vponomaryovgouthamr: absolutely any variant, wsgi just have defaults15:25
gouthamrvponomaryov: the idea was to allow common methods that most APIs implement ->  'create', 'delete', 'update', 'get',  'action' and such..15:26
*** rushil has quit IRC15:26
vponomaryovgouthamr: such double approach lead to documenting improper thing - call without args15:27
gouthamrvponomaryov: args are implied in that case.. its the method being decorated..15:27
vponomaryovgouthamr: exactly call without args - decorator()15:28
gouthamrvponomaryov: in most cases, we don't change the function names for CRUD ops..15:28
*** sgotliv has joined #openstack-manila15:28
vponomaryovtoabctl: why https://review.openstack.org/#/c/242960/ and https://review.openstack.org/#/c/242957/ in separate commits?15:36
vponomaryovtoabctl: maybe better to combine them?15:37
*** haomaiwang has joined #openstack-manila15:37
toabctlvponomaryov: one is for stable/liberty15:37
toabctlvponomaryov: and I just followed the example given on the ML15:37
vponomaryovtoabctl: oh, right15:37
vponomaryovdidn't look at branch15:38
toabctlvponomaryov: tbh I don't understand the process completly yet. as I said, I followed the example and iirc this should be merged before mitaka-115:38
*** markstur has quit IRC15:38
vponomaryovtoabctl: but then liberty branch requires original reno support anyway15:39
vponomaryovtoabctl: same commit as for master15:39
toabctlvponomaryov: yes. I think that needs to be backported if merged in master.15:39
vponomaryovcsaba: you have lots of commits, can you point to specific one that should be reviewed at first priority?15:41
*** xyang1 has joined #openstack-manila15:48
*** hodos has joined #openstack-manila15:52
*** houming-wang has joined #openstack-manila15:56
*** dustins has quit IRC15:58
*** dustins has joined #openstack-manila15:58
openstackgerritEmilien Macchi proposed openstack/puppet-manila: Bump UCA and RDO to Liberty GA  https://review.openstack.org/24164215:59
vponomaryovgouthamr:  https://bugs.launchpad.net/manila/+bug/149292015:59
openstackLaunchpad bug 1492920 in Manila " Consistency Group Support for the Generic Driver" [Undecided,New]15:59
vponomaryovgouthamr: what do you know about this?16:00
gouthamrvponomaryov: wrote that code..16:01
*** haomaiwang has quit IRC16:01
gouthamrvponomaryov: I guess 'DocImpact' wasn't needed?16:01
*** haomaiwang has joined #openstack-manila16:01
vponomaryovgouthamr: no16:01
gouthamrvponomaryov: don't understand the bug though.. it says "This bug was created here because we did not know how to map the project name "openstack/manila" to a launchpad project name."16:02
vponomaryovgouthamr: each "DocImpact" autocreates bug for openstack-manuals project16:03
vponomaryovgouthamr: so, I guess, they considered it is invalid for their project and it means "DocImpact" is redundant16:03
vponomaryovgouthamr: so, I close this bug16:03
gouthamrvponomaryov: oh. yes. sure.16:04
openstackgerritClinton Knight proposed openstack/manila: Modified last sync from oslo-incubator  https://review.openstack.org/24700316:04
gouthamrvponomaryov: https://review.openstack.org/#/c/219891/3..1//COMMIT_MSG16:05
cknightvponomaryov, u_glide1:  This is the oslo incubator sync for Manila, minus the file Igor had modified.  https://review.openstack.org/#/c/247003/16:05
cknightvponomaryov, u_glide1:  Need this so I can refactor capabilities_filter to accept lists.16:06
*** bswartz has quit IRC16:06
cknightvponomaryov, u_glide1:  I also need to pull the unit tests for the files we got from the incubator, since we own them now.  That's a separate patch.16:06
*** rushil has joined #openstack-manila16:08
*** erlon has quit IRC16:08
vponomaryovgouthamr: it should have been added to dev docs , which are located in manila repo16:13
vponomaryovgouthamr: it has nothing for DocImpact16:13
gouthamrvponomaryov: makes sense. I can close that bug with a dev docs patch16:14
vponomaryovcknight: when do you plan to port tests?16:14
vponomaryovgouthamr: already set as invalid ))16:14
gouthamrvponomaryov: alright, new bug then.16:14
*** leeantho has joined #openstack-manila16:15
cknightvponomaryov: soon :-)  But there are more files from the incubator that just what we got in the last sync, so I figured that would be a separate effort.  Doing the sync now doesn't reduce our coverage.16:15
vponomaryovgouthamr: you can modify description in this one16:15
*** houming-wang has quit IRC16:15
vponomaryovgouthamr: then we change status to correct one16:15
gouthamrvponomaryov: sure thing16:16
cknightvponomaryov: some of the unit tests don't mirror the directory/file structure of the code, so I'm wondering if I should fix that at the same time.16:16
gouthamrvponomaryov: thanks for the heads up. will change it.16:16
*** akerr has quit IRC16:16
*** yangyapeng has quit IRC16:17
*** haomaiwang has quit IRC16:51
*** dzamboni has quit IRC17:13
*** ociuhandu has quit IRC17:17
openstackgerritValeriy Ponomaryov proposed openstack/manila: Improve Tempest tests for consistency groups  https://review.openstack.org/24705217:21
openstackgerritValeriy Ponomaryov proposed openstack/manila: Improve Tempest tests for consistency groups  https://review.openstack.org/24705217:22
*** nkrinner has quit IRC17:23
openstackgerritValeriy Ponomaryov proposed openstack/manila: Improve Tempest tests for consistency groups  https://review.openstack.org/24705217:24
*** nkrinner has joined #openstack-manila17:24
*** lpetrut has quit IRC17:25
openstackgerritValeriy Ponomaryov proposed openstack/manila: Improve Tempest tests for consistency groups  https://review.openstack.org/24705217:25
*** porrua has quit IRC17:38
*** verdurin has quit IRC17:55
*** bswartz has joined #openstack-manila17:56
*** eduardo_ has quit IRC18:02
*** porrua has joined #openstack-manila18:03
*** akerr has joined #openstack-manila18:03
*** akerr_ has joined #openstack-manila18:06
*** akerr has quit IRC18:09
gansobswartz: ping18:10
*** markstur has joined #openstack-manila18:13
*** verdurin has joined #openstack-manila18:22
*** ociuhandu has joined #openstack-manila18:24
*** lpetrut has joined #openstack-manila18:48
*** erlon has joined #openstack-manila18:50
bswartzganso: pong18:50
gansobswartz: Hi18:51
gansobswartz: I am moving migration code to data copy service18:51
gansobswartz: now is the time that we will need network expertise18:52
gansobswartz: this code https://github.com/openstack/manila/blob/8247f350af7db639581c58e274e63463afe544c2/manila/share/drivers/generic.py#L218 will no longer work18:52
gansobswartz: because the machine copying data may not be the same running the driver anymore18:52
gansobswartz: for DHSS=true drivers, every share server in each share network need to be accessible by the data copy service node... currently the drivers themselves are responsible for obtaining the IP that allows such connectivity18:54
gansobswartz: I think we need to discuss some approaches that may be more feasible... like an admin creating a fixed neutron port that connects any share server (if such thing is possible), or if drivers create VIFs for each share server... or admin-network facing export locations...18:55
bswartzganso: I agree that the admin-facing network stuff is fairly urgent18:57
bswartzganso: is that something you're interested in tackling yourself or do you need help?18:58
bswartzI'll work with whoever is willing to do it, and if there's no volunteers I'll designate someone18:58
gansobswartz: any network help is much appreciated... like whoever designed the VIFs code for the generic driver may be able to help18:59
*** rushil has quit IRC18:59
gansobswartz: my favorite approach is a single dedicated neutron port for data copy service node... if that is possible19:00
gansobswartz: this may be something we can discuss on tomorrow's meeting19:00
gansobswartz: I can draw a diagram19:01
bswartzwell what I envision is pretty simply actually19:01
bswartzat least, my vision for the driver interface in the share manager19:01
bswartzhow drivers implement it could end up being complicated, but in the general case it shouldn't be19:01
*** dcain1 has joined #openstack-manila19:02
gansobswartz: what would be the general case?19:03
gansobswartz: I think we still prefer involving drivers as less as possible19:03
*** dcain has quit IRC19:06
*** tpsilva has quit IRC19:06
bswartzganso: so my proposal is simple19:08
bswartzganso: any driver that support share servers needs to ensure that it creates an additional network interface, reachable from the admin network19:09
bswartzmost likely we'll need to add a configuration parameter to allow admin to specify how drivers should obtain IP addresses and network details for that network19:09
bswartzbut it will ultimately look a lot like the existing share network code19:10
gansobswartz: one VIF per share server approach. That would make the current code I linked above compatible. The difference is that the service_ip would be that VIF ip, instead of the VIF that connects back to manila-share node19:10
bswartzthen, every share created by share-server drivers will need to create an additional export location, tagged as admin-only19:10
bswartzand the migration code will use those admin export locations19:11
bswartzwe would want to make sure the admin export locations were hidden from non-admins19:11
*** rushil has joined #openstack-manila19:15
gansobswartz: extra export locations may not be necessary, as they are not today19:17
bswartzwell I like the idea of the driver telling manila what is the best way to access shares for administrative purposes19:17
*** absubram has joined #openstack-manila19:18
gansobswartz: a code that is able to interpret several ways a driver may tell the manager is probably more complex than a way we enforce (if it works for all vendors)19:19
gansobswartz: this is the main reason I haven't gone down that road19:19
gansobswartz: and also that we haven't had vendors input on this19:19
gansobswartz: so far19:19
bswartzganso: The goal of this change is to force standardization so it's easier from yourside19:20
bswartzif we implement this new interface, then the migration code only has 2 paths -- share-servers and no-share-servers19:21
gansobswartz: that path is related to obtaining the IP the data copy service should mount, and which IP to allow access (data copy service node's). Both are already in the driver interface to be overriden19:23
openstackgerritMerged openstack/python-manilaclient: Updated from global requirements  https://review.openstack.org/24423519:28
openstackgerritMerged openstack/manila: Port admin actions extension to core API  https://review.openstack.org/23968919:40
*** rushil has quit IRC19:42
*** rushil has joined #openstack-manila19:43
*** vkmc is now known as vkmc-afk19:44
*** jasonsb has quit IRC19:55
*** lpetrut has quit IRC19:57
*** lpetrut has joined #openstack-manila20:00
*** cknight has quit IRC20:03
openstackgerritMerged openstack/manila: Port share type extensions to core API  https://review.openstack.org/23991320:04
*** sgotliv has quit IRC20:12
*** cknight has joined #openstack-manila20:15
*** lpetrut has quit IRC20:22
openstackgerritOpenStack Proposal Bot proposed openstack/manila: Updated from global requirements  https://review.openstack.org/24609520:39
openstackgerritOpenStack Proposal Bot proposed openstack/manila: Updated from global requirements  https://review.openstack.org/24609520:47
*** verdurin has quit IRC20:49
*** jasonsb has joined #openstack-manila20:50
*** jasonsb has quit IRC20:50
*** jasonsb has joined #openstack-manila20:51
openstackgerritOpenStack Proposal Bot proposed openstack/python-manilaclient: Updated from global requirements  https://review.openstack.org/24715620:51
*** rebase has joined #openstack-manila20:53
openstackgerritClinton Knight proposed openstack/manila: Last sync to Manila from oslo-incubator  https://review.openstack.org/24700320:55
*** openstack has joined #openstack-manila21:04
csababswartz: ping21:06
bswartzcsaba: pong21:07
csababswartz: hi Ben, I'm sorry if missed an official announcement on this, but has the feature proposal deadline for Mitaka been decided?21:08
bswartzcsaba: it's still 2 weeks before M-3, so I would say ~ Feb 18 most likely21:11
bswartzhowever we agreed that for BIG features we need to see WIP code long before then21:11
bswartzwe just never agreed on the specifics21:11
csababswartz: yeah, I remember that, that's why I'm asking21:12
bswartzcsaba: did you have any big new features in mind?21:12
openstackgerritMark Sturdevant proposed openstack/manila: HPE3PAR Add support for access-level (ro,rw)  https://review.openstack.org/24538321:12
openstackgerritMerged openstack/manila: Added CONTRIBUTING file in .rst format  https://review.openstack.org/24557521:13
csababswartz: of RH originated contributions, probably the upcoming Ceph driver can qualify as such21:14
bswartzyes but drivers are inherently low risk21:14
bswartzif it's a large patch we'll probably want an earlier deadline just because of the size and time to review, but if it's not a core feature I'm less worried21:15
bswartzwhat RH submissions are likely to affect the Manila core in Mitaka?21:15
csabayeah that's why I thought it's indeed still not "big"21:15
openstackgerritValeriy Ponomaryov proposed openstack/manila: Add new URLs for APIs ported from extensions  https://review.openstack.org/24022021:16
openstackgerritValeriy Ponomaryov proposed openstack/manila: [TEST] Run Tempest with v2.6 having v2.7  https://review.openstack.org/24131121:16
*** a_ta has quit IRC21:17
csababswartz: other question: does Mitaka-{1,2} have any significance21:17
csaba?21:17
xyang1bswartz: two weeks before M-3 is deadline for new feature submission?  that is the same as what we had in Liberty21:17
*** a_ta has joined #openstack-manila21:17
*** a_ta has joined #openstack-manila21:17
bswartzcsaba: those are just milestones that we use to target blueprints that will be ready for release early21:18
*** verdurin has joined #openstack-manila21:18
csababswartz: OK, thamks21:18
bswartzxyang1: yes the plan keep the existing deadline and add an additional earlier deadline for big stuff21:18
bswartzxyang1: we still haven't decided how to define "big" and how much earlier those submissions need to be21:18
bswartzxyang1: I put it on the agenda for tomorrow though21:19
xyang1bswartz: ok21:19
bswartzxyang1: do you have any ideas?21:19
xyang1bswartz: I thought we discussed about an earlier merge deadline for core features21:20
xyang1bswartz: driver is different21:20
xyang1bswartz: driver can get merged until M-321:20
bswartzxyang1: what I wanted to to require that there's a WIP submission for core features by some earlier date, but not to necessarily change the merge deadline21:20
bswartzbecause for large new features, there could be many iterations of reviews and patches21:21
xyang1bswartz: that is true21:21
bswartzas long as that process starts early enough, I see no reason to force it to end before a specific date21:21
xyang1bswartz: we need to have a deadline for early submission then21:22
bswartzyeah that's my thinking21:22
bswartzmaybe for patches that add more than 500 lines of new code (not counting refactors that add and delete lots of lines but it's all just moved around) that we require WIP by M-2 date21:23
bswartzso we have 5 weeks to review instead of 2 weeks21:23
bswartzactually 6 weeks because M-3 is that long21:24
xyang1bswartz: sure21:24
openstackgerritOpenStack Proposal Bot proposed openstack/manila: Updated from global requirements  https://review.openstack.org/24609521:24
bswartzalthough we could pick some later date that's M3 minus 4 weeks or 5 weeks21:24
xyang1bswartz: a big refactor could be painful to review too21:24
*** breitz1 is now known as breitz21:25
xyang1bswartz: vponomaryov has complained about it:)21:25
bswartzrefactors take time to review but require nowhere near as many iterations as new core features21:25
bswartzthe concern is how many times the cycle needs to be repeated21:25
xyang1bswartz: I would actually suggest a big driver refactor to be submitted and reviewed as a new driver21:26
bswartzbig reviews are always painful but refactors tend to go more quickly because it's just comparing the old to the new and looking for bugs21:26
bswartzxyang1: we don't have any special deadlines for new drivers though...21:26
xyang1bswartz: yes21:26
bswartzif a reviewer wants to look at it like a new driver, then that's fine21:27
*** hodos has quit IRC21:27
bswartzI know some people prefer to review large refactors by just pulling the change and reviewing the whole file (without looking at the old file)21:27
bswartzwould making people submit those refactor patches earlier really save us any pain? or would it just shift it earlier?21:28
xyang1bswartz: do you suggest a new driver to be submitted 2weeks before M-321:28
bswartzxyang1: that's what we did for Liberty, and the new drivers didn't cause problems like the new features did21:29
bswartzI'm open to doing it differently21:30
bswartzI just want to make sure we're solving a real problem21:30
bswartzthe big pain for me was cgroups and migration21:30
*** verdurin has quit IRC21:31
xyang1bswartz: true, I don't remember if a new driver was submitted that late21:31
bswartzglusterfs native was the last drier to come in IIRC21:31
bswartzdriver21:31
bswartzand it was right at the deadline21:31
xyang1bswartz: I thought that was some driver enhancement rather than a brand new driver, I could be wrong21:32
bswartzit sounds like redhat is trying to decide when to submit the cephfs driver -- if we don't want it on Feb 18 then we should agree on when we want to see it and let redhat know what the deadline is21:33
bswartzxyang1: you might be right21:33
xyang1bswartz: ok21:33
bswartzthere were several driver enhancement patches for the glusterfs native driver though21:34
*** a_ta has quit IRC21:39
*** a_ta has joined #openstack-manila21:39
*** a_ta has quit IRC21:44
*** dustins has quit IRC21:45
*** a_ta has joined #openstack-manila21:46
*** verdurin has joined #openstack-manila21:59
*** bswartz has quit IRC22:00
*** verdurin has quit IRC22:03
*** a_ta has quit IRC22:16
*** a_ta has joined #openstack-manila22:17
*** a_ta_ has joined #openstack-manila22:18
*** sgotliv has joined #openstack-manila22:21
*** a_ta has quit IRC22:22
*** porrua has quit IRC22:24
*** a_ta_ has quit IRC22:25
*** a_ta has joined #openstack-manila22:26
*** a_ta has quit IRC22:27
*** a_ta has joined #openstack-manila22:28
*** jasonsb has quit IRC22:28
*** a_ta has quit IRC22:30
*** jasonsb has joined #openstack-manila22:30
*** rushil has quit IRC22:31
*** verdurin has joined #openstack-manila22:31
*** a_ta_ has joined #openstack-manila22:32
*** a_ta_ has quit IRC23:27
*** a_ta has joined #openstack-manila23:28
*** xyang1 has quit IRC23:28
*** zhongjun has quit IRC23:30
*** zhongjun has joined #openstack-manila23:31
*** a_ta has quit IRC23:32
*** cknight has quit IRC23:35
*** dcain1 has quit IRC23:52

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