| *** Liang__ has joined #openstack-glance | 01:22 | |
| *** threestrands has joined #openstack-glance | 02:51 | |
| openstackgerrit | Qiu Fossen proposed openstack/python-glanceclient master: Remove Babel requirement https://review.opendev.org/723302 | 03:26 |
|---|---|---|
| *** ratailor has joined #openstack-glance | 04:24 | |
| *** evrardjp has quit IRC | 04:35 | |
| *** evrardjp has joined #openstack-glance | 04:35 | |
| *** udesale has joined #openstack-glance | 05:29 | |
| *** CeeMac has joined #openstack-glance | 06:16 | |
| openstackgerrit | Abhishek Kekane proposed openstack/glance_store master: Fix: API returns 503 if one of the store is mis-configured https://review.opendev.org/722915 | 06:47 |
| openstackgerrit | Merged openstack/glance stable/ussuri: Imported Translations from Zanata https://review.opendev.org/723141 | 07:08 |
| *** rcernin has quit IRC | 08:13 | |
| *** threestrands has quit IRC | 08:20 | |
| openstackgerrit | Merged openstack/python-glanceclient master: Add py38 package metadata https://review.opendev.org/722742 | 08:22 |
| *** tkajinam has quit IRC | 08:23 | |
| openstackgerrit | Merged openstack/glance master: Add py38 package metadata https://review.opendev.org/722695 | 08:28 |
| openstackgerrit | Merged openstack/glance_store master: Add py38 package metadata https://review.opendev.org/722696 | 08:31 |
| *** Liang__ has quit IRC | 10:15 | |
| *** smcginnis has quit IRC | 11:40 | |
| *** smcginnis has joined #openstack-glance | 11:41 | |
| *** rosmaita has left #openstack-glance | 12:34 | |
| *** rosmaita has joined #openstack-glance | 12:39 | |
| *** udesale_ has joined #openstack-glance | 13:20 | |
| *** udesale has quit IRC | 13:23 | |
| *** ratailor has quit IRC | 13:27 | |
| *** tkajinam has joined #openstack-glance | 13:42 | |
| *** irclogbot_0 has joined #openstack-glance | 14:08 | |
| abhishekk | rosmaita, around? | 14:11 |
| rosmaita | yep, good morning | 14:11 |
| rosmaita | or afternoon | 14:12 |
| abhishekk | evening :D | 14:12 |
| abhishekk | good morning to you | 14:12 |
| *** irclogbot_0 has quit IRC | 14:12 | |
| *** irclogbot_1 has joined #openstack-glance | 14:22 | |
| *** irclogbot_1 has quit IRC | 14:25 | |
| *** irclogbot_1 has joined #openstack-glance | 14:26 | |
| *** irclogbot_1 has quit IRC | 14:29 | |
| *** irclogbot_2 has joined #openstack-glance | 14:30 | |
| *** irclogbot_2 has quit IRC | 14:35 | |
| *** irclogbot_2 has joined #openstack-glance | 14:36 | |
| *** irclogbot_2 has quit IRC | 14:39 | |
| *** irclogbot_3 has joined #openstack-glance | 14:40 | |
| *** irclogbot_3 has quit IRC | 14:45 | |
| *** irclogbot_3 has joined #openstack-glance | 14:46 | |
| jokke_ | whoami-rajat: ping | 14:49 |
| *** irclogbot_3 has quit IRC | 14:51 | |
| *** irclogbot_2 has joined #openstack-glance | 14:52 | |
| *** tkajinam has quit IRC | 14:54 | |
| *** irclogbot_2 has quit IRC | 14:55 | |
| *** irclogbot_2 has joined #openstack-glance | 14:56 | |
| whoami-rajat | jokke_, hi | 14:58 |
| *** irclogbot_2 has quit IRC | 14:59 | |
| *** irclogbot_2 has joined #openstack-glance | 15:00 | |
| *** irclogbot_2 has quit IRC | 15:03 | |
| *** irclogbot_3 has joined #openstack-glance | 15:04 | |
| jokke_ | whoami-rajat: hav a look and let me know what you think http://paste.openstack.org/show/792761/ ... I just sent a mail out (internally) abot that as well as I wasn't sure if you were around | 15:05 |
| jokke_ | whoami-rajat: ^^ is a diff against your PS8 | 15:05 |
| jokke_ | I think abhishekk is testing that too as we speak | 15:06 |
| *** irclogbot_3 has quit IRC | 15:07 | |
| abhishekk | jokke_, yes tested, created 10 images so far and its working as expected | 15:07 |
| *** irclogbot_0 has joined #openstack-glance | 15:08 | |
| *** irclogbot_0 has quit IRC | 15:11 | |
| abhishekk | jokke_, after 11th image it has created new object :/ | 15:11 |
| *** irclogbot_2 has joined #openstack-glance | 15:12 | |
| jokke_ | abhishekk: but is there multiples of them? Looking the code I'd expect that to be the case when garbage collection cleans that manager away | 15:12 |
| abhishekk | jokke_, yes, I am thinking the same | 15:13 |
| abhishekk | for next images it is using same id/object of the 11th image | 15:13 |
| abhishekk | :/ | 15:14 |
| abhishekk | nope, 1st one is back again | 15:14 |
| *** irclogbot_2 has quit IRC | 15:15 | |
| *** irclogbot_0 has joined #openstack-glance | 15:16 | |
| jokke_ | abhishekk: hmm-m ... that should never happen unless we have different namespaces running simultaneously | 15:16 |
| abhishekk | not sure | 15:17 |
| abhishekk | Found new issue though | 15:17 |
| jokke_ | abhishekk: how are you monitoring that inside the process? | 15:17 |
| abhishekk | I have added debug log | 15:17 |
| jokke_ | and they are logged by same pid? | 15:18 |
| abhishekk | wait may be I haven't used entire your code, let me check it again | 15:18 |
| jokke_ | Anywhere I could have a look? | 15:18 |
| jokke_ | kk | 15:19 |
| *** irclogbot_0 has quit IRC | 15:19 | |
| *** irclogbot_3 has joined #openstack-glance | 15:20 | |
| abhishekk | give me a minute | 15:21 |
| *** irclogbot_3 has quit IRC | 15:23 | |
| *** irclogbot_2 has joined #openstack-glance | 15:24 | |
| abhishekk | I was doing this way | 15:25 |
| abhishekk | http://paste.openstack.org/show/792762/ | 15:25 |
| *** brinzhang_ has quit IRC | 15:25 | |
| *** brinzhang_ has joined #openstack-glance | 15:26 | |
| *** irclogbot_2 has quit IRC | 15:27 | |
| *** irclogbot_1 has joined #openstack-glance | 15:28 | |
| *** brinzhang_ has quit IRC | 15:28 | |
| *** brinzhang_ has joined #openstack-glance | 15:28 | |
| abhishekk | jokke_, whoami-rajat tested code as per http://paste.openstack.org/show/792764/ | 15:30 |
| abhishekk | and for 20 images it has used same object | 15:30 |
| abhishekk | http://paste.openstack.org/show/792763/ | 15:30 |
| *** irclogbot_1 has quit IRC | 15:31 | |
| *** irclogbot_3 has joined #openstack-glance | 15:37 | |
| *** udesale_ has quit IRC | 16:26 | |
| *** vesper11 has quit IRC | 16:26 | |
| *** hoonetorg has quit IRC | 16:26 | |
| *** noonedeadpunk has quit IRC | 16:26 | |
| *** abhishekk has quit IRC | 16:26 | |
| *** Blinkiz has quit IRC | 16:26 | |
| *** tonyb has quit IRC | 16:26 | |
| *** kgz has quit IRC | 16:26 | |
| *** kgz has joined #openstack-glance | 16:28 | |
| *** irclogbot_3 has quit IRC | 16:28 | |
| *** vesper11 has joined #openstack-glance | 16:29 | |
| *** irclogbot_0 has joined #openstack-glance | 16:29 | |
| *** hoonetorg has joined #openstack-glance | 16:31 | |
| *** udesale_ has joined #openstack-glance | 16:31 | |
| *** noonedeadpunk has joined #openstack-glance | 16:32 | |
| *** abhishekk has joined #openstack-glance | 16:32 | |
| *** Blinkiz has joined #openstack-glance | 16:32 | |
| *** tonyb has joined #openstack-glance | 16:32 | |
| *** evrardjp has quit IRC | 16:35 | |
| whoami-rajat | abhishekk, jokke_ the implementation mentioned uses __manager__ = _HostMountStateManager(HOST) | 16:37 |
| abhishekk | yes | 16:38 |
| whoami-rajat | abhishekk, jokke_ i thought the point of introducing metaclass singleton was to have a class who will return one instance doesn't matter how many times we initialize it | 16:38 |
| jokke_ | whoami-rajat: it does | 16:40 |
| jokke_ | as long as there is that one, if you let it to garbage collect it obviously gets recreated | 16:41 |
| jokke_ | so you don't need to use the __manager__, you will not necessarily get always the same object but there wil ever be just one | 16:42 |
| whoami-rajat | jokke_, hmm, now i don't see much difference in my implementation and this one, i achieved the same result with mine | 16:43 |
| jokke_ | also in that case you could also just call the managers (u)mount functions directly | 16:43 |
| jokke_ | whoami-rajat: The big difference is that with this implementation it literally doesn't matter how many times you call the class to create new object you get the existing one. Your original implementation relied on the module level get function and should someone call the manager class directly they would get another instance of it | 16:45 |
| jokke_ | which is very Java-like thing to do and unfortunately by no means enforced in Python | 16:46 |
| jokke_ | like it's in Java | 16:46 |
| *** evrardjp has joined #openstack-glance | 16:46 | |
| whoami-rajat | jokke_, nope. the get_manager call was for using the manager externally, it wasn't used so i removed it in my latest patchset | 16:46 |
| whoami-rajat | s/call/function | 16:46 |
| whoami-rajat | jokke_, and yes, we could get new manager object with new initialization but it wasn't called anywhere | 16:47 |
| jokke_ | whoami-rajat: like said the diff I sent to you fixed the metaclassing, you want to clean up that more, feel free. | 16:47 |
| jokke_ | didn't want to spend time to rewrite your whole patch, just the part you couldn't get working ;) | 16:48 |
| jokke_ | easier than trying to explain how it works | 16:49 |
| whoami-rajat | jokke_, ack. thanks for the help on it. i will update the patchset. | 16:49 |
| whoami-rajat | also thanks abhishekk for the help | 16:49 |
| jokke_ | whoami-rajat: the __manager__ is "handy" in a sense that it keeps the manager in memory all the time, not just when we do need to drive those mounts | 16:50 |
| *** udesale_ has quit IRC | 16:50 | |
| jokke_ | whoami-rajat: but if you're happy to just ensure we ever have only one instance of that object it's totally not needed and can be cleaned from there | 16:51 |
| jokke_ | but specially if we take the approach of cleaning up when it's initialized keeping it running might speed things up a bit | 16:52 |
| openstackgerrit | Merged openstack/glance-specs master: Deprecate 'checksum' computation https://review.opendev.org/708761 | 17:00 |
| abhishekk | jokke_, security bug has been reported as well ^^ | 17:02 |
| *** noonedeadpunk has quit IRC | 17:26 | |
| *** abhishekk has quit IRC | 17:26 | |
| *** Blinkiz has quit IRC | 17:26 | |
| *** tonyb has quit IRC | 17:26 | |
| *** noonedeadpunk has joined #openstack-glance | 17:29 | |
| *** abhishekk has joined #openstack-glance | 17:29 | |
| *** Blinkiz has joined #openstack-glance | 17:29 | |
| *** tonyb has joined #openstack-glance | 17:29 | |
| jokke_ | abhishekk: ack ... will check after dinner | 17:44 |
| abhishekk | jokke_, ack, have a nice one | 17:45 |
| abhishekk | I am signing off for the day | 17:45 |
| jokke_ | kk, gn | 17:47 |
| abhishekk | thanks, have a nice time ahead | 17:50 |
| *** hoonetorg has quit IRC | 18:02 | |
| openstackgerrit | Merged openstack/glance master: Deprecate the 'checksum' image property https://review.opendev.org/718147 | 18:19 |
| openstackgerrit | Brian Rosmaita proposed openstack/glance stable/ussuri: Deprecate the 'checksum' image property https://review.opendev.org/723638 | 18:36 |
| *** brinzhang_ has quit IRC | 20:14 | |
| *** brinzhang_ has joined #openstack-glance | 20:14 | |
| *** brinzhang has joined #openstack-glance | 21:08 | |
| *** brinzhang_ has quit IRC | 21:11 | |
| *** rcernin has joined #openstack-glance | 21:14 | |
| openstackgerrit | Rajat Dhasmana proposed openstack/glance_store master: Add lock per share for cinder nfs mount/umount https://review.opendev.org/716874 | 22:46 |
| *** tkajinam has joined #openstack-glance | 22:49 | |
| *** tkajinam has quit IRC | 22:49 | |
| *** tkajinam has joined #openstack-glance | 22:50 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!