harlowja | ok kevinbenton fixed, phew | 00:02 |
---|---|---|
harlowja | thought i busted it, lol | 00:02 |
harlowja | kevinbenton https://github.com/openstack/oslo.db/blob/master/oslo_db/sqlalchemy/exc_filters.py#L409-L411 | 00:06 |
harlowja | so that seems to be where u are going | 00:06 |
harlowja | which is basically a catch all | 00:06 |
harlowja | seems like u want a more specific catch 'filter' | 00:06 |
harlowja | my understanding is all the filters in that file are like a switch statement | 00:07 |
harlowja | with things like | 00:07 |
harlowja | case engine == mysql && exc_type == <type> && re.match(<someregex>, exc_msg) | 00:08 |
harlowja | except instead of a switch statement, u get a log of decorators | 00:08 |
harlowja | *lot of | 00:08 |
*** ihrachys has joined #openstack-oslo | 00:22 | |
kevinbenton | harlowja: so i want it to raise | 00:31 |
*** spzala has joined #openstack-oslo | 00:31 | |
kevinbenton | harlowja: but i don't want that icky LOG.exception call | 00:32 |
harlowja | ya, seems like we should just delete the LOG.exception then | 00:32 |
harlowja | not 100% sure why its there (vs in the calling code) | 00:33 |
kevinbenton | harlowja: oh | 00:34 |
kevinbenton | harlowja: i suspect the expression to turn this into DBReferenceError is not matching correctly | 00:34 |
kevinbenton | harlowja: it seems like this is the fallthrough for unmatched stuff | 00:35 |
harlowja | could be | 00:35 |
kevinbenton | so the proper fix is probably regex of some sort :) | 00:35 |
harlowja | prob :) | 00:36 |
*** mtanino has quit IRC | 00:36 | |
*** ihrachys has quit IRC | 00:40 | |
*** salv-orlando has quit IRC | 00:43 | |
*** Syed__ has quit IRC | 00:45 | |
*** yamamoto has joined #openstack-oslo | 00:51 | |
*** yamamoto has quit IRC | 00:52 | |
*** dimtruck is now known as zz_dimtruck | 01:04 | |
*** gcb has joined #openstack-oslo | 01:18 | |
amrith | harlowja, ping | 01:26 |
harlowja | sup | 01:26 |
amrith | hi joshua, did you get a chance to read the email I sent you | 01:26 |
amrith | I've been scratching my head to figure out how to get around this little mess. | 01:27 |
*** amotoki has joined #openstack-oslo | 01:27 | |
harlowja | hmmm | 01:27 |
harlowja | bad idea imho to mess with __dict__ :( | 01:27 |
amrith | well, here's our challenge. | 01:27 |
amrith | trove has a control plane and guests. | 01:27 |
amrith | when you do an upgrade, you have to upgrade one before the other. pick one. any one | 01:28 |
amrith | the issue we have is that once we do that upgrade (which typically is the control plane first) | 01:28 |
amrith | the guest could receive a context with stuff that it can't deal with | 01:28 |
amrith | this same issue can also happen in a control plane upgrade | 01:28 |
amrith | if you want to upgrade one service at a time. | 01:28 |
amrith | which is required if you want a zero-downtime upgrade mechanism. | 01:29 |
amrith | how is one supposed to handle this when a context is sent over a message (oslo_messaging.rpc) serialized as a dict? | 01:29 |
harlowja | ya, fair point | 01:30 |
harlowja | https://github.com/openstack/oslo.context/blob/master/oslo_context/context.py#L310 i don't think was altered though | 01:31 |
amrith | not this time | 01:33 |
amrith | but see https://github.com/openstack/trove/blob/master/trove/common/context.py#L67-L77 | 01:33 |
amrith | this code was put in when something changed in oslo.context | 01:33 |
*** amotoki has quit IRC | 01:33 | |
harlowja | ok | 01:33 |
harlowja | vars(cls()).keys() seems bad | 01:34 |
harlowja | why not just use to_dict().kes() | 01:34 |
harlowja | *keys | 01:34 |
amrith | see https://github.com/openstack/trove/commit/24c5e8e244380e3e7e309f0b4aaffde32dbf0789 | 01:34 |
amrith | and yes, the benefit of vars(cls()) is that you don't have to instantiate the class | 01:34 |
amrith | with to_dict() (which isn't a class method) you have to instantiate | 01:35 |
amrith | I'm trying to find some way to instantiate once and cache that set of keys | 01:35 |
amrith | because it won't change from one invocation to another. | 01:35 |
amrith | but in the fullness of time, I think we need to figure some way to make oslo.context upgrade safe or versioned or some such thing :) | 01:36 |
harlowja | ya, i'd like to know more about https://github.com/openstack/trove/commit/24c5e8e24438 | 01:37 |
harlowja | and what caused that | 01:37 |
amrith | one second, that's easy to find | 01:38 |
amrith | it was oslo.context 2.7.0 | 01:40 |
amrith | which added is_admin_project | 01:40 |
amrith | when the control plane (upgraded) sent this to the guest, the guest puked because the oslo.context (RequestContext) initialiation puked | 01:40 |
harlowja | ya, i can see that, it needs to really take a **kwargs in __init__ | 01:41 |
harlowja | *older versions that is | 01:41 |
harlowja | and just ignore the extra kwargs it may get | 01:41 |
amrith | yes, that could be one way to do it | 01:42 |
amrith | but when you change the semantics and (for example if you changed a variable from tenant to project) then that approach may not help | 01:42 |
amrith | unless there's a coexistance period | 01:43 |
amrith | where both project and tenant are sent :) | 01:43 |
amrith | in the dictionary | 01:43 |
harlowja | ya | 01:43 |
amrith | let me check with dougshelley66 | 01:44 |
amrith | he'll know for sure whether it was the is_admin thing | 01:44 |
amrith | or something else | 01:44 |
harlowja | ya, that would be my guess | 01:44 |
amrith | but, it was something like that | 01:44 |
amrith | in any event, I have to now find some solution whcih accounts for the fact that to_dict().keys() will differ from vars(cls()).keys() | 01:45 |
amrith | fun ... | 01:45 |
amrith | and I have to do it before RC1 | 01:45 |
harlowja | if u ever see `vars(cls()).keys()` just don't do it :) | 01:46 |
amrith | is that part of a Nike commercial? | 01:47 |
harlowja | should be, ha | 01:47 |
harlowja | overall yes though, we need to handle versioning this better | 01:47 |
harlowja | or make it so that older versions accept/discard the newer versions additions | 01:47 |
harlowja | via something simple like **kwargs that captures all of those | 01:47 |
amrith | ++ | 01:48 |
amrith | is it LP'worthy? | 01:48 |
harlowja | sure | 01:49 |
amrith | AHA | 01:49 |
amrith | here's a hack for you | 01:49 |
amrith | I can look at RequestContext().__init__()'s arguments and decide what to take out before calling cls(**kwargs) :) | 01:49 |
amrith | is that part of the Nike commercial as well? | 01:50 |
harlowja | ya, that seems fair imho | 01:51 |
amrith | here ya go. https://bugs.launchpad.net/oslo.context/+bug/1661790 | 01:53 |
openstack | Launchpad bug 1661790 in oslo.context "oslo.context's initializer should do a better job handing class instantiation" [Undecided,New] | 01:53 |
*** browne has quit IRC | 01:59 | |
*** kgiusti has left #openstack-oslo | 02:06 | |
openstackgerrit | lihong proposed openstack/oslo.messaging master: [zmq] Dummy value aging mechanism https://review.openstack.org/428663 | 02:08 |
openstackgerrit | xhzhf proposed openstack/taskflow master: python3.0 has disable LOG.warn https://review.openstack.org/412174 | 02:11 |
amrith | harlowja, have a good evening! thanks for listening :) | 02:15 |
harlowja | np :) | 02:15 |
amrith | I have the whole day ahead of me to fix this ... while y'all take a nap. | 02:16 |
amrith | I will tag you on the review | 02:16 |
harlowja | cools | 02:16 |
*** yamamoto has joined #openstack-oslo | 02:18 | |
*** yamamoto has quit IRC | 02:21 | |
*** e0ne has joined #openstack-oslo | 02:30 | |
*** e0ne has quit IRC | 02:39 | |
*** yamahata has quit IRC | 03:23 | |
*** e0ne has joined #openstack-oslo | 03:35 | |
*** e0ne has quit IRC | 03:40 | |
*** spzala has quit IRC | 04:16 | |
*** nicolasbock has quit IRC | 04:17 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/tooz master: Updated from global requirements https://review.openstack.org/346782 | 04:25 |
*** spzala has joined #openstack-oslo | 04:34 | |
*** spzala has quit IRC | 04:36 | |
*** e0ne has joined #openstack-oslo | 04:36 | |
*** e0ne has quit IRC | 04:41 | |
*** e0ne has joined #openstack-oslo | 04:46 | |
*** lpetrut has joined #openstack-oslo | 04:49 | |
*** e0ne has quit IRC | 04:55 | |
*** e0ne has joined #openstack-oslo | 04:56 | |
*** e0ne has quit IRC | 04:56 | |
*** e0ne has joined #openstack-oslo | 04:57 | |
*** e0ne has quit IRC | 04:57 | |
*** harlowja has quit IRC | 05:08 | |
*** amotoki has joined #openstack-oslo | 05:15 | |
*** lpetrut has quit IRC | 05:16 | |
*** amotoki has quit IRC | 05:20 | |
*** lpetrut has joined #openstack-oslo | 05:36 | |
*** lpetrut has quit IRC | 05:40 | |
*** Rocky_g has quit IRC | 05:48 | |
*** harlowja has joined #openstack-oslo | 06:17 | |
*** salv-orlando has joined #openstack-oslo | 06:18 | |
*** salv-orlando has quit IRC | 06:22 | |
*** salv-orlando has joined #openstack-oslo | 06:42 | |
*** harlowja has quit IRC | 06:50 | |
*** amotoki has joined #openstack-oslo | 07:16 | |
*** amotoki has quit IRC | 07:21 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.messaging master: Test:Use unittest.mock on Python 3 https://review.openstack.org/429196 | 07:45 |
*** eezhova has joined #openstack-oslo | 07:48 | |
*** salv-orl_ has joined #openstack-oslo | 07:49 | |
*** harlowja has joined #openstack-oslo | 07:50 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.messaging master: Move decorator updated_kwarg_default_value to right place https://review.openstack.org/428693 | 07:50 |
*** salv-orlando has quit IRC | 07:51 | |
*** harlowja has quit IRC | 07:57 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.messaging master: Test:Use unittest.mock on Python 3 https://review.openstack.org/429196 | 08:17 |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/oslo.messaging master: Move decorator updated_kwarg_default_value to right place https://review.openstack.org/428693 | 08:17 |
*** eezhova has quit IRC | 08:17 | |
*** zzzeek has quit IRC | 09:00 | |
*** zzzeek has joined #openstack-oslo | 09:00 | |
*** amotoki has joined #openstack-oslo | 09:04 | |
*** amotoki has quit IRC | 09:08 | |
*** sheeprine has quit IRC | 09:37 | |
*** spzala has joined #openstack-oslo | 09:37 | |
*** sheeprine has joined #openstack-oslo | 09:41 | |
*** spzala has quit IRC | 09:42 | |
*** spzala has joined #openstack-oslo | 09:48 | |
*** spzala has quit IRC | 09:53 | |
*** lpetrut has joined #openstack-oslo | 10:06 | |
*** yamahata has joined #openstack-oslo | 10:42 | |
*** lpetrut has quit IRC | 10:51 | |
*** amotoki has joined #openstack-oslo | 11:03 | |
*** amotoki has quit IRC | 11:07 | |
*** yamahata has quit IRC | 11:13 | |
*** nicolasbock has joined #openstack-oslo | 11:24 | |
*** eezhova has joined #openstack-oslo | 12:00 | |
*** salv-orl_ has quit IRC | 12:07 | |
*** kberger has quit IRC | 12:16 | |
*** e0ne has joined #openstack-oslo | 12:19 | |
*** e0ne has quit IRC | 12:24 | |
*** e0ne has joined #openstack-oslo | 12:25 | |
*** gcb has quit IRC | 12:34 | |
*** e0ne has quit IRC | 12:36 | |
*** amotoki has joined #openstack-oslo | 13:04 | |
*** amotoki has quit IRC | 13:08 | |
*** lucas-afk has quit IRC | 13:46 | |
*** mriedem has joined #openstack-oslo | 13:49 | |
*** amotoki has joined #openstack-oslo | 13:53 | |
*** salv-orlando has joined #openstack-oslo | 14:00 | |
*** spzala has joined #openstack-oslo | 14:03 | |
*** salv-orlando has quit IRC | 15:14 | |
*** ansmith has joined #openstack-oslo | 15:32 | |
*** nicolasbock has quit IRC | 16:10 | |
*** kberger has joined #openstack-oslo | 16:16 | |
*** ansmith has quit IRC | 16:20 | |
*** ansmith has joined #openstack-oslo | 16:31 | |
*** ansmith has quit IRC | 16:54 | |
*** ansmith has joined #openstack-oslo | 17:07 | |
*** yamahata has joined #openstack-oslo | 17:09 | |
*** HenryG has quit IRC | 17:17 | |
*** salv-orlando has joined #openstack-oslo | 17:31 | |
*** salv-orlando has quit IRC | 17:36 | |
*** zz_dimtruck is now known as dimtruck | 17:39 | |
*** HenryG has joined #openstack-oslo | 17:39 | |
*** HenryG has quit IRC | 17:51 | |
*** HenryG has joined #openstack-oslo | 17:55 | |
*** kberger has quit IRC | 18:32 | |
*** kberger has joined #openstack-oslo | 18:32 | |
*** dimtruck is now known as zz_dimtruck | 18:34 | |
*** salv-orlando has joined #openstack-oslo | 18:34 | |
*** salv-orlando has quit IRC | 18:39 | |
*** zz_dimtruck is now known as dimtruck | 18:43 | |
*** e0ne has joined #openstack-oslo | 19:13 | |
*** e0ne has quit IRC | 19:17 | |
*** salv-orlando has joined #openstack-oslo | 19:35 | |
*** salv-orlando has quit IRC | 19:39 | |
*** dimtruck is now known as zz_dimtruck | 19:40 | |
*** yamahata has quit IRC | 20:12 | |
*** zz_dimtruck is now known as dimtruck | 20:25 | |
*** salv-orlando has joined #openstack-oslo | 20:36 | |
*** salv-orlando has quit IRC | 20:40 | |
*** spzala has quit IRC | 20:55 | |
*** dimtruck is now known as zz_dimtruck | 21:00 | |
*** david-lyle_ has joined #openstack-oslo | 21:02 | |
*** kberger has quit IRC | 21:04 | |
*** david-lyle has quit IRC | 21:05 | |
*** lpetrut has joined #openstack-oslo | 21:07 | |
*** oanson has quit IRC | 21:13 | |
*** oanson has joined #openstack-oslo | 21:13 | |
*** ansmith has quit IRC | 21:32 | |
*** ansmith has joined #openstack-oslo | 22:13 | |
*** eezhova has quit IRC | 22:37 | |
*** salv-orlando has joined #openstack-oslo | 22:38 | |
*** salv-orlando has quit IRC | 22:42 | |
*** spzala has joined #openstack-oslo | 22:56 | |
*** zz_dimtruck is now known as dimtruck | 22:56 | |
*** amotoki has quit IRC | 22:59 | |
*** spzala has quit IRC | 23:00 | |
*** david-lyle_ has quit IRC | 23:05 | |
*** pkoniszewski has quit IRC | 23:06 | |
*** lpetrut has quit IRC | 23:06 | |
*** pkoniszewski has joined #openstack-oslo | 23:08 | |
*** dulek has quit IRC | 23:29 | |
*** dulek has joined #openstack-oslo | 23:32 | |
*** salv-orlando has joined #openstack-oslo | 23:38 | |
*** salv-orlando has quit IRC | 23:43 | |
*** salv-orlando has joined #openstack-oslo | 23:48 | |
*** salv-orl_ has joined #openstack-oslo | 23:50 | |
*** salv-orlando has quit IRC | 23:53 | |
*** salv-orl_ has quit IRC | 23:54 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!