| *** flaper87 is now known as flaper87|afk | 00:05 | |
| harlowja | anyone with any oslo.messaging experience around | 00:15 |
|---|---|---|
| *** rpodolyaka has quit IRC | 00:17 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/oslo.messaging: Add a zookeeper based matchmaker https://review.openstack.org/79825 | 00:22 |
| *** rpodolyaka has joined #openstack-oslo | 00:27 | |
| morganfainberg | dhellmann, ping re https://review.openstack.org/#/c/72291/3/openstack/common/cache/memoization.py, your comment about objects is related to the decorator function (line 85) or the memoize function? | 00:32 |
| *** arnaud has quit IRC | 00:39 | |
| *** arnaud has joined #openstack-oslo | 00:39 | |
| dims_ | harlowja, is kazoo being used in any other project? (just see it in requirements/global-requirements.txt but nowhere else) | 00:41 |
| harlowja | dims_ taskflow at least, stackforge/tooz most defintly, nova uses a out of date evzookeeper, there was a review up for moving it to use kazoo instead for https://github.com/openstack/nova/blob/master/nova/servicegroup/drivers/zk.py | 00:43 |
| harlowja | but not sure where that review went | 00:43 |
| harlowja | it seems the service group code in nova is similar to the matchmaker code (also similar to tooz code doing similar stuff) | 00:44 |
| dims_ | harlowja, i see. also taskflow seems to need >=1.3.1 (https://github.com/openstack/taskflow/blob/master/optional-requirements.txt#L22) | 00:44 |
| harlowja | it can likely work with older versions | 00:45 |
| harlowja | how old, not really sure | 00:45 |
| harlowja | https://pypi.python.org/pypi/kazoo#changelog | 00:46 |
| dims_ | just thinking if requirements/global-requirements.txt needs to be bumped up | 00:46 |
| harlowja | hmmm, thought it did/was, or i had a review for that, where'd it go, lol | 00:46 |
| * harlowja finding | 00:47 | |
| dims_ | found the kazoo one - https://review.openstack.org/#/c/28951/ | 00:47 |
| harlowja | k, hmmm, ya, seems like it was mostly done | 00:48 |
| dims_ | looks like test case comments were not followed up on - https://review.openstack.org/#/c/28951/17/nova/tests/servicegroup/test_kazoo_driver.py,unified | 00:49 |
| harlowja | ah, dims_ https://review.openstack.org/#/c/69815/ | 00:49 |
| harlowja | so that was the unpin | 00:49 |
| harlowja | shall i bump | 00:49 |
| harlowja | i'm ok with that | 00:50 |
| dims_ | +1 to bump up global. thanks! | 00:51 |
| harlowja | k | 00:51 |
| harlowja | https://review.openstack.org/#/c/79796/ was another weird one to | 00:51 |
| harlowja | pip requirements checking/validation doesn't seem so great, lol | 00:52 |
| dims_ | lol :) | 00:52 |
| harlowja | ya, i made a updated an anvil tool that correctly identifies these kind of weirdnesses and after adding some functionality was wondering what was happening | 00:54 |
| harlowja | https://review.openstack.org/#/c/79203/ | 00:54 |
| harlowja | anyways | 00:54 |
| harlowja | dims_ https://review.openstack.org/80134 | 00:55 |
| dims_ | harlowja, another one, i was checking if the matchmakers were pluggable and listed in setup.cfg looks like they are not | 00:55 |
| harlowja | dims_ ya, that makes it hard to discover these | 00:55 |
| harlowja | i am wondering if oslo.messaging should sort of adopt the tooz group membership model and replace the matchmaker stuff with it | 00:56 |
| harlowja | groups seems like a better abstraction (if it fits) | 00:56 |
| harlowja | https://github.com/stackforge/tooz/blob/master/tooz/coordination.py#L26 | 00:56 |
| dims_ | harlowja, dunno, markmc may have a better idea about that. | 00:57 |
| harlowja | in a way the matchmaker stuff is doign group memebership (where the group is a set of hosts in a topic, or fanout) | 00:57 |
| harlowja | jd__ might also, think some folks from evonance are doing some of the tooz work | 00:57 |
| harlowja | dims_ let me see if someone at at&t (i believe they are using the nova + zookeeper driver) or maybe myself can finish https://review.openstack.org/#/c/28951/ | 01:01 |
| harlowja | it'd be nice to get off https://pypi.python.org/pypi/evzookeeper which doesn't appear updated in 2 years | 01:01 |
| harlowja | kazoo at least is actively maintained (by twitter i think?) | 01:02 |
| dims_ | cool. fyi https://bugs.launchpad.net/oslo/+bug/1291701 please add your thoughts as well | 01:03 |
| dims_ | brb | 01:04 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Avoid calling callbacks while holding locks https://review.openstack.org/80138 | 01:25 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Avoid calling callbacks while holding locks https://review.openstack.org/80138 | 01:26 |
| *** rpodolyaka has quit IRC | 01:58 | |
| openstackgerrit | lifeless proposed a change to openstack-dev/pbr: Return the real class in VersionInfo __repr__. https://review.openstack.org/79856 | 02:00 |
| openstackgerrit | lifeless proposed a change to openstack-dev/pbr: Fix up some docstrings. https://review.openstack.org/79855 | 02:00 |
| *** mriedem has left #openstack-oslo | 02:05 | |
| *** mriedem has quit IRC | 02:05 | |
| *** rpodolyaka has joined #openstack-oslo | 02:08 | |
| *** rpodolyaka has quit IRC | 02:21 | |
| *** hdd_ has joined #openstack-oslo | 02:25 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Update lock_utils to use the posix IPC lock https://review.openstack.org/80155 | 02:25 |
| *** arnaud has quit IRC | 02:26 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Update comment + six.text_type instead of str for name https://review.openstack.org/80158 | 02:35 |
| *** rpodolyaka has joined #openstack-oslo | 02:53 | |
| *** rpodolyaka has quit IRC | 02:54 | |
| *** rpodolyaka has joined #openstack-oslo | 02:54 | |
| openstackgerrit | lifeless proposed a change to openstack-dev/pbr: TEST: Deleting thought-to-be-dead code. https://review.openstack.org/80161 | 02:55 |
| *** gcb_ has joined #openstack-oslo | 02:55 | |
| *** gcb has quit IRC | 02:55 | |
| *** gcb__ has joined #openstack-oslo | 02:56 | |
| *** saikrishna_ has joined #openstack-oslo | 02:58 | |
| *** gcb_ has quit IRC | 03:00 | |
| *** zhiyan_ is now known as zhiyan | 03:06 | |
| *** saikrishna_ has quit IRC | 03:17 | |
| openstackgerrit | Cyril Roelandt proposed a change to openstack-dev/hacking: Add a six-related checker https://review.openstack.org/80171 | 03:34 |
| *** saikrishna_ has joined #openstack-oslo | 03:45 | |
| *** saikrishna_ has quit IRC | 03:48 | |
| *** rpodolyaka has quit IRC | 03:51 | |
| *** lbragstad1 has joined #openstack-oslo | 03:57 | |
| *** lbragstad has quit IRC | 03:59 | |
| *** lbragstad1 is now known as lbragstad | 04:07 | |
| *** lbragstad is now known as lbragstad__ | 04:16 | |
| *** harlowja is now known as harlowja_away | 04:20 | |
| *** rpodolyaka has joined #openstack-oslo | 04:22 | |
| *** harlowja_away is now known as harlowja | 04:24 | |
| *** harlowja is now known as harlowja_away | 05:53 | |
| *** hdd_ has quit IRC | 05:55 | |
| openstackgerrit | Zhi Yan Liu proposed a change to openstack/oslo-incubator: Making table utf-8 charset checking be optional for DB migration https://review.openstack.org/75356 | 06:18 |
| *** yamahata has quit IRC | 06:21 | |
| *** gcb__ has quit IRC | 07:10 | |
| *** rpodolyaka has quit IRC | 07:12 | |
| *** gcb__ has joined #openstack-oslo | 07:12 | |
| *** gcb_ has joined #openstack-oslo | 07:13 | |
| *** gcb__ has quit IRC | 07:17 | |
| *** morganfainberg is now known as morganfainberg_Z | 07:18 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/oslo-incubator: Re-raise exception of unloadable library https://review.openstack.org/77934 | 07:43 |
| *** gcb__ has joined #openstack-oslo | 07:59 | |
| *** gcb_ has quit IRC | 08:02 | |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Ingore 'taskflow.' prefix when sorting module index https://review.openstack.org/80205 | 08:09 |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Relative links improvements https://review.openstack.org/80206 | 08:09 |
| *** flaper87|afk is now known as flaper87 | 08:13 | |
| *** bogdando has joined #openstack-oslo | 08:15 | |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: Remove extra quote symbol https://review.openstack.org/80211 | 08:32 |
| *** I159_ has joined #openstack-oslo | 08:37 | |
| openstackgerrit | Oleg Bondarev proposed a change to openstack/oslo-incubator: save_and_reraise_exception: add a flag to omit logging original exception https://review.openstack.org/80222 | 09:34 |
| openstackgerrit | Oleg Bondarev proposed a change to openstack/oslo-incubator: save_and_reraise_exception: add a flag to omit logging https://review.openstack.org/80222 | 10:04 |
| *** sergmelikyan has joined #openstack-oslo | 10:24 | |
| sergmelikyan | Hi, guys. openstack.messaging has openstack.common.ssutils module imported, same is done by murano, that also used oslo.messaging. As expected DuplicateOptError exception is raised in oslo.config. How do you deal with such cases? | 10:26 |
| sileht | sergmelikyan, update oslo.messaging to latest 1.3.0aX or sync the oslo-incubator sslutils module | 10:49 |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Add page describing atom arguments and results https://review.openstack.org/80242 | 10:53 |
| sergmelikyan | sileht, thx! | 10:58 |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Improve BaseTask method docstrings https://review.openstack.org/80248 | 11:08 |
| sergmelikyan | sileht, What do you think about updating requirement for oslo.messaging in global-requirements? | 11:10 |
| sergmelikyan | I think requirement should be updated in IceHouse, since it is quite important fix | 11:11 |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Add page describing atom arguments and results https://review.openstack.org/80242 | 11:13 |
| *** gcb__ has quit IRC | 11:23 | |
| *** gcb__ has joined #openstack-oslo | 11:25 | |
| *** gcb_ has joined #openstack-oslo | 11:28 | |
| *** viktors_ is now known as viktors | 11:29 | |
| *** gcb__ has quit IRC | 11:32 | |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Add page describing atom arguments and results https://review.openstack.org/80242 | 11:48 |
| sergmelikyan | https://review.openstack.org/80259 - guys, what do you think about this patch? | 12:04 |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Add page describing atom arguments and results https://review.openstack.org/80242 | 12:06 |
| *** beav has quit IRC | 12:07 | |
| *** dims_ has quit IRC | 12:30 | |
| *** dims_ has joined #openstack-oslo | 12:34 | |
| *** arnaud has joined #openstack-oslo | 12:55 | |
| *** yamahata has joined #openstack-oslo | 13:03 | |
| *** mriedem has joined #openstack-oslo | 13:12 | |
| *** lbragstad__ has quit IRC | 13:19 | |
| *** beav has joined #openstack-oslo | 13:29 | |
| openstackgerrit | Nikola Dipanov proposed a change to openstack-dev/hacking: Add TestCase assertion methods for testing checks https://review.openstack.org/80278 | 13:30 |
| openstackgerrit | Nikola Dipanov proposed a change to openstack-dev/hacking: Add an import whitelist helper factory https://review.openstack.org/80018 | 13:30 |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Add page describing atom arguments and results https://review.openstack.org/80242 | 13:33 |
| *** lbragstad has joined #openstack-oslo | 13:47 | |
| flaper87 | hey guys, can anyone help me replicate this failure ? http://logs.openstack.org/58/76458/4/check/gate-oslo-incubator-python27/e6be562/console.html | 14:15 |
| flaper87 | I can't replicate it in my box | 14:15 |
| flaper87 | so, not sure what's going on there | 14:16 |
| *** hdd_ has joined #openstack-oslo | 14:23 | |
| openstackgerrit | Stanislav Kudriashev proposed a change to openstack/taskflow: Introduce message types for WBE protocol https://review.openstack.org/79712 | 14:37 |
| *** hdd_ has quit IRC | 14:46 | |
| *** browne has joined #openstack-oslo | 14:52 | |
| *** hartsocks has joined #openstack-oslo | 15:01 | |
| dhellmann | flaper87: I haven't seen that one before | 15:04 |
| openstackgerrit | Stanislav Kudriashev proposed a change to openstack/taskflow: [WBE] Collect information from workers https://review.openstack.org/79713 | 15:07 |
| dhellmann | morganfainberg_Z: I'm not sure which comment you mean? | 15:08 |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/oslo-incubator: Do not use the 'extend' method on a dict_items object https://review.openstack.org/80307 | 15:09 |
| flaper87 | dhellmann: :/ I'll keep digging, it's not related to the patch for sure | 15:10 |
| flaper87 | (says flaper87 before finding out it's his fault) | 15:10 |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/oslo-incubator: Do not use the 'extend' method on a dict_items object https://review.openstack.org/80307 | 15:16 |
| *** rpodolyaka has joined #openstack-oslo | 15:18 | |
| *** rpodolyaka has left #openstack-oslo | 15:35 | |
| openstackgerrit | Luis de Bethencourt proposed a change to openstack/oslo.config: deprecated cli option has _ rather than - https://review.openstack.org/80319 | 15:43 |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/oslo-incubator: Python3: do not use contextlib.nested() https://review.openstack.org/80321 | 15:46 |
| openstackgerrit | Stanislav Kudriashev proposed a change to openstack/taskflow: Update request `expired` property docsting https://review.openstack.org/80322 | 15:54 |
| zhiyan | dhellmann: hello, not sure you could take a look on my response at https://bugs.launchpad.net/glance/+bug/1279000 and let me know position? the issue is really a blocker now for glance this release. | 15:55 |
| zhiyan | thanks | 15:55 |
| *** arnaud has quit IRC | 16:09 | |
| openstackgerrit | Stanislav Kudriashev proposed a change to openstack/taskflow: Introduce message types for WBE protocol https://review.openstack.org/79712 | 16:12 |
| dhellmann | zhiyan: I'm afraid I don't know the right answer to this problem. It sounds like the table encoding can't just be changed because of the current encoding, but it seems like allowing migrations to take place before that problem is resolved only makes the issue worse -- possibly creating new tables and adding more data. | 16:12 |
| dhellmann | zhiyan: If the switch that disables the check is meant to let deployers run migrations before fixing the problem, why don't we just have them fix the problem first? | 16:15 |
| *** rpodolyaka has joined #openstack-oslo | 16:19 | |
| *** bogdando has quit IRC | 16:22 | |
| *** bogdando has joined #openstack-oslo | 16:24 | |
| zhiyan | dhellmann: what in you mind to ask deployer fix the problem first? be manually? | 16:24 |
| zhiyan | dhellmann: i totally understand your concerns and agree what you proposed: to have deploy fix the table charset problem first then to do other migration. But i think the key is that IMO we'd better allow migration script fix it automatically, so this idea need that switch. | 16:28 |
| zhiyan | s/deploy/deployer | 16:29 |
| dhellmann | zhiyan: Based on something said on the mailing list, I thought a migration couldn't fix this because it might corrupt the existing data. | 16:29 |
| dhellmann | zhiyan: To be honest, I'm not a MySQL expert. We do have experts in the community, so maybe you can get a clear path defined as part of that mailing list thread with their help? | 16:30 |
| zhiyan | dhellmann: i'm confused, iiuc, we need a migration to fix that issue in anyway. | 16:32 |
| zhiyan | dhellmann: ok, i will prepare a short mail to get more input. btw, could you pls give me the ML link? or its title..i'm not sure i know that clearly. | 16:33 |
| dhellmann | zhiyan: give me a minute and I'll find that ML message | 16:33 |
| zhiyan | thanks | 16:33 |
| bnemec | zhiyan: According to Clint's post, we can't do this in a migration safely: http://lists.openstack.org/pipermail/openstack-dev/2014-March/029708.html | 16:33 |
| zhiyan | ah, recalled | 16:34 |
| dhellmann | bnemec: thanks, that's the one | 16:34 |
| bnemec | Unfortunately this is just a very tricky problem from what I'm hearing. | 16:34 |
| dhellmann | yeah | 16:34 |
| bnemec | Maybe we could complain loudly if someone does a migration with the wrong encoding, but provide the option to continue anyway. | 16:35 |
| zhiyan | bnemec: hum..so are you saying deployer need recreate whole table and dump all rows back? | 16:35 |
| dhellmann | so it seems like the only correct solution is to have deployers fix their databases before any more migrations are run | 16:35 |
| *** saikrishna_ has joined #openstack-oslo | 16:36 | |
| bnemec | zhiyan: I'm honestly not sure. I think it depends on the data and the encodings involved, which is why this is tricky. | 16:36 |
| zhiyan | i'm very interested in the steps of how to fix their database for deployers | 16:36 |
| zhiyan | you know, even we deiced to fix this issue by document, we still need to give deployer a advising | 16:37 |
| bnemec | zhiyan: Right, but it's not clear to me how to fix the "latin1 db with utf8 data inserted" problem that Clint raised. | 16:39 |
| bnemec | I suppose recreating the table from scratch could work, but if there are foreign keys on it I think that could be difficult. | 16:39 |
| zhiyan | very difficult | 16:39 |
| bnemec | Basically it's very, very bad that we ever let this happen because people's existing dbs may already be corrupted. | 16:40 |
| zhiyan | nod | 16:40 |
| zhiyan | back to the topic, i will talk this with markwash later and know his comments. anyway, i think you currently trend to like deployer do the fix by themself. | 16:42 |
| bnemec | I agree that we need good documentation of the problem to let people proceed as best they can, but outside of Clint I'm not sure who would be qualified to write it. | 16:42 |
| zhiyan | and if so, this bug will go to invalid. | 16:42 |
| zhiyan | bnemec: if so do you think we can add the auto fix for this issue as a step in oslo db_sync? which code logic follow Clint's solution? | 16:44 |
| bnemec | zhiyan: I don't think we can auto-fix the problem at all. We could provide a fix that would go through and change all the tables to utf8 (I think that's what nova did at some point in the past), but it doesn't sound like that will work right in every case. | 16:46 |
| bnemec | AIUI anyway | 16:46 |
| zhiyan | bnemec: tbh i'm confused at here, if you understand correct, you are saying glance project can make a built-in migration script to (partial, probably) fix this issue by changing all the tables to utf8, right? that just is what I proposed here.. | 16:49 |
| dhellmann | zhiyan: automating the solution to this problem will be quite complex | 16:50 |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/oslo-incubator: Python 3: enable tests/unit/middleware/test_request_id.py https://review.openstack.org/80336 | 16:50 |
| dhellmann | zhiyan: this is why database administrators exist -- we need to explain to them what their database should look like, but I don't think it's safe to try to write a migration to fix the database because we can't be sure of the starting conditions | 16:51 |
| zhiyan | dhellmann: agreed, i totally understand. | 16:51 |
| zhiyan | dhellmann: do you agree we fix it by a document? | 16:52 |
| dhellmann | zhiyan: yes, I think that's the best way to handle it, in the release notes | 16:53 |
| zhiyan | in project's release notes, like glance? or a common one, not sure | 16:53 |
| bnemec | Yeah, we could write a tool to handle the trivial case, but the more complex ones need manual intervention. | 16:54 |
| openstackgerrit | Luis de Bethencourt proposed a change to openstack/oslo.config: deprecated cli option has _ rather than - https://review.openstack.org/73694 | 16:54 |
| dhellmann | bnemec: right, and everyone would want the tool to handle every case :-/ | 16:56 |
| dhellmann | zhiyan: that's a question for annegentle, but maybe both locations | 16:56 |
| bnemec | dhellmann: And that way lies madness :-) | 16:56 |
| dhellmann | certainly this needs to be documented in the upgrade "manual" | 16:57 |
| dhellmann | bnemec: openstack is madness all the way down | 16:57 |
| bnemec | Indeed. :-) | 16:58 |
| zhiyan | dhellmann: bnemec thanks for your input btw | 16:59 |
| *** arnaud has joined #openstack-oslo | 17:01 | |
| *** viktors is now known as viktors_ | 17:06 | |
| *** morganfainberg_Z is now known as morganfainberg | 17:09 | |
| *** harlowja_away is now known as harlowja | 17:10 | |
| *** saikrishna_ has quit IRC | 17:15 | |
| *** lbragstad has quit IRC | 17:17 | |
| openstackgerrit | Anastasia Karpinska proposed a change to openstack/taskflow: Move flow states to developers socumentation https://review.openstack.org/80350 | 17:22 |
| openstackgerrit | Anastasia Karpinska proposed a change to openstack/taskflow: Add retry action to execute retries https://review.openstack.org/68112 | 17:22 |
| openstackgerrit | Anastasia Karpinska proposed a change to openstack/taskflow: Flow smart revert with retry controller https://review.openstack.org/71621 | 17:22 |
| openstackgerrit | Anastasia Karpinska proposed a change to openstack/taskflow: Add atom intentions for tasks and retries https://review.openstack.org/71013 | 17:22 |
| openstackgerrit | Anastasia Karpinska proposed a change to openstack/taskflow: Extend logbook and storage to work with retry https://review.openstack.org/68111 | 17:22 |
| *** lbragstad has joined #openstack-oslo | 17:23 | |
| *** hartsocks has left #openstack-oslo | 17:37 | |
| *** hartsocks has joined #openstack-oslo | 17:37 | |
| harlowja | dhellmann lol, openstack is madness all the way down ;) | 17:37 |
| harlowja | lol | 17:37 |
| dims_ | continuously tested madness :) | 17:37 |
| harlowja | ;) | 17:37 |
| morganfainberg | dhellmann, the one on line 85 | 17:38 |
| harlowja | dims_ mr.jenkins must be a very tired man, lol | 17:38 |
| dims_ | :) | 17:38 |
| morganfainberg | dhellmann, https://review.openstack.org/#/c/72291/3/openstack/common/cache/memoization.py should the whole decorator (memoize) me be a class? or just the "decorator" function | 17:38 |
| morganfainberg | dhellmann, i wasn't sure which you meant. | 17:38 |
| dhellmann | morganfainberg: I think I meant memoize() should be a class, let me verify | 17:41 |
| morganfainberg | dhellmann, i thought so, but i wanted to be 100% sure. | 17:41 |
| dhellmann | morganfainberg: yeah, that way we can document invalidate() etc. as methods of memoize | 17:42 |
| dhellmann | as they stand now, there's no way to discover those names to document them automatically | 17:42 |
| morganfainberg | dhellmann, i'll need to look into that, there is some oddities based on how "self" gets eaten in those cases | 17:42 |
| morganfainberg | but i think i can do it | 17:42 |
| dhellmann | morganfainberg: also, regarding the API, I wonder if it would be better match lru_cache than dogpile, since the latter is a stdlib "standard" | 17:42 |
| morganfainberg | lru_cache is a wholesale cache clear for the method | 17:43 |
| morganfainberg | afaict | 17:43 |
| morganfainberg | dogpile is true memoization | 17:43 |
| dhellmann | morganfainberg: there was a good thread on planet.python.org recently about writing decorators | 17:43 |
| morganfainberg | so arguments are the key. | 17:43 |
| morganfainberg | dhellmann, thanks! i'll check out that thread | 17:43 |
| dhellmann | morganfainberg: http://blog.dscpl.com.au/2014/01/how-you-implemented-your-python.html | 17:44 |
| morganfainberg | dhellmann, thanks! | 17:45 |
| dhellmann | morganfainberg: as far as the API, I just meant cache_clear() instead of invalidate() | 17:46 |
| morganfainberg | dhellmann, ahh | 17:46 |
| morganfainberg | dhellmann, i'll just alias both | 17:46 |
| dhellmann | morganfainberg: and maybe add cache_info()? | 17:46 |
| morganfainberg | dhellmann, that is easy | 17:46 |
| morganfainberg | dhellmann, hmm. sure. | 17:46 |
| morganfainberg | dhellmann, both good ideas. | 17:46 |
| dhellmann | morganfainberg: I'd have to read the docs again closely to see if there are other appropriate changes, but you get the idea | 17:46 |
| morganfainberg | dhellmann, sure, the lru_cache interfaces seem sane | 17:47 |
| morganfainberg | dhellmann, i'm also pulling the crypto stuff out so that can be specifically reviewed by simo and other more crypto-facing people | 17:48 |
| morganfainberg | independantly of the core changeset | 17:48 |
| dhellmann | morganfainberg: ok, that sounds good | 17:48 |
| morganfainberg | dhellmann, :) it's slow work because it's a big changeset, but it's moving. | 17:49 |
| dhellmann | morganfainberg: we're basically frozen now anyway, right? | 17:49 |
| morganfainberg | dhellmann, aye | 17:49 |
| morganfainberg | dhellmann, no one would sync it in | 17:49 |
| morganfainberg | dhellmann, but i want it ready for J1, if i can have keystone's cache and KVS systems moved over to it in Juno i'll be pleased | 17:50 |
| morganfainberg | and then i can help other projects leverage it as well | 17:50 |
| *** beav- has joined #openstack-oslo | 17:52 | |
| *** beav has quit IRC | 17:53 | |
| *** arnaud has quit IRC | 17:53 | |
| *** rpodolyaka has quit IRC | 17:57 | |
| *** rpodolyaka has joined #openstack-oslo | 18:00 | |
| *** flaper87 is now known as flaper87IsNotAwa | 18:04 | |
| *** flaper87IsNotAwa is now known as flaper87DoesNotK | 18:04 | |
| *** flaper87DoesNotK is now known as flaper87WTF | 18:05 | |
| *** flaper87WTF is now known as flaper87 | 18:05 | |
| dhellmann | morganfainberg: +1 | 18:06 |
| openstackgerrit | Ivan Melnikov proposed a change to openstack/taskflow: docs: Add page describing atom arguments and results https://review.openstack.org/80242 | 18:09 |
| *** zhiyan is now known as zhiyan_ | 18:10 | |
| openstackgerrit | Luis de Bethencourt proposed a change to openstack/oslo.config: deprecated cli option has _ rather than - https://review.openstack.org/73694 | 18:12 |
| *** I159_ has quit IRC | 18:13 | |
| *** dshulyak_ has joined #openstack-oslo | 18:30 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack/oslo.messaging: Full support of multiple hosts in transport url https://review.openstack.org/78948 | 18:37 |
| openstackgerrit | Cyril Roelandt proposed a change to openstack/oslo-incubator: Python3: enable test_jsonutils.py https://review.openstack.org/80370 | 19:05 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Introduce a CorrelatedFailure exception https://review.openstack.org/80372 | 19:11 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Introduce a CorrelatedFailure exception https://review.openstack.org/80372 | 19:12 |
| *** rpodolyaka has quit IRC | 19:13 | |
| *** dshulyak_ has quit IRC | 19:20 | |
| *** hartsocks has left #openstack-oslo | 19:21 | |
| *** hartsocks has joined #openstack-oslo | 19:21 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Introduce a CorrelatedFailure exception https://review.openstack.org/80372 | 19:24 |
| *** rpodolyaka has joined #openstack-oslo | 20:14 | |
| *** rpodolyaka has quit IRC | 20:19 | |
| bnemec | So https://bugs.launchpad.net/oslo/+bug/1285530 | 20:20 |
| bnemec | The mailing list discussion didn't get a lot of love, but I think it's something we need to discuss, especially in the lazy translation world we live in now. | 20:21 |
| *** beav- has quit IRC | 20:25 | |
| *** rpodolyaka has joined #openstack-oslo | 20:32 | |
| dhellmann | bnemec: the logic behind not translating the exceptions eludes me | 20:34 |
| dhellmann | if the translation machinery is "broken" at the time the error is reported, then it just won't be translated | 20:35 |
| dhellmann | by the logic presented, we shouldn't report errors because error reporting may be broken when an error happens | 20:35 |
| bnemec | dhellmann: Couldn't broken translation mask the original error though? | 20:35 |
| dhellmann | bnemec: I'd have to look, but I'm reasonably sure that if translation fails for any reason we just use the original message string | 20:36 |
| bnemec | Although on second thought I'm wondering if it's even possible to raise a translated exception at a point where translation is broken. | 20:36 |
| bnemec | I'd think we'd blow up on a log message or something long before that. | 20:36 |
| dhellmann | bnemec: yeah | 20:37 |
| dhellmann | bnemec: we should add some exception handling in http://git.openstack.org/cgit/openstack/oslo-incubator/tree/openstack/common/gettextutils.py#n176 | 20:38 |
| *** rpodolyaka has quit IRC | 20:39 | |
| dhellmann | bnemec: but otherwise, I think we should go ahead and translate errors and assume that most of the time the error is going to be something related to a user action and not MemoryError or some other catastrophic thing | 20:39 |
| bnemec | dhellmann: Yeah, I think that makes sense. | 20:39 |
| dhellmann | bnemec: I will reply to the email thread | 20:40 |
| bnemec | We should probably update the wiki to basically say "when in Rome, translate exceptions as the Romans do" | 20:40 |
| bnemec | Because I'm not sure _every_ project translates exceptions today, but I know quite a few do. | 20:40 |
| *** rpodolyaka has joined #openstack-oslo | 20:40 | |
| bnemec | dhellmann: Thanks | 20:40 |
| dhellmann | bnemec: sent -- I made one update to the wiki about log translations a while back and forgot to reply to the thread at that point. I will leave the thing about exceptions to give people time to reply to my message. | 20:46 |
| bnemec | dhellmann: Sounds good. | 20:46 |
| * bnemec is pretty sure no one else is going to go anywhere near this topic ;-) | 20:47 | |
| *** browne has left #openstack-oslo | 20:48 | |
| * dhellmann nods | 20:52 | |
| *** gilliard_ has joined #openstack-oslo | 21:05 | |
| gilliard_ | dhellmann: would you mind having another look at https://review.openstack.org/#/c/71559/9 when you have time? | 21:06 |
| dhellmann | gilliard_: there's no priority on that for nova, are you working on it for rc1? | 21:26 |
| *** arnaud has joined #openstack-oslo | 21:27 | |
| *** mriedem has left #openstack-oslo | 21:53 | |
| *** lbragstad has quit IRC | 21:54 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Adjust the exception hierachy https://review.openstack.org/80372 | 21:58 |
| *** hartsocks has quit IRC | 22:04 | |
| *** rpodolyaka has quit IRC | 22:15 | |
| *** rpodolyaka has joined #openstack-oslo | 22:18 | |
| *** arnaud has quit IRC | 22:53 | |
| *** dims_ has quit IRC | 22:58 | |
| *** rpodolyaka has quit IRC | 22:59 | |
| *** rpodolyaka has joined #openstack-oslo | 23:04 | |
| *** mriedem has joined #openstack-oslo | 23:04 | |
| *** mriedem has left #openstack-oslo | 23:05 | |
| *** mriedem has joined #openstack-oslo | 23:07 | |
| *** dims_ has joined #openstack-oslo | 23:12 | |
| *** rpodolyaka has quit IRC | 23:17 | |
| *** rpodolyaka has joined #openstack-oslo | 23:20 | |
| *** flaper87 is now known as flaper87|afk | 23:37 | |
| *** wingwj has quit IRC | 23:45 | |
| *** wingwj has joined #openstack-oslo | 23:46 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!