Monday, 2013-09-23

« Sunday, 2013-09-22 Index Tuesday, 2013-09-24 »
*** ekarlso has quit IRC00:23
*** ekarlso has joined #openstack-dns00:29
*** CaptTofu has joined #openstack-dns01:37
*** msisk has joined #openstack-dns01:47
*** msisk has quit IRC02:54
*** msisk has joined #openstack-dns02:59
*** msisk has quit IRC03:01
*** CaptTofu has quit IRC03:11
*** CaptTofu has joined #openstack-dns03:12
*** zane has quit IRC05:23
*** CaptTofu has quit IRC11:33
*** CaptTofu has joined #openstack-dns11:42
*** CaptTofu has quit IRC11:45
*** CaptTofu has joined #openstack-dns11:46
*** CaptTofu has quit IRC11:49
*** CaptTofu has joined #openstack-dns11:49
*** CaptTofu has quit IRC11:51
*** CaptTofu has joined #openstack-dns11:51
*** CaptTofu has quit IRC12:11
*** CaptTofu has joined #openstack-dns12:11
*** CaptTofu has quit IRC13:15
*** CaptTofu has joined #openstack-dns13:16
*** CaptTofu has quit IRC13:17
*** CaptTofu has joined #openstack-dns13:17
*** msisk has joined #openstack-dns13:19
*** eankutse has joined #openstack-dns13:33
*** zane has joined #openstack-dns14:01
*** CaptTofu has quit IRC14:01
*** CaptTofu has joined #openstack-dns14:02
mugsieinitial docs for the phase one of server pool is at https://wiki.openstack.org/wiki/Designate/Server_Pools14:05
mugsiewill be updating it with more docs for the next while14:05
*** zane1 has joined #openstack-dns14:08
*** zane has quit IRC14:08
*** CaptTofu has quit IRC14:22
*** CaptTofu has joined #openstack-dns14:22
eankutsemugsie: Thx14:24
eankutse:-)14:24
*** CaptTofu has quit IRC14:25
mugsienp14:26
*** CaptTofu has joined #openstack-dns14:26
*** betsy has quit IRC14:26
*** CaptTofu has quit IRC14:31
*** CaptTofu has joined #openstack-dns14:32
*** betsy has joined #openstack-dns15:09
eankutseKiall: regarding https://review.openstack.org/#/c/45078/8. I made suggested changes and moved the check for last server into central. As a result, tests (like designate/tests/test_central/test_service.py") are failing where they used to be able to delete last servers. I tried to modify the test to create more than one server in designate/tests/test_central/test_service.py" but get "server already exists" errors.16:14
eankutseAny ideas on how to proceed on changing those tests?16:14
eankutseOther failures:16:14
eankutsedesignate/tests/test_api/test_v1/test_servers.py", line 202, in test_delete_server16:14
eankutse/tests/test_api/test_v1/test_servers.py", line 227, in test_delete_server_missing16:14
kiallHey eankutse16:15
eankutsehey16:15
kiallSo .. yea.. those tests are going to complain :)16:15
kiallThe trick to creating a second, is to supply a fixture=0/1/2 param to the self.create_server() call16:16
kiall0 == the default..16:16
kiall1 and 2 (or however many samples we have..) will give different values16:16
eankutsek. Thx. :-)16:16
*** vinodmr has joined #openstack-dns16:20
eankutseKiall: For some reason, any attempt to access any fields of old_server such as getting the 'content' field results in the below exception. Could this be due to some db schema mismatch?:17:01
eankutse            old_server = self.session.query(models.Record)\17:01
eankutse                .filter_by(type='NS', designate_id=server['id'])\17:01
eankutse                .first()17:01
eankutse            old_server_name = old_server['content']17:01
eankutse2013-09-23 16:56:13    ERROR [designate.openstack.common.rpc.amqp] Exception during message handling17:01
eankutseTraceback (most recent call last):17:01
eankutse  File "/home/vagrant/designate/designate_src/designate/openstack/common/rpc/amqp.py", line 430, in _process_data17:01
eankutse    rval = self.proxy.dispatch(ctxt, version, method, **args)17:01
eankutse  File "/home/vagrant/designate/designate_src/designate/openstack/common/rpc/dispatcher.py", line 133, in dispatch17:01
eankutse    return getattr(proxyobj, method)(ctxt, **kwargs)17:01
eankutse  File "/home/vagrant/designate/designate_src/designate/central/service.py", line 301, in update_server17:01
eankutse    self.backend.update_server(context, server)17:01
eankutse  File "/usr/lib/python2.7/contextlib.py", line 35, in __exit__17:01
eankutse    self.gen.throw(type, value, traceback)17:01
eankutse  File "/home/vagrant/designate/designate_src/designate/central/service.py", line 42, in wrap_backend_call17:01
eankutse    raise exceptions.Backend('Unknown backend failure: %r' % exc)17:01
eankutseBackend: Unknown backend failure: TypeError("'NoneType' object has no attribute '__getitem__'",)17:01
kiallAh17:19
kiallthose are SQLA ResultProxy objects.. So it would be old_server_name = old_server.content17:20
kiallAlso .. Humm17:23
kiall"NoneType .."17:23
kiallThat means the query isn't finding any matching records in the DB17:24
kiallActually - that first bit is off.. We're getting a models.Record object back ..17:26
kiallIt's that it's not matching any records..17:26
kiall.first() returns the first matching result, or None17:26
eankutseHmm, I have the query that is generated and when I ran that I get the first row of the list from the db so the query seemed to be working:17:27
eankutse2013-09-23 16:56:13     INFO [sqlalchemy.engine.base.Engine] SELECT records.id AS records_id, records.designate_id AS records_designate_id, records.domain_id AS records_domain_id, records.name AS records_name, records.type AS records_type, records.content AS records_content, records.ttl AS records_ttl, records.prio AS records_prio, records.change_date AS records_change_date, records.ordername AS records_ordername, records.auth AS records_auth, records.17:27
eankutseFROM records17:27
eankutseWHERE records.designate_id = ? AND records.type = ?17:27
eankutse LIMIT ? OFFSET ?17:27
kiall.one() returns the only matching result, or raises a NoResultFound exception if there is 0 results, or a MultipleResultsFound if there is more than 1 result17:27
eankutse2013-09-23 16:56:13     INFO [sqlalchemy.engine.base.Engine] ('7738d735dfa644ce8e172ee562c2cf97', 'NS', 1, 0)17:27
kiallHumm - So there is a record in the DB at the time of querying?17:28
kiall(a matching record of course :))17:29
eankutseyes - more than one that match the query17:29
eankutseLike you pointed out, the problem is me trying to access the proxy object as though it was a python object. It works now :-)17:32
kiallReally? Okay .. I thought I was mistaken on that. Heh!17:32
eankutseYou were right :-)17:33
eankutseThx17:33
kiallSending this anyway .. Since I typed it: Humm - that makes no sense! If in the [backend:powerdns] section you set connection_debug = 100, and connection_trace = True, you should get much more verbose info about the whats happening.. Including logging of the actual SQL rows returned..17:33
kiall(It'll all get logged in triplicate.. Never bothered anything enough to fix it!)17:34
kialls/anything/anyone/17:35
eankutseYea - I've been wondering about that as I see that triplicate logging already.17:35
eankutseWhat is causing that?17:35
eankutsedo you know?17:35
kiallWell, I know where 2 of the 3 copies come from.. It's the third that I have no clue ;)17:36
eankutseok. I'll file a but for that.17:36
kiallPython logging handlers are hierarchical, where events propagate up to the "root"17:36
eankutseok. So the logging is probably being done at each of the levels17:37
kiallSo .. SQLA is installing an handler at "sqlalchemy" and the "sqlalchemy.engine.base.Engine" logs pass via these handlers, if they exist:17:37
kiallsqlalchemy.engine.base.Engine17:37
kiallsqlalchemy.engine.base17:37
kiallsqlalchemy.engine17:37
kiallsqlalchemy17:37
kialland "the root handler"17:37
kiallso .. we have logging config for the root handler, and sqla has some at "sqlalchemy".. That explains the duplicate.. The 3rd copy, I've no idea :)17:38
eankutseOk. we'll have to look for it at some point - maybe low priority right now but worth looking into17:39
kiallYea, It's something that would never be used outside of debugging a sql query, so it's been lower than lowest priority!17:39
vinodmrkiall:  I had a process related question - what does it mean to be a part of designate-core?  Are there other groups for designate?17:41
kiallvinodmr: no other groups, and it just means you have approval rights on reviews17:44
kiallWell17:44
kiallthere's designate-milestone as well17:44
kiallI've never looked at what the gives access to though ;)17:45
vinodmris there a group for contributors?17:48
kiallNo, that's just everyone on LP17:51
kiallThe openstack-infra folks set the patterns we follow for all these.. We just "fit in" :)17:52
vinodmrSo do other openstack projects too have 2 similar groups - a core group and a milestone group?17:53
kiallYea - Exactly the same groups :)17:54
vinodmrThanks kiall17:54
*** tsimmons has joined #openstack-dns18:36
*** vinodmr has quit IRC18:45
*** CaptTofu has quit IRC18:49
*** vipul is now known as vipul-away18:49
*** vipul-away is now known as vipul18:50
*** J-_2--- has joined #openstack-dns18:58
*** J-_2--- has left #openstack-dns18:59
*** zane1 has quit IRC18:59
openstackgerritBetsy Luzader proposed a change to stackforge/designate: docs: Correct errors in the Create Record examples  https://review.openstack.org/4764819:12
*** david-lyle has joined #openstack-dns19:17
*** betsy has quit IRC19:41
*** zane has joined #openstack-dns19:49
*** vipul is now known as vipul-away20:13
*** vipul-away is now known as vipul20:13
eankutseKiall: Notice some interesting behavior regarding server ids the way they show up in the UI versus in the database: GET servers/2e12f5fa-463b-4cfd-8499-07350637cc4f will return a server20:45
eankutseBut in the designate database I see 2e12f5fa463b4cfd849907350637cc4f (without the hyphens) is that correct?20:46
eankutseAnd it looks like which of the two types of values gets into the database depends on whether one is using ORM level SQLALchemy queries versus lower level SQL Alchemy Expressions20:47
eankutsethat 2e12f5fa-463b-4cfd-8499-07350637cc4f versus 2e12f5fa463b4cfd849907350637cc4f20:48
*** shakayumi has joined #openstack-dns20:52
*** shakayumi has quit IRC20:53
eankutses/that /that is/21:01
*** msisk has quit IRC21:23
*** CaptTofu has joined #openstack-dns21:55
*** eankutse has quit IRC22:15
*** vipul is now known as vipul-away22:20
*** zane has quit IRC22:20
*** vipul-away is now known as vipul22:22
*** msisk has joined #openstack-dns22:26
*** msisk has quit IRC22:26
*** openstack has joined #openstack-dns22:32
*** ChanServ sets mode: +v openstack22:32
*** tsimmons has quit IRC22:32
*** betsy has joined #openstack-dns22:36
*** shakayumi has quit IRC22:38
*** betsy has quit IRC22:45
*** vipul is now known as vipul-away22:51
*** vipul-away is now known as vipul23:01
*** betsy has joined #openstack-dns23:25
*** yidclare has joined #openstack-dns23:50
*** matsuhashi has joined #openstack-dns23:50
« Sunday, 2013-09-22 Index Tuesday, 2013-09-24 »

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