*** kfox1111 has quit IRC | 00:50 | |
*** johnbelamaric has joined #openstack-dns | 00:57 | |
*** johnbelamaric has quit IRC | 00:58 | |
*** mariusv has quit IRC | 01:00 | |
*** rektide has quit IRC | 01:02 | |
*** mariusv has joined #openstack-dns | 01:02 | |
*** mariusv has quit IRC | 01:02 | |
*** mariusv has joined #openstack-dns | 01:02 | |
*** rektide has joined #openstack-dns | 01:02 | |
*** dcmartin has joined #openstack-dns | 01:11 | |
*** ericpeterson has joined #openstack-dns | 01:16 | |
*** ericpeterson has quit IRC | 01:26 | |
*** ericpeterson has joined #openstack-dns | 01:29 | |
*** MentalRay has quit IRC | 01:46 | |
*** MentalRay has joined #openstack-dns | 01:53 | |
*** elarson has quit IRC | 02:00 | |
*** MentalRay has quit IRC | 02:08 | |
*** GonZo2K has joined #openstack-dns | 02:31 | |
*** mwagner_lap has quit IRC | 02:46 | |
*** mwagner_lap has joined #openstack-dns | 02:59 | |
*** ericpeterson has quit IRC | 03:11 | |
*** chlong has joined #openstack-dns | 03:48 | |
*** chlong has quit IRC | 03:49 | |
*** dcmartin has quit IRC | 04:20 | |
*** MentalRay has joined #openstack-dns | 04:28 | |
*** richm has quit IRC | 04:37 | |
*** penick has joined #openstack-dns | 04:41 | |
*** penick has quit IRC | 04:42 | |
*** penick has joined #openstack-dns | 04:46 | |
*** penick has quit IRC | 04:46 | |
*** GonZoPT has joined #openstack-dns | 04:50 | |
*** GonZo2K has quit IRC | 04:52 | |
*** penick has joined #openstack-dns | 05:15 | |
*** penick has quit IRC | 05:15 | |
*** penick has joined #openstack-dns | 05:33 | |
*** penick has quit IRC | 05:56 | |
*** penick has joined #openstack-dns | 07:32 | |
*** penick has quit IRC | 07:40 | |
*** fawadkhaliq has joined #openstack-dns | 07:52 | |
*** GonZoPT has quit IRC | 07:58 | |
*** penick has joined #openstack-dns | 08:02 | |
*** fawadk has joined #openstack-dns | 08:08 | |
*** MentalRay has quit IRC | 08:11 | |
*** fawadkhaliq has quit IRC | 08:12 | |
*** fawadk has quit IRC | 08:13 | |
*** penick has quit IRC | 08:19 | |
*** jordanP has joined #openstack-dns | 08:20 | |
*** fawadkhaliq has joined #openstack-dns | 08:54 | |
*** untriaged-bot has joined #openstack-dns | 10:01 | |
untriaged-bot | Untriaged bugs so far: | 10:01 |
---|---|---|
untriaged-bot | https://bugs.launchpad.net/designate/+bug/1437699 | 10:01 |
openstack | Launchpad bug 1437699 in Designate "mDNS should Handle Bad File Descriptor" [Medium,New] - Assigned to Ron Rickard (rjrjr) | 10:01 |
*** untriaged-bot has quit IRC | 10:01 | |
*** fawadkhaliq has quit IRC | 11:22 | |
*** km has quit IRC | 11:32 | |
openstackgerrit | Graham Hayes proposed openstack/designate: Fixed sort key to not sort by an un-indexed field by default https://review.openstack.org/170213 | 11:50 |
*** richm has joined #openstack-dns | 12:16 | |
*** ericpeterson has joined #openstack-dns | 12:20 | |
*** fawadkhaliq has joined #openstack-dns | 12:23 | |
*** fawadkhaliq has quit IRC | 12:28 | |
*** lkoranda has quit IRC | 12:29 | |
*** johnbelamaric has joined #openstack-dns | 12:31 | |
*** fawadkhaliq has joined #openstack-dns | 12:46 | |
*** ericpeterson has quit IRC | 12:46 | |
*** kei_yama has quit IRC | 12:57 | |
*** jmcbride has joined #openstack-dns | 13:17 | |
*** elarson has joined #openstack-dns | 13:18 | |
*** ericpeterson has joined #openstack-dns | 13:33 | |
*** openstackgerrit has quit IRC | 13:36 | |
*** openstackgerrit has joined #openstack-dns | 13:36 | |
*** ChanServ sets mode: +v openstackgerrit | 13:36 | |
*** jmcbride has quit IRC | 13:56 | |
*** jmcbride has joined #openstack-dns | 13:56 | |
*** vinod1 has joined #openstack-dns | 14:00 | |
*** ericpeterson has quit IRC | 14:07 | |
*** paul_glass has joined #openstack-dns | 14:13 | |
*** jordanP has quit IRC | 14:26 | |
*** openstackgerrit has quit IRC | 14:39 | |
*** openstackgerrit has joined #openstack-dns | 14:39 | |
*** ChanServ sets mode: +v openstackgerrit | 14:39 | |
paul_glass | what is the difference between a pool_nameserver and a pool_target in designate.conf? | 14:48 |
*** fawadkhaliq has quit IRC | 14:49 | |
mugsie | paul_glass: a target is where we write changes, nameserver is where we read from | 14:53 |
timsim | A nameserver is the thing that gets polled, so where the change actually resolves, the target is where the change is made. So for BIND9, that'd be where you'd put options like rndc host/port. For powerdns, that'd be where you put a database connection string, for the agent, that's where you put the host/port that is different than the host/port to poll | 14:53 |
mugsie | (e.g. with akamai, we write to the REST API, and read from the nameservers) | 14:54 |
mugsie | or PowerDNS we write the new zone in DB, but could read from totally differnt servers | 14:54 |
paul_glass | that makes sense. thanks. | 14:55 |
*** dcmartin has joined #openstack-dns | 15:16 | |
*** dcmartin has quit IRC | 15:17 | |
*** dcmartin has joined #openstack-dns | 15:18 | |
*** Kiall has quit IRC | 15:22 | |
*** Kiall has joined #openstack-dns | 15:22 | |
openstackgerrit | Paul Glass proposed openstack/designate: Remove zone wrapper object in Tempest tests https://review.openstack.org/170538 | 15:30 |
*** vinod1 has quit IRC | 15:39 | |
*** ericpeterson has joined #openstack-dns | 15:41 | |
elarson | I'm trying to create a zone and getting this error. http://paste.openstack.org/show/198143/ is this b/c there aren't any masters defined? | 15:41 |
timsim | Most likely elarson | 15:42 |
elarson | we should probably update the docs | 15:42 |
* elarson looks for a ticket | 15:42 | |
timsim | The sample config? | 15:43 |
timsim | Oh jk, yeah | 15:43 |
elarson | timsim: oh, I was thinking about the docs for creating a domain, but you're right, I should double check the config | 15:44 |
timsim | Yeah you need https://github.com/openstack/designate/blob/master/etc/designate/designate.conf.sample#L268 probably | 15:44 |
elarson | that said, it arguably could default to whatever ns has been configured right? | 15:44 |
timsim | For masters? I don't think so, because most of the time that'd be itself...I think | 15:45 |
elarson | timsim: you and I should have a rap session about this ;) | 15:45 |
* timsim walks 3 steps | 15:45 | |
* elarson takes 2 steps back | 15:46 | |
jbratton | is this going to turn in to a dance off? | 15:46 |
* Kiall backs away slowly | 15:47 | |
elarson | already happened... I won | 15:49 |
timsim | I would argue that result | 15:49 |
elarson | dangit!!! | 15:49 |
*** penick has joined #openstack-dns | 15:50 | |
elarson | timsim: so what is this error about then? http://paste.openstack.org/show/198146/ | 15:54 |
paul_glass | elarson: I just ran into that... I think I had the pool_target and pool_nameserver sections swapped. | 15:55 |
timsim | I think that's this line https://github.com/openstack/designate/blob/master/etc/designate/designate.conf.sample#L269 | 15:55 |
dcmartin | couchdb | 15:56 |
elarson | maybe it is couchdb | 15:56 |
elarson | whow knows | 15:56 |
elarson | timsim: I have that uncommented for the target | 15:56 |
*** dcmartin has quit IRC | 15:57 | |
mugsie | well that was odd | 15:58 |
*** andrewbogott_afk is now known as andrewbogott | 16:01 | |
*** untriaged-bot has joined #openstack-dns | 16:01 | |
untriaged-bot | Untriaged bugs so far: | 16:01 |
untriaged-bot | https://bugs.launchpad.net/designate/+bug/1437699 | 16:01 |
openstack | Launchpad bug 1437699 in Designate "mDNS should Handle Bad File Descriptor" [Medium,New] - Assigned to Ron Rickard (rjrjr) | 16:01 |
*** untriaged-bot has quit IRC | 16:01 | |
*** penick has quit IRC | 16:04 | |
elarson | here is a tidbit | 16:18 |
timsim | couchdb? | 16:18 |
elarson | in oslo.config (and I suspect configparser as well) if you have `foo = bar # a helpful comment` the value of foo is {'foo': 'bar # a helpful comment'} | 16:19 |
elarson | I suppose that makes sense since you often would have characters after the `=` that you'd want to include | 16:19 |
elarson | in any case, knowing is half the battle | 16:19 |
*** bharath_ has joined #openstack-dns | 16:28 | |
openstackgerrit | Paul Glass proposed openstack/designate: First Tempest tests for recordsets https://review.openstack.org/167356 | 16:29 |
*** vinod1 has joined #openstack-dns | 16:48 | |
bharath_ | We are trying to setup designate juno (in ebay) and we are running into an issue for a single nova boot ... There is nothing else happening at the same time and so it seems strange that it's waiting on a lock !!! Here's the stack trace: | 17:08 |
bharath_ | Stack trace for create: 2015-04-03 01:28:07.740 11180 TRACE oslo.messaging.rpc.dispatcher Traceback (most recent call last): 2015-04-03 01:28:07.740 11180 TRACE oslo.messaging.rpc.dispatcher File "/opt/openstack/designate/2015.2.0.r1/lib/python2.7/site-packages/oslo/messaging/rpc/dispatcher.py", line 134, in _dispatch_and_reply 2015-04-03 01:28:07.740 11180 TRACE oslo.messaging.rpc.dispatcher incoming.message)) 2015-04 | 17:08 |
bharath_ | 2015-04-03 01:28:07.740 11180 TRACE oslo.messaging.rpc.dispatcher OperationalError: (OperationalError) (1205, 'Lock wait timeout exceeded; try restarting transaction') 'INSERT INTO recordsets (id, version, created_at, tenant_id, domain_id, name, type, ttl) VALUES (%s, %s, %s, %s, %s, %s, %s, %s)' ('772bdc16437343ae874b259e510e6eac', 1, datetime.datetime(2015, 4, 3, 1, 27, 16, 152409), 'a9fac4ab8c314ed786f547a7facfa5ca', '387 | 17:09 |
bharath_ | Any pointers would be helpful ? | 17:10 |
*** penick has joined #openstack-dns | 17:28 | |
*** pk_ has joined #openstack-dns | 17:38 | |
Kiall | bharath_: just had a chance to have a quick look, so the underlying bug we talked about a few days back could still be causing that - even with exactly nova boot, depending on your sink handler... The bug effectivly meant a request came into central, central would start a TX and say, insert a record, it would then yield to another thread, which could be a another request from the sink handler etc, which would open a TX and then stall trying to | 17:38 |
Kiall | bump the serial number up, leading to a deadlocked process which can't recover until the timeout. | 17:38 |
Kiall | It was a total "how on earth did this bug go unnoticed for so long" - we had that bug in HP Cloud DNS producton for quite some time, and it didn't even show itself there :( | 17:39 |
bharath_ | If it's deadlocked on bumping the serial number and if I do increment_serial = False from the sink handlers, then it shouldn't run into the issue, rite ? | 17:42 |
Kiall | bharath_: Ehh - I think so, the serial is the only place I know of where we deadlocked. | 17:44 |
Kiall | (Also - I never tried doing that, so it's possible it won't work..) | 17:44 |
bharath_ | Yup, I tried that and it didn't work :( | 17:45 |
Kiall | Oh, I just looked at the paste above and noticed the query.. | 17:45 |
Kiall | that's an `insert into recordsets` getting a lock wait timeout .. Humm | 17:45 |
bharath_ | yes | 17:45 |
bharath_ | in case of nova delete, it's 'delete from records' | 17:46 |
Kiall | Can you fire one off, and during the 30 second window collect the output of this? SHOW ENGINE INNODB STATUS; | 17:47 |
Kiall | Theres also .. And I always have to go find this query on StackOverflow every time I need it ;) | 17:49 |
Kiall | SELECT r.trx_id waiting_trx_id, r.trx_mysql_thread_id waiting_thread, | 17:49 |
Kiall | r.trx_query waiting_query, | 17:49 |
Kiall | b.trx_id blocking_trx_id, b.trx_mysql_thread_id blocking_thread, | 17:49 |
Kiall | b.trx_query blocking_query | 17:49 |
Kiall | FROM information_schema.innodb_lock_waits w | 17:49 |
Kiall | INNER JOIN information_schema.innodb_trx b ON b.trx_id = w.blocking_trx_id | 17:49 |
Kiall | INNER JOIN information_schema.innodb_trx r ON r.trx_id = w.requesting_trx_id; | 17:49 |
*** rrickard has joined #openstack-dns | 17:52 | |
*** richm has quit IRC | 17:53 | |
rrickard | @kiall: are you working with bharath_? | 17:55 |
Kiall | Yea, asked him to gather some details during a lock.. | 17:55 |
openstackgerrit | Merged openstack/designate: Remove zone wrapper object in Tempest tests https://review.openstack.org/170538 | 17:55 |
Kiall | http://eavesdrop.openstack.org/irclogs/%23openstack-dns/%23openstack-dns.2015-04-03.log has some details ;) | 17:56 |
Kiall | (not pasting that big query twice :P) | 17:56 |
Kiall | About to hop on another call, be back in a bit. | 17:57 |
*** pk_ has quit IRC | 17:58 | |
rrickard | @mugsie: our chat message doesn't show the location of the logs anymore? | 17:58 |
rrickard | i'm just seeing a message about Zuul being stuck. | 17:59 |
Kiall | rrickard: something broke while something was broke and the topic has been like that ever since.. lol | 18:06 |
*** ChanServ sets mode: +o Kiall | 18:06 | |
*** Kiall changes topic to "OpenStack Designate - Logged @ http://eavesdrop.openstack.org/irclogs/%23openstack-dns" | 18:06 | |
*** ChanServ sets mode: -o Kiall | 18:06 | |
*** richm has joined #openstack-dns | 18:11 | |
bharath_ | | 39A9D6 | 19257 | INSERT INTO recordsets (id, version, created_at, tenant_id, domain_id, name, type, ttl) VALUES ('cede4541c31d4ac9a8ed1bb9b25cf990', 1, '2015-04-03 18:09:23.748365', 'a9fac4ab8c314ed786f547a7facfa5ca', '3876b908dffb47a181a09570ef21387f', 'dev-test30-5655.slc01.dev.ebayc3.com.', 'A', NULL) | 39A9D4 | 19255 | NULL | 18:13 |
bharath_ | that's the output for the sql query | 18:13 |
bharath_ | only one row | 18:13 |
*** dcmartin has joined #openstack-dns | 18:18 | |
bharath_ | http://cryptb.in/Vx6vP (key - 52c4c4ac5d516c415c67eabd330e83af) | 18:26 |
*** pk_ has joined #openstack-dns | 18:44 | |
*** penick has quit IRC | 19:00 | |
*** pk_ has quit IRC | 19:01 | |
*** pk_ has joined #openstack-dns | 19:02 | |
*** pk_ has quit IRC | 19:06 | |
*** dcmartin has quit IRC | 19:07 | |
*** MentalRay has joined #openstack-dns | 19:12 | |
*** penick has joined #openstack-dns | 19:14 | |
*** dcmartin has joined #openstack-dns | 19:14 | |
*** dcmartin has quit IRC | 19:55 | |
*** dcmartin has joined #openstack-dns | 19:55 | |
*** ericpeterson has quit IRC | 20:00 | |
*** ericpeterson has joined #openstack-dns | 20:00 | |
*** ericpete_ has joined #openstack-dns | 20:01 | |
*** pk_ has joined #openstack-dns | 20:03 | |
*** ericpeterson has quit IRC | 20:06 | |
*** pk_ has quit IRC | 20:07 | |
*** MentalRay has quit IRC | 20:07 | |
*** dcmartin has quit IRC | 20:16 | |
*** bharath_ has quit IRC | 20:30 | |
*** fawadkhaliq has joined #openstack-dns | 20:39 | |
rrickard | @Kiall: we have figured out why we are getting the lock contention. | 20:40 |
rrickard | we have Designate sink listening on a topic called 'dns'. | 20:40 |
rrickard | nova is publishing on 'notifications, dns'. | 20:41 |
rrickard | Designate sink is consuming both 'notifications.info' and 'dns.info' even though we want it just consuming 'dns.info' and our config shows just 'dns' topic. | 20:41 |
rrickard | this was not a problem in Designate Icehouse, but appeared when we upgraded to Designate Juno. | 20:42 |
rrickard | so, 2 notifications are consumed and we have lock contention. | 20:43 |
*** penick has quit IRC | 20:56 | |
*** MentalRay has joined #openstack-dns | 21:02 | |
ekarlso | rrickard: how's life at ebay ? :p | 21:04 |
*** MentalRay has quit IRC | 21:13 | |
openstackgerrit | Tim Simmons proposed openstack/designate: WIP: Refactor notify logic into backends https://review.openstack.org/170612 | 21:22 |
*** vinod1 has quit IRC | 21:26 | |
*** paul_glass has quit IRC | 21:41 | |
*** pk has joined #openstack-dns | 21:44 | |
*** penick has joined #openstack-dns | 21:49 | |
*** rrickard has quit IRC | 21:58 | |
*** andrewbogott is now known as andrewbogott_afk | 22:07 | |
*** MentalRay has joined #openstack-dns | 22:07 | |
*** ericpete_ has quit IRC | 22:08 | |
*** pk has quit IRC | 22:09 | |
*** penick has quit IRC | 22:30 | |
*** pk has joined #openstack-dns | 22:32 | |
openstackgerrit | Kiall Mac Innes proposed openstack/designate: Split out Optional OpenStack services in localrc https://review.openstack.org/170620 | 22:34 |
*** penick has joined #openstack-dns | 22:35 | |
*** pk has quit IRC | 22:37 | |
*** pk has joined #openstack-dns | 22:43 | |
*** fawadkhaliq has quit IRC | 22:48 | |
Kiall | timsim: if your still about, https://review.openstack.org/#/c/166191/ should finally be good to turn back on again :) | 22:48 |
openstackgerrit | Kiall Mac Innes proposed openstack/designate: DevStack: Akamai/Dyn Support for Jenkins Credentials and Exercises https://review.openstack.org/169891 | 22:52 |
Kiall | and ^ ;) | 22:53 |
*** jmcbride has quit IRC | 22:56 | |
openstackgerrit | Kiall Mac Innes proposed openstack/designate: DevStack: Akamai/Dyn Support for Jenkins Credentials and Exercises https://review.openstack.org/169891 | 22:56 |
*** pk has quit IRC | 23:03 | |
*** MentalRay has quit IRC | 23:09 | |
boris-42 | Kiall ekarlso howdy | 23:11 |
Kiall | heya | 23:13 |
boris-42 | Kiall: how are you?) | 23:19 |
Kiall | Good, but about to head to bed :P | 23:20 |
boris-42 | Kiall: -) | 23:21 |
boris-42 | Kiall: soo I made recently a lib that should be interesting for you=) | 23:21 |
boris-42 | Kiall: https://github.com/boris-42/profimp | 23:21 |
Kiall | Oh? | 23:21 |
boris-42 | Kiall: yep to speed up import of python-designateclient | 23:22 |
*** pk has joined #openstack-dns | 23:23 | |
Kiall | Interesting.. Not something I've considered optomizing before, given it's a once off startup cost - but.. Yea, the CLI starts up every command ;) | 23:23 |
Kiall | (and all the OS CLIs are terribly slow to "boot" ;)) | 23:23 |
boris-42 | Kiall: http://boris-42.github.io/profimp/designate.html | 23:25 |
boris-42 | Kiall: so this is trace for designate client ^ | 23:26 |
boris-42 | Kiall: so now we see how imports are done and where the most time is consumed | 23:26 |
boris-42 | Kiall: so adding in proper places lazy imports can fix this | 23:26 |
Kiall | 208.41 ms ain't bad in fairness! | 23:27 |
Kiall | (We're dealing with python compile times ;)) | 23:27 |
Kiall | Anyway, I'm nearly fast asleep! I'll have a look tomorrow :) | 23:28 |
boris-42 | Kiall: good night=) | 23:28 |
boris-42 | Kiall: so ping me when you find time | 23:28 |
Kiall | :) | 23:28 |
boris-42 | Kiall: and yep 270 is bad for libs | 23:28 |
boris-42 | Kiall: it should be something like 20ms=) | 23:28 |
Kiall | At a glance, Most of our time is spent importing PBR (we can probably kill that one) and Keystone (Can't do much about that one, 99% of use cases will need us to load it anyway, so lazy loading just defers it) | 23:30 |
Kiall | actually, PBR seems (from the graph) to be coming in via Keystone Client | 23:31 |
Kiall | " from keystoneclient import adapter .. 207.47 ms" out total is 208.41 ms | 23:31 |
Kiall | Anyway, I'm probably reading it wrong at this point of the evening :P | 23:32 |
Kiall | cya - will check it out properly over the weekend :) | 23:32 |
*** penick has quit IRC | 23:35 | |
boris-42 | Kiall: -) | 23:39 |
*** pk has quit IRC | 23:53 | |
*** pk has joined #openstack-dns | 23:55 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!