| *** EricGonczer_ has joined #openstack-dns | 00:04 | |
| *** dtx00ff has quit IRC | 00:08 | |
| *** dtx00ff has joined #openstack-dns | 00:09 | |
| *** dtx00ff has quit IRC | 00:10 | |
| *** dtx00ff has joined #openstack-dns | 00:11 | |
| *** jmcbride has quit IRC | 00:11 | |
| *** EricGonczer_ has quit IRC | 00:18 | |
| *** amcrn has quit IRC | 00:32 | |
| *** amcrn has joined #openstack-dns | 00:34 | |
| *** rjrjr has quit IRC | 00:39 | |
| *** jmcbride has joined #openstack-dns | 00:49 | |
| *** jmcbride has quit IRC | 00:53 | |
| *** jmcbride has joined #openstack-dns | 00:53 | |
| *** jmcbride has quit IRC | 00:58 | |
| *** rmoe has quit IRC | 01:07 | |
| *** amcrn has quit IRC | 01:13 | |
| *** dtx00ff has quit IRC | 01:14 | |
| *** jmcbride has joined #openstack-dns | 01:24 | |
| *** rmoe has joined #openstack-dns | 01:25 | |
| *** jmcbride has quit IRC | 01:32 | |
| *** amcrn has joined #openstack-dns | 01:36 | |
| *** amcrn has quit IRC | 01:47 | |
| *** jmcbride has joined #openstack-dns | 01:59 | |
| *** openstackgerrit has quit IRC | 02:01 | |
| *** EricGonczer_ has joined #openstack-dns | 02:05 | |
| *** EricGonczer_ has quit IRC | 02:07 | |
| *** jmcbride has quit IRC | 02:09 | |
| *** openstackgerrit has joined #openstack-dns | 02:20 | |
| *** ChanServ sets mode: +v openstackgerrit | 02:20 | |
| *** jmcbride has joined #openstack-dns | 03:07 | |
| *** jmcbride has quit IRC | 03:20 | |
| *** k4n0 has joined #openstack-dns | 04:46 | |
| *** HenryG has quit IRC | 05:09 | |
| *** HenryG has joined #openstack-dns | 05:22 | |
| *** alokj has joined #openstack-dns | 05:31 | |
| *** ttrumm has joined #openstack-dns | 06:08 | |
| *** ttrumm_ has joined #openstack-dns | 06:19 | |
| *** ttrumm has quit IRC | 06:22 | |
| *** ttrumm_ has quit IRC | 06:32 | |
| *** ttrumm has joined #openstack-dns | 06:33 | |
| *** diga has quit IRC | 06:33 | |
| *** ttrumm has quit IRC | 06:51 | |
| *** ttrumm has joined #openstack-dns | 06:52 | |
| *** alokj has quit IRC | 06:55 | |
| *** ttrumm has quit IRC | 07:06 | |
| *** alokj has joined #openstack-dns | 07:09 | |
| *** simonmcc has quit IRC | 07:44 | |
| *** simonmcc has joined #openstack-dns | 07:45 | |
| *** jordanP has joined #openstack-dns | 08:20 | |
| *** k4n0 has quit IRC | 08:37 | |
| *** k4n0 has joined #openstack-dns | 08:53 | |
| *** k4n0 has quit IRC | 09:22 | |
| *** k4n0 has joined #openstack-dns | 09:35 | |
| *** EricGonczer_ has joined #openstack-dns | 11:37 | |
| *** mwagner_lap has quit IRC | 11:41 | |
| *** EricGonczer_ has quit IRC | 11:55 | |
| *** k4n0 has quit IRC | 12:01 | |
| *** alokj has quit IRC | 12:27 | |
| *** alokj has joined #openstack-dns | 12:43 | |
| *** alokj is now known as Guest96667 | 12:44 | |
| *** Guest96667 has quit IRC | 12:45 | |
| *** richm has joined #openstack-dns | 13:04 | |
| *** mwagner_lap has joined #openstack-dns | 13:13 | |
| *** nkinder has quit IRC | 13:23 | |
| *** jmcbride has joined #openstack-dns | 13:25 | |
| *** EricGonczer_ has joined #openstack-dns | 13:53 | |
| *** jmcbride has quit IRC | 13:55 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/designate: Updated from global requirements https://review.openstack.org/122419 | 13:58 |
|---|---|---|
| *** EricGonczer_ has quit IRC | 13:59 | |
| *** EricGonczer_ has joined #openstack-dns | 14:03 | |
| *** timsim has joined #openstack-dns | 14:07 | |
| *** jmcbride has joined #openstack-dns | 14:08 | |
| *** nkinder has joined #openstack-dns | 14:10 | |
| timsim | Kiall: Around? I wanted to talk about bind9 briefly. | 14:11 |
| Kiall | Hyea | 14:11 |
| Kiall | Heya* | 14:11 |
| timsim | So with this review: https://review.openstack.org/#/c/119768 | 14:12 |
| timsim | I just wanted to point out how I would think bind9 zones would be added with MiniDNS in the fold. | 14:12 |
| timsim | Basically I'm ok with putting a one second sleep in there for now. | 14:13 |
| timsim | With bind, you can issue an rndc addzone mydomain.com '{ type slave; masters { master_ip; }; };' Where the master_ip is Mdns. And it magically kicks of an AXFR for that zone. No need to touch or make any zone file. | 14:13 |
| Kiall | It get's easier with mDNS to be fair - we just `rndc addzone bla.com "{type: seconadary; masters: 10.0.0.1; }"` (or something along those lines | 14:14 |
| Kiall | timsim: yea, exactly :) | 14:14 |
| timsim | Right. I just wanted to make sure that was common knowledge. So I guess from my point of view, adding a little sleep in the current backend wouldn't be the worst thing in the world. I was able to replicate that issue yesterday on 14.04 with the same bind setup that devstack was using. | 14:15 |
| Kiall | Have you guys tested if you can reproduce the issue under the in-tree vagrant/devstack stuff? Because, I just can't -_- I'm stumped at why though.. | 14:15 |
| Kiall | brb - 2 mins | 14:15 |
| timsim | So I did. | 14:15 |
| timsim | sure | 14:15 |
| Kiall | back | 14:19 |
| timsim | I'm wondering what kind of VM you're using? Maybe it's slow enough that that the time difference doesn't manifest itself. With the same setup you had, I actually had it work correctly once, but fail most of the time. http://paste.openstack.org/raw/112869/ | 14:19 |
| Kiall | Interesting - The machine I'm testing on is "mighty fast" .. so I would expect the VM performance to be as good as any VM.. Looking at your logs, it might be that the first reload happened right at the "end" of a second | 14:21 |
| Kiall | This really really seems like a bind bug :/ | 14:21 |
| Kiall | reload | 14:22 |
| Kiall | All zones with a changed timestamp are reloaded at the next incoming query for them. | 14:22 |
| Kiall | lol | 14:22 |
| Kiall | oh | 14:22 |
| Kiall | that's the pdns docs -_- | 14:22 |
| Kiall | (for the "bind" backend to pdns) | 14:22 |
| timsim | Yeah that's kind of what it looks like, but why would it work fine for you? | 14:22 |
| Kiall | Luck of the draw? Maybe all of mine have, by fluke, landed towards the end of a second boundary | 14:23 |
| timsim | Yeah. It's so odd. | 14:23 |
| Kiall | "[102] When you reload the primary, it loads the updated zone file regardless of whether you've changed the serial number. It checks the file's timestamp, sees that it's been modified since it last loaded the data, and reads the file." | 14:23 |
| Kiall | from http://docstore.mik.ua/orelly/networking_2ndEd/dns/ch14_03.htm | 14:23 |
| timsim | Well shouldn't it load the updated file if you DID actually change the serial number? | 14:24 |
| Kiall | According to that, bind only re-reads the file IF the the file's timestamp has changed | 14:25 |
| Kiall | Which is exactly what we're seeing | 14:25 |
| timsim | -_- | 14:25 |
| timsim | If that's the case it seems like putting a sleep in there is the only way for that to work properly every time? | 14:26 |
| Kiall | I think you might just be right.. Unless bind has a config option to disable that | 14:26 |
| Kiall | I'm going to add the sleep - it's somewhat harmless as A) it's only on creating a zone and B) it's inside eventlet - so we're not actually blocking the whole process for 1sec - just that specific request. | 14:27 |
| timsim | Yeah. Plus, this is going to be replaced soon anyway. | 14:28 |
| Kiall | ++ | 14:28 |
| timsim | I've been using Mdns with bind a little bit and it's quite nice. I'm looking forward to putting some load on it here soon. | 14:30 |
| Kiall | Nice :) | 14:30 |
| *** EricGonczer_ has quit IRC | 14:30 | |
| Kiall | Now.. The Q is, where do we put the time.sleep(1) .. | 14:35 |
| timsim | Yeah because we don't actually issue that reload call do we? | 14:38 |
| Kiall | Well, we do issue it, it's just from outside the bind9 driver | 14:39 |
| Kiall | (well - the reload is inside bind9 driver, we call into bind9 driver twice) | 14:40 |
| *** paul_glass has joined #openstack-dns | 14:43 | |
| timsim | The two reloads are for create_ns/create_soa yeah? | 14:44 |
| Kiall | yea - I think I have a fix that's somewhat acceptable.. | 14:45 |
| jordanP | timsim, Hi. I have a problem with bind9 and the floatingIP PTR records. ( https://bugs.launchpad.net/designate/+bug/1370621 ) My question is : | 14:49 |
| uvirtbot | Launchpad bug 1370621 in designate "Bing9 PTR Zone error : multiple RRs of singleton type" [Undecided,New] | 14:49 |
| jordanP | could a bind9 zone have several SOA records ? | 14:50 |
| jordanP | if not how come only the reverse resolution is affected by this bug whereas the forward resolution works great with bind 9 ? | 14:50 |
| jordanP | if you have any idea, that would be great :) | 14:51 |
| timsim | jordanP: I'll take a look, I'm no bind expert. | 14:52 |
| jordanP | :) | 14:52 |
| jordanP | thanks | 14:52 |
| jbratton | timsim: when you are testing this, how much logging do you have turned on for BIND? | 14:53 |
| jbratton | I'm sort of wondering if you're not creating an invalid zone and that's why the reload is failing | 14:53 |
| jordanP | the designate code is clear : central/service.py::create_domain() calls _create_soa() but there's already an SOA entry in bind9-zone.jinja2 template | 14:54 |
| timsim | jordanP: you might have to edit that template file to suit your needs. | 14:54 |
| jordanP | well that seems weird that the default template doesn't work only for the reverse dns resolution | 14:55 |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Backends need to implement c/u/d_recordset https://review.openstack.org/119768 | 14:56 |
| Kiall | timsim: .. ^ might just do it... | 14:56 |
| timsim | jbratton: It's just the default logging. There was an issue where it was failing with 'bad zone', and the reload actually works if the timestamps are different from the addzone (I think) | 14:56 |
| Kiall | jbratton: we actually are creating an invalid zone - momentarily.. | 14:56 |
| Kiall | due to some other internals, we end up creating a zonefile with just the NS records first, then add the SOA after.. It all disappears with mDNS in a couple of weeks though! | 14:57 |
| jbratton | well good, then I won't even worry about it then :) | 14:59 |
| *** jmcbride1 has joined #openstack-dns | 14:59 | |
| jbratton | because we do have a support contract with ISC, so we can open up bug reports with them and all that | 14:59 |
| Kiall | Yea, it's really a total quirk that will disappear once we stop writing zonefiles and bind is always a slave | 15:00 |
| jbratton | but if this becomes a non-issue shortly, it doesn't seem worth it | 15:00 |
| jbratton | cool | 15:00 |
| *** jmcbride has quit IRC | 15:01 | |
| timsim | Kiall: That worked for me. http://paste.openstack.org/show/112890/ brb | 15:02 |
| *** paul_glass1 has joined #openstack-dns | 15:04 | |
| Kiall | that's with my new patchset? | 15:04 |
| *** paul_glass has quit IRC | 15:07 | |
| Kiall | timsim: ^ | 15:13 |
| timsim | Kiall: Yup! | 15:24 |
| *** jmcbride1 has quit IRC | 15:25 | |
| timsim | And the dig showed the correct ns/soa records. | 15:25 |
| *** openstackgerrit has quit IRC | 15:33 | |
| *** jmcbride has joined #openstack-dns | 15:36 | |
| *** betsy has joined #openstack-dns | 15:43 | |
| *** rmoe has quit IRC | 15:48 | |
| *** EricGonczer_ has joined #openstack-dns | 15:49 | |
| Kiall | timsim: Cool :) | 15:49 |
| *** nkinder has quit IRC | 16:00 | |
| *** openstackgerrit has joined #openstack-dns | 16:02 | |
| *** ChanServ sets mode: +v openstackgerrit | 16:02 | |
| *** rmoe has joined #openstack-dns | 16:04 | |
| *** EricGonc_ has joined #openstack-dns | 16:05 | |
| *** EricGonczer_ has quit IRC | 16:05 | |
| *** EricGonczer_ has joined #openstack-dns | 16:13 | |
| *** nkinder has joined #openstack-dns | 16:13 | |
| *** EricGonc_ has quit IRC | 16:14 | |
| jordanP | Kiall, somehow https://review.openstack.org/#/c/119768/10 fixes https://bugs.launchpad.net/designate/+bug/1370621 | 16:20 |
| uvirtbot | Launchpad bug 1370621 in designate "Bing9 PTR Zone error : multiple RRs of singleton type" [Undecided,New] | 16:20 |
| Kiall | jordanP: I suspect it might :) | 16:20 |
| Kiall | suspected* | 16:20 |
| jordanP | I believe it's due to the fact that you rewrote the jinja2 template | 16:20 |
| Kiall | Well - Before, we had some really really silly things happening :) and yes, the template rewrite would have got rid of the second SOA | 16:21 |
| jordanP | thanks a lot | 16:21 |
| jordanP | i can move forward with my user story at $my_job :) | 16:21 |
| Kiall | :) | 16:23 |
| openstackgerrit | Kiall Mac Innes proposed a change to openstack/designate: Backends need to implement c/u/d_recordset https://review.openstack.org/119768 | 16:24 |
| Kiall | (Updated commit message to reflect the second bug..) | 16:24 |
| *** vinod1 has joined #openstack-dns | 16:27 | |
| *** jmcbride has quit IRC | 16:30 | |
| *** jmcbride has joined #openstack-dns | 16:31 | |
| *** jordanP has quit IRC | 16:35 | |
| *** vinod1 has quit IRC | 16:55 | |
| *** vinod1 has joined #openstack-dns | 17:04 | |
| Kiall | vinod1: so, https://review.openstack.org/#/c/119768 should be good to go now.. timsim has tested it works reliably with your bind setup :) | 17:06 |
| Kiall | heh.. just say jordanP's comment on it :( | 17:07 |
| Kiall | saw* | 17:07 |
| * Kiall needs to give up coding and become a farmer | 17:08 | |
| vinod1 | If you create multiple recordsets within a second - your change prevents any reloads not failing | 17:11 |
| *** photocyte has joined #openstack-dns | 17:14 | |
| *** paul_glass1 has quit IRC | 17:29 | |
| *** openstackgerrit has quit IRC | 17:31 | |
| *** openstackgerrit has joined #openstack-dns | 17:32 | |
| *** ChanServ sets mode: +v openstackgerrit | 17:32 | |
| *** nkinder has quit IRC | 17:38 | |
| *** amcrn has joined #openstack-dns | 17:47 | |
| *** jmcbride has quit IRC | 17:48 | |
| *** rossk has joined #openstack-dns | 17:53 | |
| *** nkinder has joined #openstack-dns | 17:54 | |
| *** photocyte has quit IRC | 17:56 | |
| *** vinod1 has quit IRC | 18:00 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack/designate-specs: Use OpenStack incubated branding https://review.openstack.org/122481 | 18:01 |
| *** photocyte has joined #openstack-dns | 18:02 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack/designate-specs: Use OpenStack incubated branding https://review.openstack.org/122481 | 18:04 |
| *** jmcbride has joined #openstack-dns | 18:06 | |
| *** dtx00ff has joined #openstack-dns | 18:10 | |
| *** mwagner_lap has quit IRC | 18:18 | |
| *** jmcbride has quit IRC | 18:31 | |
| openstackgerrit | Andreas Jaeger proposed a change to openstack/designate-specs: Added olsosphinx theme, and enable incubation option for specs output https://review.openstack.org/113535 | 18:31 |
| *** jmcbride has joined #openstack-dns | 18:40 | |
| *** paul_glass has joined #openstack-dns | 18:40 | |
| *** photocyte has quit IRC | 18:44 | |
| *** photocyte has joined #openstack-dns | 18:49 | |
| *** jmcbride has quit IRC | 18:55 | |
| *** dtx00ff has quit IRC | 18:55 | |
| *** jmcbride has joined #openstack-dns | 19:04 | |
| *** paul_glass has left #openstack-dns | 19:10 | |
| *** mwagner_lap has joined #openstack-dns | 19:10 | |
| *** paul_glass1 has joined #openstack-dns | 19:11 | |
| *** EricGonczer_ has quit IRC | 19:33 | |
| openstackgerrit | A change was merged to openstack/designate: Downgrade in 039 does not delete NS recordsets https://review.openstack.org/119890 | 19:33 |
| *** EricGonczer_ has joined #openstack-dns | 19:33 | |
| *** EricGonczer_ has quit IRC | 19:35 | |
| *** EricGonczer_ has joined #openstack-dns | 19:36 | |
| *** photocyte has quit IRC | 19:40 | |
| *** openstackgerrit has quit IRC | 19:45 | |
| *** vinod1 has joined #openstack-dns | 19:57 | |
| *** jmcbride has quit IRC | 20:00 | |
| *** dtx00ff has joined #openstack-dns | 20:01 | |
| *** vinod1 has quit IRC | 20:04 | |
| *** jmcbride has joined #openstack-dns | 20:05 | |
| *** jmcbride1 has joined #openstack-dns | 20:07 | |
| *** jmcbride has quit IRC | 20:10 | |
| *** serverascode has quit IRC | 20:23 | |
| *** vinod1 has joined #openstack-dns | 20:25 | |
| *** serverascode has joined #openstack-dns | 20:25 | |
| *** paul_glass1 has quit IRC | 20:26 | |
| *** paul_glass has joined #openstack-dns | 20:26 | |
| vinod1 | mugsie: i don't see doc/sources/diagrams/*.shapes referenced anywhere else. Are they intended to be standalone? | 20:43 |
| *** dtx00ff has quit IRC | 20:49 | |
| *** dtx00ff has joined #openstack-dns | 20:50 | |
| *** nkinder has quit IRC | 20:54 | |
| vinod1 | ekarlso: are you around? | 20:58 |
| ekarlso | vinod1: little bit :p | 20:58 |
| vinod1 | Re: bug 1366953 | 20:59 |
| uvirtbot | Launchpad bug 1366953 in designate "Creating duplicate Zone in V2 API gives 500" [High,New] https://launchpad.net/bugs/1366953 | 20:59 |
| ekarlso | yes | 21:00 |
| *** vinod1 has quit IRC | 21:03 | |
| *** vinod1 has joined #openstack-dns | 21:04 | |
| *** openstackgerrit has joined #openstack-dns | 21:07 | |
| *** ChanServ sets mode: +v openstackgerrit | 21:07 | |
| *** vinod1 has quit IRC | 21:08 | |
| *** vinod1 has joined #openstack-dns | 21:09 | |
| vinod1 | Looks like my previous messages did not make it - with constant disconnects | 21:10 |
| vinod1 | ekarlso: What backend were you using when you hit bug 1366953? | 21:11 |
| uvirtbot | Launchpad bug 1366953 in designate "Creating duplicate Zone in V2 API gives 500" [High,New] https://launchpad.net/bugs/1366953 | 21:11 |
| vinod1 | I have seen such failures when an entry was present in the backend but not in storage | 21:11 |
| ekarlso | vinod1: pdns | 21:12 |
| vinod1 | was it that the (duplicated) entry was present in the pdns db but not in the storage db? | 21:12 |
| ekarlso | vinod1: dont remember sadly :| | 21:13 |
| vinod1 | np - have a good night | 21:13 |
| *** betsy has quit IRC | 21:27 | |
| *** vinod1 has quit IRC | 21:29 | |
| *** vinod1 has joined #openstack-dns | 21:29 | |
| *** jmcbride1 has quit IRC | 21:32 | |
| *** vinod1 has quit IRC | 21:40 | |
| *** amcrn has quit IRC | 21:55 | |
| *** EricGonczer_ has quit IRC | 22:00 | |
| *** paul_glass has quit IRC | 22:04 | |
| *** timsim has quit IRC | 22:06 | |
| *** nkinder has joined #openstack-dns | 22:19 | |
| *** amcrn has joined #openstack-dns | 22:23 | |
| *** vinod1 has joined #openstack-dns | 22:24 | |
| *** dtx00ff has quit IRC | 22:52 | |
| *** dtx00ff has joined #openstack-dns | 22:54 | |
| openstackgerrit | OpenStack Proposal Bot proposed a change to openstack/designate: Updated from global requirements https://review.openstack.org/122419 | 22:55 |
| *** amcrn has quit IRC | 23:04 | |
| *** arborism has joined #openstack-dns | 23:08 | |
| openstackgerrit | Endre Karlson proposed a change to openstack/python-designateclient: Use keystone sessions for v1 client https://review.openstack.org/122561 | 23:12 |
| *** timsim has joined #openstack-dns | 23:19 | |
| *** rossk has quit IRC | 23:29 | |
| *** EricGonczer_ has joined #openstack-dns | 23:32 | |
| *** richm has quit IRC | 23:34 | |
| *** EricGonczer_ has quit IRC | 23:37 | |
| *** timsim has quit IRC | 23:37 | |
| *** EricGonczer_ has joined #openstack-dns | 23:37 | |
| *** vinod1 has quit IRC | 23:41 | |
| *** EricGonczer_ has quit IRC | 23:45 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!