| *** bbowen has quit IRC | 00:29 | |
| *** ozzzo has quit IRC | 00:39 | |
| *** gyee has quit IRC | 00:43 | |
| *** markvoelker has quit IRC | 00:57 | |
| *** xenos76 has quit IRC | 01:13 | |
| *** Kuirong has joined #openstack | 01:29 | |
| Kuirong | hi all, I would like to know if I want install OpenStack stein version in CentOS, should I enable the OpenStack Stein repository? | 01:30 |
|---|---|---|
| Kuirong | because I didn't find "centos-release-openstack-stein" in the Installation guide(https://docs.openstack.org/install-guide/environment-packages-rdo.html) | 01:30 |
| cz2 | Kuirong: http://mirror.centos.org/centos/7/extras/x86_64/Packages/centos-release-openstack-stein-1-1.el7.centos.noarch.rpm | 01:38 |
| cz2 | Kuirong: it's in CentOS-Extras repo | 01:38 |
| cz2 | make sure it's enabled | 01:38 |
| cz2 | most likely documentation is not updated | 01:42 |
| *** aakarsh|2 has quit IRC | 01:43 | |
| Kuirong | thanks cz2 | 01:44 |
| *** a0s510 has quit IRC | 01:49 | |
| *** markvoelker has joined #openstack | 01:52 | |
| *** markvoelker has quit IRC | 02:02 | |
| *** markvoelker has joined #openstack | 02:03 | |
| *** a0s510 has joined #openstack | 02:10 | |
| *** spsurya has joined #openstack | 02:11 | |
| *** markvoelker has quit IRC | 02:13 | |
| *** morazi has quit IRC | 02:13 | |
| *** zaitcev has joined #openstack | 02:13 | |
| *** markvoelker has joined #openstack | 02:14 | |
| *** markvoelker has quit IRC | 02:24 | |
| *** a0s510 has quit IRC | 02:41 | |
| *** bbowen has joined #openstack | 02:44 | |
| *** BXS has joined #openstack | 03:04 | |
| *** Kuirong has quit IRC | 03:05 | |
| *** maddtux has joined #openstack | 03:06 | |
| *** ianychoi has quit IRC | 03:29 | |
| *** ianychoi has joined #openstack | 03:30 | |
| *** gkadam has joined #openstack | 03:36 | |
| *** gkadam has quit IRC | 03:36 | |
| *** dsneddon has quit IRC | 03:51 | |
| *** arnoldoree has quit IRC | 03:52 | |
| *** markvoelker has joined #openstack | 04:05 | |
| *** zaitcev has quit IRC | 04:06 | |
| *** markvoelker has quit IRC | 04:10 | |
| *** zaitcev has joined #openstack | 04:20 | |
| *** dsneddon has joined #openstack | 04:24 | |
| *** BXS has quit IRC | 04:28 | |
| *** darvon has quit IRC | 04:40 | |
| *** gshippey_ has quit IRC | 04:40 | |
| *** darvon has joined #openstack | 04:41 | |
| *** brault has joined #openstack | 04:50 | |
| *** brault has quit IRC | 04:54 | |
| *** soniya29 has joined #openstack | 04:54 | |
| *** dr_feelgood has joined #openstack | 05:05 | |
| *** odicha has joined #openstack | 05:08 | |
| *** zhubx has joined #openstack | 05:09 | |
| *** dr_feelgood has quit IRC | 05:09 | |
| *** boxiang has quit IRC | 05:10 | |
| *** markvoelker has joined #openstack | 05:20 | |
| *** markvoelker has quit IRC | 05:25 | |
| *** sauvin has joined #openstack | 05:26 | |
| *** jhesketh has quit IRC | 05:34 | |
| *** dr_feelgood has joined #openstack | 05:38 | |
| *** shyamb has joined #openstack | 05:39 | |
| *** yaawang has quit IRC | 05:39 | |
| *** yaawang has joined #openstack | 05:40 | |
| *** Warped has joined #openstack | 05:41 | |
| *** jhesketh has joined #openstack | 05:42 | |
| *** dr_feelgood has quit IRC | 05:43 | |
| *** shyamb has quit IRC | 05:44 | |
| *** links has joined #openstack | 05:46 | |
| gregwork | does tripleo/rdo/osp support policy route configurations with os-net-config ? | 05:46 |
| *** miloa has joined #openstack | 05:46 | |
| *** ymasson has quit IRC | 05:47 | |
| gregwork | im trying to figure out how to deploy a network config that will correctly set ip route/rule rules for multi homed servers | 05:48 |
| gregwork | ive got this lovely asymmetric routing issue because both vlan101 and vlan103 on my controllers can both reach vlan71 on my switch via their svi's and the default route is vlan101 .. so packets come in vlan103 and go out vlan101. This of course gets logged as martian traffic and dropped until i set rp_filter to 2 .. but id rather figure out how to push an ip rule/route map that will survive a stack update | 05:49 |
| *** stamps has quit IRC | 05:50 | |
| *** lpetrut has joined #openstack | 06:00 | |
| *** Kuirong has joined #openstack | 06:06 | |
| *** dsneddon has quit IRC | 06:14 | |
| *** zhubx has quit IRC | 06:18 | |
| *** zhubx has joined #openstack | 06:18 | |
| *** surpatil has joined #openstack | 06:18 | |
| *** shyamb has joined #openstack | 06:20 | |
| *** jtomasek has joined #openstack | 06:26 | |
| *** slaweq has joined #openstack | 06:27 | |
| *** zhubx has quit IRC | 06:28 | |
| *** zhubx has joined #openstack | 06:28 | |
| *** yaawang has quit IRC | 06:31 | |
| *** slaweq has quit IRC | 06:32 | |
| *** yaawang has joined #openstack | 06:33 | |
| *** dgurtner has joined #openstack | 06:36 | |
| *** dsneddon has joined #openstack | 06:42 | |
| *** dsneddon has quit IRC | 06:47 | |
| *** dsneddon has joined #openstack | 06:48 | |
| *** slaweq has joined #openstack | 06:48 | |
| *** macz has joined #openstack | 06:52 | |
| *** dsneddon has quit IRC | 06:54 | |
| *** dsneddon has joined #openstack | 06:54 | |
| *** macz has quit IRC | 06:56 | |
| *** avivgt has joined #openstack | 06:59 | |
| *** dsneddon has quit IRC | 07:00 | |
| *** mlycka has joined #openstack | 07:00 | |
| *** dsneddon has joined #openstack | 07:00 | |
| *** crazzy has joined #openstack | 07:00 | |
| *** zhubx has quit IRC | 07:01 | |
| *** zhubx has joined #openstack | 07:01 | |
| *** avivgta has joined #openstack | 07:01 | |
| *** dsneddon has quit IRC | 07:06 | |
| *** dsneddon has joined #openstack | 07:07 | |
| *** trident has quit IRC | 07:08 | |
| *** zhubx has quit IRC | 07:11 | |
| *** zhubx has joined #openstack | 07:11 | |
| *** rcernin has quit IRC | 07:12 | |
| yoctozepto | watersj: hey, re: kolla, we have fixed that in https://review.opendev.org/677453 - though it only applies to cluster expansion and some cases of redeployments after failure, clean deploys worked normally | 07:13 |
| yoctozepto | let me know whether that fixes your problem | 07:13 |
| yoctozepto | also, join us on #openstack-kolla, we are more vigilant in there :-) | 07:13 |
| *** trident has joined #openstack | 07:17 | |
| *** zhubx has quit IRC | 07:20 | |
| *** zhubx has joined #openstack | 07:20 | |
| *** fsimonce has joined #openstack | 07:25 | |
| *** mrechtma__ has joined #openstack | 07:27 | |
| *** ivve has joined #openstack | 07:28 | |
| *** mrechtma__ has quit IRC | 07:29 | |
| *** mrechtma__ has joined #openstack | 07:29 | |
| *** avivgt has quit IRC | 07:29 | |
| *** zhubx has quit IRC | 07:30 | |
| *** zhubx has joined #openstack | 07:30 | |
| *** shyamb has quit IRC | 07:42 | |
| *** aojea has joined #openstack | 07:42 | |
| *** zaitcev has quit IRC | 07:50 | |
| *** aojeagarcia has joined #openstack | 08:00 | |
| *** aojea has quit IRC | 08:01 | |
| *** zaitcev has joined #openstack | 08:02 | |
| *** jangutter has joined #openstack | 08:03 | |
| *** avivgt has joined #openstack | 08:05 | |
| *** dgurtner has quit IRC | 08:05 | |
| *** zaitcev_ has joined #openstack | 08:06 | |
| *** mrechtma__ has quit IRC | 08:07 | |
| *** zaitcev has quit IRC | 08:09 | |
| *** dgurtner has joined #openstack | 08:11 | |
| *** suuuper has joined #openstack | 08:14 | |
| *** alexmcleod has joined #openstack | 08:22 | |
| *** zhubx has quit IRC | 08:26 | |
| *** zhubx has joined #openstack | 08:27 | |
| *** zhubx has quit IRC | 08:27 | |
| *** boxiang has joined #openstack | 08:29 | |
| *** imega has joined #openstack | 08:29 | |
| *** zaitcev_ has quit IRC | 08:30 | |
| *** shyamb has joined #openstack | 08:33 | |
| *** boxiang has quit IRC | 08:33 | |
| *** boxiang has joined #openstack | 08:34 | |
| *** dsneddon has quit IRC | 08:38 | |
| *** zaitcev_ has joined #openstack | 08:43 | |
| *** phyrox has joined #openstack | 08:48 | |
| *** dgurtner has quit IRC | 08:56 | |
| *** jraju__ has joined #openstack | 08:56 | |
| *** links has quit IRC | 08:57 | |
| *** xenos76 has joined #openstack | 09:02 | |
| *** xenos76 has quit IRC | 09:07 | |
| *** xenos76 has joined #openstack | 09:09 | |
| *** dsneddon has joined #openstack | 09:12 | |
| *** brault has joined #openstack | 09:17 | |
| *** dsneddon has quit IRC | 09:17 | |
| *** shyamb has quit IRC | 09:22 | |
| *** phyrox has quit IRC | 09:27 | |
| *** abirba has quit IRC | 09:41 | |
| *** dgurtner has joined #openstack | 09:43 | |
| *** links has joined #openstack | 09:51 | |
| *** jraju__ has quit IRC | 09:52 | |
| *** zaitcev_ has quit IRC | 09:55 | |
| *** zhubx has joined #openstack | 10:00 | |
| *** sshnaidm|afk is now known as sshnaidm | 10:02 | |
| *** zhubx has quit IRC | 10:03 | |
| *** zhubx has joined #openstack | 10:03 | |
| *** boxiang has quit IRC | 10:03 | |
| *** zhubx has quit IRC | 10:04 | |
| *** zhubx has joined #openstack | 10:04 | |
| *** macz has joined #openstack | 10:05 | |
| *** zhubx has quit IRC | 10:06 | |
| *** zhubx has joined #openstack | 10:07 | |
| *** zaitcev_ has joined #openstack | 10:07 | |
| *** zhubx has quit IRC | 10:08 | |
| *** zhubx has joined #openstack | 10:09 | |
| *** macz has quit IRC | 10:09 | |
| *** zhubx has quit IRC | 10:09 | |
| *** zhubx has joined #openstack | 10:10 | |
| *** shyamb has joined #openstack | 10:10 | |
| *** dsneddon has joined #openstack | 10:18 | |
| *** melsakhawy has joined #openstack | 10:26 | |
| *** arnoldoree has joined #openstack | 10:31 | |
| *** melsakhawy has quit IRC | 10:32 | |
| *** forgotmynick has joined #openstack | 10:38 | |
| *** shyamb has quit IRC | 10:41 | |
| *** shyamb has joined #openstack | 11:00 | |
| *** tesseract has joined #openstack | 11:12 | |
| *** tonythomas has joined #openstack | 11:12 | |
| LarsErikP | Is there still no client support for neutron port forwarding? | 11:16 |
| LarsErikP | Can't really find much info, besides this pretty old RFE https://bugs.launchpad.net/neutron/+bug/1811352 | 11:16 |
| openstack | Launchpad bug 1811352 in neutron "[RFE] Include neutron CLI floatingip port-forwarding support" [High,In progress] - Assigned to LIU Yulong (dragon889) | 11:16 |
| slaweq | LarsErikP: it seems that this should be added with https://review.opendev.org/#/c/650062/ but it's still not finished | 11:18 |
| *** zaitcev_ has quit IRC | 11:20 | |
| LarsErikP | wow.. that was a pretty long history :p | 11:21 |
| *** dudek has joined #openstack | 11:32 | |
| *** zaitcev_ has joined #openstack | 11:32 | |
| *** Lucas_Gray has joined #openstack | 11:33 | |
| *** dsneddon has quit IRC | 11:36 | |
| *** xenos76 has quit IRC | 11:39 | |
| *** mcornea has joined #openstack | 11:39 | |
| *** shyamb has quit IRC | 11:41 | |
| *** shyam89 has joined #openstack | 11:41 | |
| *** markvoelker has joined #openstack | 11:43 | |
| *** dsneddon has joined #openstack | 11:45 | |
| *** shyam89 has quit IRC | 11:45 | |
| *** morazi has joined #openstack | 11:50 | |
| *** surpatil has quit IRC | 12:01 | |
| *** shyamb has joined #openstack | 12:03 | |
| *** maddtux has quit IRC | 12:40 | |
| *** shyamb has quit IRC | 12:42 | |
| *** ttsiouts has joined #openstack | 12:43 | |
| *** shyamb has joined #openstack | 12:47 | |
| *** soniya29 has quit IRC | 12:50 | |
| *** surpatil has joined #openstack | 12:51 | |
| *** shyamb has quit IRC | 12:52 | |
| *** mcriswel1 has joined #openstack | 13:01 | |
| *** xenos76 has joined #openstack | 13:05 | |
| *** surpatil has quit IRC | 13:12 | |
| *** RickDeckard has joined #openstack | 13:15 | |
| *** Kuirong has quit IRC | 13:16 | |
| *** forgotmynick has quit IRC | 13:17 | |
| *** dsneddon has quit IRC | 13:20 | |
| *** dsneddon has joined #openstack | 13:28 | |
| *** dsneddon has quit IRC | 13:33 | |
| *** ScrumpyJack has joined #openstack | 13:34 | |
| ScrumpyJack | in a swift cluster, a node had to be rebuilt and drives wiped. it's been out of the cluster for over a week. what's the best way to re-introduce it back into the cluster? | 13:36 |
| ScrumpyJack | ringfiles have not been touch since the node was wiped | 13:37 |
| DHE | make sure that you also formatted all the drives referenced in the rings, not just the OS drives | 13:37 |
| *** Goneri has joined #openstack | 13:42 | |
| ScrumpyJack | yes, I have done that | 13:43 |
| ScrumpyJack | but starting up swift on that node seems to bare a lot of load on the swift cluster | 13:44 |
| ScrumpyJack | will swift try to move all the container/objects back to that node? | 13:44 |
| ScrumpyJack | (after weeks on that node being down) | 13:45 |
| *** mcriswel1 has quit IRC | 13:46 | |
| *** mcriswell has joined #openstack | 13:46 | |
| *** ttsiouts has quit IRC | 13:48 | |
| *** ttsiouts has joined #openstack | 13:48 | |
| *** odicha has quit IRC | 13:50 | |
| ScrumpyJack | should i remove the node from the ring files and have swift tread the rebuilt node as new? | 13:52 |
| *** ttsiouts has quit IRC | 13:53 | |
| DHE | well it's rebuilding all the data it's missing, so yeah I can see it getting slammed | 13:56 |
| DHE | maybe you just want to bring the weight to a small number like 1/10 its normal value and raise it slowly over a few days. | 13:56 |
| *** DK2 has joined #openstack | 13:57 | |
| ScrumpyJack | is it best to do that on all the disks on that node? or perhaps just add a few disks at a time with a 1/10th weight? | 13:57 |
| *** aojeagarcia has quit IRC | 13:58 | |
| DHE | depends on your capacity I guess. this would be a network issue | 14:03 |
| *** dsneddon has joined #openstack | 14:07 | |
| ScrumpyJack | a node is 1PB | 14:08 |
| ScrumpyJack | and swift-recon shows 30 nodes | 14:09 |
| *** aakarsh has joined #openstack | 14:13 | |
| *** boxiang has joined #openstack | 14:14 | |
| *** zhubx has quit IRC | 14:17 | |
| *** sshnaidm has quit IRC | 14:17 | |
| *** sshnaidm has joined #openstack | 14:19 | |
| DHE | impressive... | 14:20 |
| *** boxiang has quit IRC | 14:20 | |
| ScrumpyJack | the node has been out of the cluster for a long time, so swift has created a new third copy of each object right? So If i remove the node from the ringfiles and rebalance, what happends? | 14:20 |
| *** boxiang has joined #openstack | 14:21 | |
| ScrumpyJack | do those objects (the third copy, presumably offloaded somewhere) move somewhere else again? or does swift just mark those third copies as "good, you're staying here now" | 14:23 |
| ScrumpyJack | DHE: I used the term exabyte for the first time the other day. It was only half an exabyte, but still ... :) | 14:25 |
| DHE | and here i am looking at my 1/2 petabyte ZFS array and thinking it's big.... | 14:25 |
| *** boxiang has quit IRC | 14:25 | |
| *** boxiang has joined #openstack | 14:26 | |
| *** markvoelker has quit IRC | 14:26 | |
| ScrumpyJack | so when a node dies, after a certain time (a week?) swift will create a new third copy of all the objects on that node right? | 14:27 |
| ScrumpyJack | onto other nodes in the cluster | 14:27 |
| DHE | the risk is that when a file is deleted, swift writes a tombstone object to disk so replication has something physical to look at and understand the point in time a delete happened. but that tombstone is only kept around for a short while, usually about a week | 14:28 |
| ScrumpyJack | and when i add the node back (blank drives) swift will copy those objects back right? | 14:28 |
| DHE | so if a machine is down for too long and brought back with its disks intact, you can have zombie objects taking up space but not supposed to be there, not in container listings, etc. | 14:29 |
| DHE | this is why I said you should wipe all disks, not just the OS | 14:29 |
| *** xubozhang has joined #openstack | 14:29 | |
| *** markvoelker has joined #openstack | 14:29 | |
| *** xubozhang has quit IRC | 14:30 | |
| DHE | if a node is down, replication will select a machine to act as the temporary replacement, and when the node returns it will be sync'd up | 14:30 |
| *** hukyld has joined #openstack | 14:30 | |
| *** aconole has joined #openstack | 14:32 | |
| ScrumpyJack | right, but can make the temp replacement permanent and avoid the sync up? | 14:33 |
| ScrumpyJack | (when i had the node with blank disks back) | 14:33 |
| *** gillesMo has quit IRC | 14:34 | |
| *** miloa has quit IRC | 14:34 | |
| *** miloa has joined #openstack | 14:36 | |
| *** markvoelker has quit IRC | 14:37 | |
| *** markvoelker has joined #openstack | 14:38 | |
| *** gagan662 has quit IRC | 14:39 | |
| *** gagan662 has joined #openstack | 14:40 | |
| *** Lucas_Gray has quit IRC | 14:51 | |
| *** lpetrut has quit IRC | 14:52 | |
| *** noogie has quit IRC | 14:53 | |
| *** igordc has joined #openstack | 14:57 | |
| *** links has quit IRC | 15:01 | |
| *** xenos76 has quit IRC | 15:03 | |
| *** zaitcev_ has quit IRC | 15:04 | |
| timburke | oh, this sounds like a fun conversation :-) | 15:07 |
| timburke | "the node has been out of the cluster for a long time, so swift has created a new third copy of each object right?" yes, as i recall. definitely if the node was up but had drives unmounted | 15:08 |
| timburke | (the main question in my mind is whether we'd replicate to a handoff if the primary is temporarily unavailable, but certainly all new data, swift will try to write three replicas wherever it can) | 15:09 |
| timburke | "do those objects (the third copy, presumably offloaded somewhere) move somewhere else again?" yes -- swift's ring has a notion of "primary assignments" for objects; if a primary isn't available for a PUT, it'll start going through the remaining devices in a consistent order and write to a "handoff" location | 15:11 |
| DHE | but you can't really take a host out of the ring and assure what use to be handoffs are promoted to primary nodes in the order they would be used. | 15:12 |
| timburke | when the replicator comes around and finds the handoff data, it'll try to move it back to the primaries | 15:12 |
| timburke | DHE, absolutely correct | 15:13 |
| *** PagliaccisCloud has joined #openstack | 15:13 | |
| timburke | in short, there's got to be a bunch of movement regardless of what you do | 15:14 |
| DHE | and my weight thing was about making it a bit slow as the drives re-populate with the data they're supposed to have, hopefully keeping the disks from getting too slammed with requests during replication | 15:14 |
| *** lpetrut has joined #openstack | 15:15 | |
| *** zaitcev_ has joined #openstack | 15:16 | |
| timburke | yeah -- but the rebalance is going to move traffic elsewhere. it'll probably improve your queue depths on the freshly-formatted node, but i'm not sure it'll get the cluster to the desired end state any faster | 15:16 |
| *** xenos76 has joined #openstack | 15:17 | |
| *** markvoelker has quit IRC | 15:18 | |
| timburke | so, ScrumpyJack -- at this point i'd recommend leaving the rings alone. there's a bunch of load in your swift cluster because there's some pent up work that we've been putting off. one way or another, it needs to get resolved | 15:18 |
| *** mcriswell has quit IRC | 15:19 | |
| *** markvoelker has joined #openstack | 15:21 | |
| timburke | if it starts to impact client-facing traffic, you might consider tuning down the replicators -- but personally, i'd be partial toward thinking of the client impact as back-pressure to give the cluster enough headroom that it can get back to full health | 15:21 |
| timburke | unfortunately, it takes time to move 1PB ;-) | 15:22 |
| *** ivve has quit IRC | 15:24 | |
| *** gyee has joined #openstack | 15:26 | |
| * ScrumpyJack nods | 15:26 | |
| *** ymasson has joined #openstack | 15:27 | |
| *** dsneddon has quit IRC | 15:27 | |
| watersj | is there a way to import images contained in ceph to a new openstack deployment ? | 15:29 |
| gregwork | new openstack has ceph ? | 15:30 |
| gregwork | if so then you could rbd-mirror | 15:30 |
| watersj | yes, new openstack, existing ceph | 15:31 |
| watersj | images contained were from a (cough , messed up install) | 15:31 |
| gregwork | oh | 15:31 |
| gregwork | ceph is external to openstack i take it | 15:31 |
| gregwork | eg. not managed ceph | 15:31 |
| watersj | yes, external storage | 15:32 |
| *** RickDeckard has quit IRC | 15:32 | |
| gregwork | i mean you define the pools where things are, have you tried pointing at your old pools ? (maybe after making a copy of them) | 15:32 |
| watersj | getting ready to re-deploy kolla-ansible | 15:33 |
| watersj | I guess i'll find out soon | 15:33 |
| gregwork | so full disclosure im only really versed in tripleO based openstack installs | 15:33 |
| gregwork | so this might not work at all for you :) | 15:33 |
| gregwork | theoretically tho .. ceph is ceph | 15:34 |
| watersj | thinking worst case, i might need to do some manual db edits in cinder | 15:34 |
| ScrumpyJack | timburke: are there any docs about "tuning down the replicators"? | 15:34 |
| watersj | or I export and reimport | 15:34 |
| watersj | from ceph | 15:35 |
| watersj | gregwork, why did you choose triple0 for you deployment tool? | 15:36 |
| *** niceplace_ has joined #openstack | 15:39 | |
| gregwork | it made a lot of sense for doing this in prod | 15:40 |
| timburke | ScrumpyJack, you can fiddle with concurrency/workers: https://github.com/openstack/swift/blob/master/etc/object-server.conf-sample#L240-L248 and ionice: https://github.com/openstack/swift/blob/master/etc/object-server.conf-sample#L309-L319 | 15:40 |
| gregwork | template driven deployment with heat managing ansible/puppet, a big name vendor using it as their distro base .. the ability to scale up/down easily | 15:41 |
| timburke | ScrumpyJack, various replication-related settings on the server may be useful, too: https://github.com/openstack/swift/blob/master/etc/object-server.conf-sample#L170-L194 | 15:42 |
| ScrumpyJack | and if i were to leave the ring files alone, and just add the node back, it keep it load low, i should add, say, 10 disks at a time? | 15:42 |
| watersj | kolla seemed to have ability to scale easily, issue I had was with mariadb cluster was hosed | 15:43 |
| *** dsneddon has joined #openstack | 15:44 | |
| timburke | ScrumpyJack, IME more spindles makes for a happier cluster -- i'd be inclined to just add them all back. you could start by just mounting a few at a time, though, see how the system reacts | 15:46 |
| timburke | do you have a separate replication network? | 15:46 |
| *** sshnaidm is now known as sshnaidm|afk | 15:49 | |
| ScrumpyJack | no seperate network | 15:50 |
| *** RickDeckard has joined #openstack | 15:50 | |
| timburke | might be worth looking into. the big advantage there is that you can saturate the replication network without much impacting client traffic | 15:55 |
| timburke | out of curiosity, what are you using swift for? i love hearing about people's use-cases :-) | 15:55 |
| *** imega has quit IRC | 15:58 | |
| *** zaitcev_ has quit IRC | 16:01 | |
| *** hukyld has left #openstack | 16:02 | |
| *** miloa has quit IRC | 16:03 | |
| *** RickDeckard has quit IRC | 16:03 | |
| *** tesseract has quit IRC | 16:08 | |
| *** med_ has quit IRC | 16:10 | |
| *** bobh has joined #openstack | 16:11 | |
| *** zaitcev_ has joined #openstack | 16:13 | |
| *** alexmcleod has quit IRC | 16:14 | |
| *** bobh has quit IRC | 16:15 | |
| *** bobh has joined #openstack | 16:16 | |
| *** bobh has quit IRC | 16:20 | |
| *** jathan has joined #openstack | 16:23 | |
| *** watersj has quit IRC | 16:25 | |
| *** donnyd has joined #openstack | 16:26 | |
| *** brault has quit IRC | 16:27 | |
| *** dgurtner has quit IRC | 16:28 | |
| *** iclon has quit IRC | 16:32 | |
| *** macz has joined #openstack | 16:33 | |
| gregwork | if you deploy a stack with a particular template version, can you update the template version of that stack with a stack update? | 16:37 |
| gregwork | im thinking no given my limited testing | 16:37 |
| gregwork | but was looking for other opnions | 16:37 |
| *** jonaspaulo has joined #openstack | 16:39 | |
| *** suuuper has quit IRC | 16:48 | |
| *** miloa has joined #openstack | 16:49 | |
| *** mlycka has quit IRC | 16:51 | |
| *** salmankhan has joined #openstack | 16:51 | |
| *** satanist has quit IRC | 16:52 | |
| *** Warped has quit IRC | 16:52 | |
| *** satanist has joined #openstack | 16:52 | |
| *** avivgta has quit IRC | 16:54 | |
| *** RickDeckard has joined #openstack | 17:02 | |
| *** RickDeckard has quit IRC | 17:07 | |
| *** RickDeckard has joined #openstack | 17:09 | |
| *** zaitcev_ has quit IRC | 17:12 | |
| *** salmankhan has quit IRC | 17:13 | |
| *** ozzzo has joined #openstack | 17:19 | |
| *** zaitcev_ has joined #openstack | 17:24 | |
| *** brault has joined #openstack | 17:30 | |
| *** miloa has quit IRC | 17:32 | |
| *** brault has quit IRC | 17:34 | |
| *** sajoupa has quit IRC | 17:35 | |
| *** shibboleth has joined #openstack | 17:43 | |
| *** brault has joined #openstack | 17:44 | |
| *** brault has quit IRC | 17:45 | |
| *** jmlowe has quit IRC | 17:51 | |
| *** zaitcev__ has joined #openstack | 17:55 | |
| *** shibboleth has quit IRC | 17:57 | |
| *** mmethot has quit IRC | 17:58 | |
| *** zaitcev_ has quit IRC | 17:58 | |
| *** mmethot has joined #openstack | 17:58 | |
| *** jmlowe has joined #openstack | 18:06 | |
| dsneddon | gregwork, Hi, I see in my scrollback that you were asking about policy-based routing with os-net-config last night. Did you get your answer? | 18:16 |
| dsneddon | gregwork, I added policy-based routing fairly recently, so you'll need a recent build. I can find the minimum package which has the code if you want. It's not really fully tested yet. | 18:17 |
| *** igordc has quit IRC | 18:23 | |
| *** phyrox has joined #openstack | 18:24 | |
| *** morazi has quit IRC | 18:27 | |
| *** RickDeckard has quit IRC | 18:48 | |
| *** zaitcev_ has joined #openstack | 18:51 | |
| *** bbowen has quit IRC | 18:51 | |
| *** zaitcev__ has quit IRC | 18:55 | |
| *** RickDeckard has joined #openstack | 18:56 | |
| *** lpetrut has quit IRC | 19:02 | |
| *** godlike has quit IRC | 19:08 | |
| *** godlike has joined #openstack | 19:08 | |
| *** godlike has quit IRC | 19:09 | |
| *** godlike has joined #openstack | 19:10 | |
| *** RickDeckard has quit IRC | 19:11 | |
| *** godlike has quit IRC | 19:16 | |
| *** godlike has joined #openstack | 19:16 | |
| *** zaitcev_ has quit IRC | 19:18 | |
| *** ivve has joined #openstack | 19:20 | |
| *** zaitcev_ has joined #openstack | 19:30 | |
| *** morazi has joined #openstack | 19:31 | |
| *** aakarsh has quit IRC | 19:35 | |
| *** PagliaccisCloud has quit IRC | 19:35 | |
| *** morazi has quit IRC | 19:38 | |
| gregwork | dsneddon: no i didnt | 19:42 |
| gregwork | dsneddon: hmmn so for now i guess i need to do a post config hook in my overcloud deploy | 19:43 |
| gregwork | im using redhat osp13 (queens) | 19:43 |
| gregwork | to create the ip route / ip rule entries | 19:43 |
| dsneddon | gregwork, Yeah, let me check and see if we backported the PBR to OSP 13. | 19:44 |
| dsneddon | gregwork, OK, it looks like it only made it into 15+. | 19:44 |
| *** sajoupa has joined #openstack | 19:45 | |
| dsneddon | gregwork, There were some workarounds, like adding the route table identifier in "route_options" of the route itself. Then you need to manually update the /etc/iproute2/rt_tables file separately or via a script that gets run during deploy. | 19:47 |
| dsneddon | gregwork, I know a few production OSP 13 deployments used that method, before one user requested that we make it possible to create route tables in os-net-config. | 19:47 |
| gregwork | yeah | 19:49 |
| gregwork | my network is breaking pretty badly without a proper policy config | 19:49 |
| gregwork | asymmetric and all that | 19:49 |
| gregwork | ive had to set rpfilter to 2 | 19:49 |
| gregwork | for the time being | 19:49 |
| *** morazi has joined #openstack | 19:56 | |
| *** gyee has quit IRC | 20:05 | |
| *** atmark has quit IRC | 20:07 | |
| *** mloza has joined #openstack | 20:08 | |
| *** slaweq has quit IRC | 20:09 | |
| *** zaitcev_ has quit IRC | 20:17 | |
| *** zhubx has joined #openstack | 20:20 | |
| *** gyee has joined #openstack | 20:23 | |
| *** boxiang has quit IRC | 20:24 | |
| *** aconole has quit IRC | 20:26 | |
| *** zaitcev_ has joined #openstack | 20:29 | |
| *** brault has joined #openstack | 20:32 | |
| *** jmlowe has quit IRC | 20:36 | |
| *** brault has quit IRC | 20:37 | |
| *** aakarsh has joined #openstack | 20:39 | |
| *** xenos76 has quit IRC | 20:53 | |
| *** bbowen has joined #openstack | 20:53 | |
| *** markvoelker has quit IRC | 20:58 | |
| *** markvoelker has joined #openstack | 20:58 | |
| *** belmoreira has joined #openstack | 20:59 | |
| *** markvoelker has quit IRC | 21:10 | |
| *** markvoelker has joined #openstack | 21:10 | |
| *** slaweq has joined #openstack | 21:11 | |
| *** Goneri has quit IRC | 21:15 | |
| *** zaitcev_ has quit IRC | 21:16 | |
| *** slaweq has quit IRC | 21:16 | |
| *** whyz has quit IRC | 21:16 | |
| *** belmoreira has quit IRC | 21:16 | |
| *** jmlowe has joined #openstack | 21:18 | |
| *** ircuser-1 has quit IRC | 21:22 | |
| *** aakarsh has quit IRC | 21:23 | |
| *** belmoreira has joined #openstack | 21:25 | |
| *** imega has joined #openstack | 21:27 | |
| *** markvoelker has quit IRC | 21:39 | |
| *** markvoelker has joined #openstack | 21:40 | |
| *** markvoelker has quit IRC | 21:45 | |
| *** dayou has quit IRC | 21:51 | |
| *** jbadiapa has quit IRC | 21:52 | |
| *** ivve has quit IRC | 22:03 | |
| *** dayou has joined #openstack | 22:05 | |
| *** markvoelker has joined #openstack | 22:05 | |
| *** dudek has quit IRC | 22:05 | |
| *** markvoelker has quit IRC | 22:11 | |
| *** slaweq has joined #openstack | 22:11 | |
| *** belmoreira has quit IRC | 22:11 | |
| *** ircuser-1 has joined #openstack | 22:15 | |
| *** fsimonce has quit IRC | 22:16 | |
| *** slaweq has quit IRC | 22:17 | |
| *** jonaspaulo has quit IRC | 22:24 | |
| *** avivgt has quit IRC | 22:33 | |
| *** notmyname has quit IRC | 22:38 | |
| *** mcornea has quit IRC | 22:43 | |
| *** notmyname has joined #openstack | 22:48 | |
| *** macz has quit IRC | 22:57 | |
| *** arnoldoree has quit IRC | 23:04 | |
| *** jathan has quit IRC | 23:10 | |
| *** slaweq has joined #openstack | 23:11 | |
| *** slaweq has quit IRC | 23:15 | |
| *** rcernin has joined #openstack | 23:21 | |
| *** PagliaccisCloud has joined #openstack | 23:25 | |
| *** imega has quit IRC | 23:30 | |
| *** phyrox has quit IRC | 23:30 | |
| *** Warped has joined #openstack | 23:49 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!