-1D - RGW Geo-Replication and Disaster Recovery » History » Version 1
Jessica Mack, 06/22/2015 05:44 AM
1 | 1 | Jessica Mack | h1. -1D - RGW Geo-Replication and Disaster Recovery |
---|---|---|---|
2 | |||
3 | <pre> |
||
4 | *** rturk is now known as rturk-away 09:32 |
||
5 | *** davidzlap has joined #ceph-summit1 09:32 |
||
6 | *** davidzlap has left #ceph-summit1 09:32 |
||
7 | *** rturk-away is now known as rturk 09:33 |
||
8 | scuttlemonkey florian, want in to speak on this one? 09:33 |
||
9 | gregaf fghaas: ^ 09:33 |
||
10 | loicd ccourtaut: is having troubles with his video connection 09:33 |
||
11 | loicd video / audio 09:34 |
||
12 | fghaas I'm here, yes :) 09:34 |
||
13 | fghaas it's ok for me to lurk though :) 09:34 |
||
14 | scuttlemonkey k 09:34 |
||
15 | ccourtaut i'm currently having problem between my isp/google services :/ 09:34 |
||
16 | *** wwformat has quit IRC 09:35 |
||
17 | *** henrycc has quit IRC 09:35 |
||
18 | loicd http://wiki.ceph.com/01Planning/02Blueprints/Dumpling/RGW_Geo-Replication_and_Disaster_Recovery 09:36 |
||
19 | loicd http://pad.ceph.com/p/RADOS_Gateway_refactor 09:36 |
||
20 | *** liwang has quit IRC 09:36 |
||
21 | gregaf loicd: hey, I finally have video of you and it's a beautiful hallway :P 09:38 |
||
22 | *** yehuda_hm has quit IRC 09:39 |
||
23 | lmb Are the replicas consistent? (As in, they represent a valid, consistent state of the object at any given time, even if outdated) 09:39 |
||
24 | *** yehuda_hm has joined #ceph-summit1 09:40 |
||
25 | sagewk lmb: yes 09:40 |
||
26 | sagewk the (replicated) updates are atomic 09:40 |
||
27 | sagewk but not necessarily ordered (across objects) 09:40 |
||
28 | loicd gregaf: there you see me for real ;-) 09:41 |
||
29 | lmb sagewk: aye 09:41 |
||
30 | rturk hi, loic! 09:41 |
||
31 | loicd :-D 09:41 |
||
32 | loicd it would be useful to have links to the http://tracker.ceph.com/projects/ceph issues related to http://wiki.ceph.com/01Planning/02Blueprints/Dumpling/RGW_Geo-Replication_and_Disaster_Recovery in the pad 09:45 |
||
33 | dmick loicd: add them! :) 09:45 |
||
34 | rturk loicd: I'm working on a dekiwiki -> redmine plugin 09:45 |
||
35 | rturk so you'll be able to embed them the way you embed etherpads 09:45 |
||
36 | rturk as part of the post-summit work, I planned to share it so ppl can cross-link their work items 09:46 |
||
37 | paravoid they're not new 09:46 |
||
38 | paravoid but it doesn't really work 09:46 |
||
39 | paravoid I opened several bugs about it (incl. performance ones) and they said that they weren't really well designed 09:47 |
||
40 | paravoid they're implementing geo-replication themselves and they're not reusing this code for this 09:47 |
||
41 | paravoid that was on swift 1.7.x, I don't know how it has evolved since 09:48 |
||
42 | joao loicd, wave for us 09:48 |
||
43 | scuttlemonkey paravoid: so was the feeling you got that they were abandoning that code? 09:48 |
||
44 | scuttlemonkey or just not repurposing it for geo-replication 09:48 |
||
45 | loicd :-) 09:48 |
||
46 | paravoid I don't think they're abandoning it since it's there, but it's really for a different purpose from the get go 09:49 |
||
47 | michael_dreamhost the swift-sync seems like a higher level user controrable interface, rather than a cluster policy 09:49 |
||
48 | *** liwang has joined #ceph-summit1 09:50 |
||
49 | scuttlemonkey makes sense 09:50 |
||
50 | paravoid the problem with not doing it in the rados layer is that we'll end up with having 6 copies for each object 09:50 |
||
51 | joshd fghaas: glance has their own plans for doing data transfer between regions 09:51 |
||
52 | gregaf paravoid: not necessarily, you could do 2+1 for instance 09:51 |
||
53 | paravoid well, not really 09:51 |
||
54 | loicd http://tracker.ceph.com/projects/rgw/search?utf8=%E2%9C%93&issues=1&q=multisite does that cover it ? 09:51 |
||
55 | paravoid because if I have two lost drives at the same time on one of the clusters 09:51 |
||
56 | paravoid (probable enough) 09:51 |
||
57 | paravoid the pg would be lost 09:52 |
||
58 | paravoid and radosgw replication won't recover it 09:52 |
||
59 | hskinner Please consider geo-dispersion of erasure-encoded objects as you design this 09:53 |
||
60 | fghaas gregaf: feel free to correct me if my summary of your reply is incorrect in the pad 09:53 |
||
61 | gregaf good enough for me, but I'm blunter than they are ;) 09:53 |
||
62 | scuttlemonkey hskinner: that would be a good thing to add to the etherpad as an "other" or something 09:54 |
||
63 | scuttlemonkey perhaps under coding tasks and yehuda can move it around where he wants 09:54 |
||
64 | *** chee has quit IRC 09:54 |
||
65 | fghaas gregaf: better? 09:55 |
||
66 | gregaf lol, I did *not* say that 09:55 |
||
67 | *** chee has joined #ceph-summit1 09:57 |
||
68 | pioto so, this geo replication being discussed is up at the rgw layer, and not, say, rados? 09:57 |
||
69 | scuttlemonkey pioto: yes 09:57 |
||
70 | pioto meaning, it won't be able to support cephfs or rbd, then? hm 09:58 |
||
71 | scuttlemonkey pioto: the hope is to eventually move it down the stack 09:58 |
||
72 | pioto ok 09:58 |
||
73 | fghaas pioto: I'm sure gregaf has a "blunt" comment for you ;) 09:58 |
||
74 | scuttlemonkey pioto: I think it's more about _using_ rgw to sync all data 09:58 |
||
75 | lmb scuttlemonkey: the "ain't gonna happen" comment on the pad? ;-) 09:58 |
||
76 | scuttlemonkey not limit it to rgw 09:58 |
||
77 | nwl pioto: the driver is disaster recovery. you can use incremental snapshots of rbd with export/import between clusters to do RBD DR 09:58 |
||
78 | scuttlemonkey but someone else can clarify 09:58 |
||
79 | fghaas lmb: I just attributed that one :) 09:58 |
||
80 | pioto nwl: yeah. i thin kthere's also whole-cluster snapshots i saw somewhere 09:59 |
||
81 | loicd what task would you advise a new coder willing to help to try to help with the implementation of geo replication ? 09:59 |
||
82 | pioto but i dunno if there was a good way to replicatte that directly to another cluster.. .think it wanted a whole giant local directory to export to first? 09:59 |
||
83 | lmb scuttlemonkey: access must go through rgw though, otherwise it can't log/journal, right? so you can't use rgw-replication on top of an arbitrary block object 09:59 |
||
84 | fghaas lmb: that is the plan, as it stands 10:00 |
||
85 | paravoid will it scale? 10:00 |
||
86 | paravoid :) 10:00 |
||
87 | paravoid for both large amount of containers and objects 10:01 |
||
88 | lmb fghaas: too bad, misses our use case, but clearly its a great first step 10:01 |
||
89 | fghaas lmb, I know -- surely there's plenty of folks that would love to see this in rbd 10:01 |
||
90 | fghaas which is what I guess you're relating to 10:02 |
||
91 | paravoid sagewk: ^ 10:02 |
||
92 | rturk paravoid: I think there's a min or so delay w/the feed FYI 10:02 |
||
93 | lmb fghaas: yeah. similarly though, perhaps it could be implemented in the client-side libraries 10:02 |
||
94 | fghaas librbd I/O multiplex? joshd? 10:02 |
||
95 | *** nwat has quit IRC 10:03 |
||
96 | *** nwat has joined #ceph-summit1 10:04 |
||
97 | *** nwat has quit IRC 10:04 |
||
98 | </pre> |