Hi Casey, thanks for this info. It’s been doing something for 36 hours, but not 
updating the status at all. So it either takes a really long time for 
“preparing for full sync” or I’m doing something wrong. This is helpful 
information, but there’s a myriad of states that the system could be in. 

With that, I’m going to set up a lab rig and see if I can build a fully 
replicated state. At that point, I’ll have a better understanding of what a 
working system responds like and maybe I can at least ask better questions, 
hopefully figure it out myself. 

Thanks again! Brian

> On Apr 16, 2019, at 08:38, Casey Bodley <[email protected]> wrote:
> 
> Hi Brian,
> 
> On 4/16/19 1:57 AM, Brian Topping wrote:
>>> On Apr 15, 2019, at 5:18 PM, Brian Topping <[email protected] 
>>> <mailto:[email protected]>> wrote:
>>> 
>>> If I am correct, how do I trigger the full sync?
>> 
>> Apologies for the noise on this thread. I came to discover the 
>> `radosgw-admin [meta]data sync init` command. That’s gotten me with 
>> something that looked like this for several hours:
>> 
>>> [root@master ~]# radosgw-admin  sync status
>>>           realm 54bb8477-f221-429a-bbf0-76678c767b5f (example)
>>>       zonegroup 8e33f5e9-02c8-4ab8-a0ab-c6a37c2bcf07 (us)
>>>            zone b6e32bc8-f07e-4971-b825-299b5181a5f0 (secondary)
>>>   metadata sync preparing for full sync
>>>                 full sync: 64/64 shards
>>>                 full sync: 0 entries to sync
>>>                 incremental sync: 0/64 shards
>>>                 metadata is behind on 64 shards
>>>                 behind shards: 
>>> [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63]
>>>       data sync source: 35835cb0-4639-43f4-81fd-624d40c7dd6f (master)
>>>                         preparing for full sync
>>>                         full sync: 1/128 shards
>>>                         full sync: 0 buckets to sync
>>>                         incremental sync: 127/128 shards
>>>                         data is behind on 1 shards
>>>                         behind shards: [0]
>> 
>> I also had the data sync showing a list of “behind shards”, but both of them 
>> sat in “preparing for full sync” for several hours, so I tried 
>> `radosgw-admin [meta]data sync run`. My sense is that was a bad idea, but 
>> neither of the commands seem to be documented and the thread I found them on 
>> indicated they wouldn’t damage the source data.
>> 
>> QUESTIONS at this point:
>> 
>> 1) What is the best sequence of commands to properly start the sync? Does 
>> init just set things up and do nothing until a run is started?
> The sync is always running. Each shard starts with full sync (where it lists 
> everything on the remote, and replicates each), then switches to incremental 
> sync (where it polls the replication logs for changes). The 'metadata sync 
> init' command clears the sync status, but this isn't synchronized with the 
> metadata sync process running in radosgw(s) - so the gateways need to restart 
> before they'll see the new status and restart the full sync. The same goes 
> for 'data sync init'.
>> 2) Are there commands I should run before that to clear out any previous bad 
>> runs?
> Just restart gateways, and you should see progress via 'sync status'.
>> 
>> *Thanks very kindly for any assistance. *As I didn’t really see any 
>> documentation outside of setting up the realms/zones/groups, it seems like 
>> this would be useful information for others that follow.
>> 
>> best, Brian
>> 
>> _______________________________________________
>> ceph-users mailing list
>> [email protected]
>> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
> _______________________________________________
> ceph-users mailing list
> [email protected]
> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to