Hi Cephs
Several nodes of our Ceph 14.2.5 are fully dedicated to host cold storage /
backups information.
Today checking the data usage with a customer found that rgw-admin is reporting:
{
"bucket": "XXXXXX",
"tenant": "",
"zonegroup": "4d8c7c5f-ca40-4ee3-b5bb-b2cad90bd007",
"placement_rule": "default-placement",
"explicit_placement": {
"data_pool": "default.rgw.buckets.data",
"data_extra_pool": "default.rgw.buckets.non-ec",
"index_pool": "default.rgw.buckets.index"
},
"id": "48efb8c3-693c-4fe0-bbe4-fdc16f590a82.15946848.1",
"marker": "48efb8c3-693c-4fe0-bbe4-fdc16f590a82.3886182.18",
"index_type": "Normal",
"owner": "XXXXXXXX",
"ver": "0#410482,1#441516,2#401062,3#371595",
"master_ver": "0#0,1#0,2#0,3#0",
"mtime": "2019-06-08 00:26:06.266567Z",
"max_marker": "0#,1#,2#,3#",
"usage": {
"rgw.none": {
"size": 0,
"size_actual": 0,
"size_utilized": 0,
"size_kb": 0,
"size_kb_actual": 0,
"size_kb_utilized": 0,
"num_objects": 0
},
"rgw.main": {
"size": 5118399148914,
"size_actual": 5118401548288,
"size_utilized": 5118399148914,
"size_kb": 4998436669,
"size_kb_actual": 4998439012,
"size_kb_utilized": 4998436669,
"num_objects": 293083
},
"rgw.multimeta": {
"size": 0,
"size_actual": 0,
"size_utilized": 378,
"size_kb": 0,
"size_kb_actual": 0,
"size_kb_utilized": 1,
"num_objects": 1688
}
},
"bucket_quota": {
"enabled": false,
"check_on_raw": false,
"max_size": -1024,
"max_size_kb": 0,
"max_objects": -1
}
That's near 5TB used space in CEPH, and the external tools are reporting just
1.42TB.
Just in this case is more than a 300%. As the platform is billed by usage that
cause an internal problem with customers.
Our setup don't use EC nodes, all are replica. All nodes use 14.2.5. 6 SSD
fully dedicated to RGW-index .
No error at rgw logs or something that can explain this huge difference.
Magnitude in our case is that customer report us he use near 70-80TB in
multiple buckets, but our CEPH report 163TB.
Im planning to move out all the customer information to a NAS to cleanup this
bucket/space and re-upload but the process is not very transparent or smooth
for customer.
Suggestions accepted.
Regards
Manuel
_______________________________________________
ceph-users mailing list
[email protected]
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
