Hi,
--
I am trying to test disaster recovery of rados gateways.
I setup a federated architecture for rados gateway as explained in the docs.
I am using ceph version - 0.80.7
I have setup only one region, "us", with two zones.
"us-west" slave zone having user "us-east"
"us-east" master zone having user "us-east"
The details of specific users are given below.
Details of user for us-east-1 gateway.
{
- "user_id":"us-east",
- "display_name":"Region-US Zone-East",
- "email":"",
- "suspended":0,
- "max_buckets":1000,
- "auid":0,
- "subusers":[
- {
- "id":"us-east:swift",
- "permissions":"full-control"
- {
- "keys":[
- {
- "user":"us-east:swift",
- "access_key":"0DQH33TDOLDPNUOHDGLX",
- "secret_key":""
- {
- "user":"us-east",
- "access_key":"PAA0BEG7ALEEDYXOJ7NE",
- "secret_key":"BBQNeJ9il5lVWU0u897KK3oJRcifQcQdntuqNufu"
- {
- "swift_keys":[
- {
- "user":"us-east:swift",
- "secret_key":"yLbRVIs7QIWcSYLS8KMqzdGWyc3LaKqqvaXJNdF6"
- {
- "caps":[
- "op_mask":"read, write, delete",
- "system":"true",
- "default_placement":"",
- "placement_tags":[
- "bucket_quota":{
- "enabled":false,
- "max_size_kb":-1,
- "max_objects":-1
- "user_quota":{
- "enabled":false,
- "max_size_kb":-1,
- "max_objects":-1
- "temp_url_keys":[
Details of user for us-west-1 gateway
{
- "user_id":"us-west",
- "display_name":"Region-US Zone-West",
- "email":"",
- "suspended":0,
- "max_buckets":1000,
- "auid":0,
- "subusers":[
- {
- "id":"us-west:swift",
- "permissions":"full-control"
- {
- "keys":[
- {
- "user":"us-west:swift",
- "access_key":"ABAI43X3JZ2LE734XC71",
- "secret_key":""
- {
- "user":"us-west",
- "access_key":"98VDZ8ZTWZMFAT1YWXIL",
- "secret_key":"wKQfBqJtYCZ4VSK26JIYN9tad2GC6t9BKyUsHEb3"
- {
- "swift_keys":[
- {
- "user":"us-west:swift",
- "secret_key":"KrjdheLazRpMRzUIpzLgxd0pjN81quFlnp97pwHs"
- {
- "caps":[
- "op_mask":"read, write, delete",
- "system":"true",
- "default_placement":"",
- "placement_tags":[
- "bucket_quota":{
- "enabled":false,
- "max_size_kb":-1,
- "max_objects":-1
- "user_quota":{
- "enabled":false,
- "max_size_kb":-1,
- "max_objects":-1
- "temp_url_keys":[
Now I created a bucket in "us-east" zone with read permissions for all.
vinod@LT05:~$ swift --verbose -A http://us-east-1.lt.com/auth -U us-east:swift
-K yLbRVIs7QIWcSYLS8KMqzdGWyc3LaKqqvaXJNdF6 stat Container1
Account : v1
Container : Container1
Objects : 0
Byte : 0
Read ACL: .r:*
Write ACL :
Sync To :
Sync Key :
Vary : Accept-Encoding
Serve : Apache/2.2.22 (Ubuntu)
X-Container-Bytes-Used-Actual: 0
Content-Type : text/plain; charset=utf-8
There are no containers on the us-west zone.
When i try to create a new container directly in the us-west zone, it returns status-403.
I guess this is because its the slave zone.
But the doc says "You may read objects from secondary zones.
Currently, the Gateway does not prevent you from writing to a secondary zone, but DON’T DO IT."
I am just curious why am I not able to create containers!
Now I sync zones using 'radosgw-agent' , using the command
sudo radosgw-agent --dest-access-key=wKQfBqJtYCZ4VSK26JIYN9tad2GC6t9BKyUsHEb3
--dest-secret-key=wKQfBqJtYCZ4VSK26JIYN9tad2GC6t9BKyUsHEb3
--src-access-key=PAA0BEG7ALEEDYXOJ7NE
--src-secret-key=BBQNeJ9il5lVWU0u897KK3oJRcifQcQdntuqNufu
--source=http://us-east-1.lt.com
--sync-scope=full --log-file=/var/log/radosgw/zone-sync-us-east-west.log
There are no error logged during this process.
But I am not able to see this container on us-west zone.
vinod@LT05:~$ swift --verbose -A http://us-west-1.lt.com/auth -U us-west:swift
-K KrjdheLazRpMRzUIpzLgxd0pjN81quFlnp97pwHs stat
StorageURL: http://us-west-1.lt.com/swift/v1
Auth Token: AUTH_rgwtk0d00000075732d776573743a7377696674080418ee247db6d6c5986c54a00cc1145bcd8fba363322c25ba6508535b5f513c29b3a53
Account : v1
Containers : 0
Objects : 0
Bytes : 0
Vary : Accept-Encoding
Server : Apache/2.2.22 (Ubuntu)
X-Account-Bytes-Used-Actual: 0
Content-Type: text/plain; charset=utf-8
How can I access the container from us-west-1 rgw instance.
Do I need to manually create the us-east user on the us-west-1 instance also?
Now the there is common storage cluster for both the zones. Is it that the replication will work
only when the storage clusters are different?
Vinod H I
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com