Survey: Do you deploy MemcacheDB with replication?
4 views
Skip to first unread message
Steve Chu
Oct 9, 2008, 3:49:46 AM10/9/08
to memca...@googlegroups.com
Hi, all,
As a critical feature, replication is used to gain HA for your storage
in MemcacheDB. In order to improve replication, and make it more
reliable, Please let me know if you are using replication feature in
your production environment.
Please follow this mail if you do, and more detail is better. Also any
advice on replication is welcome and appreciated. Besides, I can tell
you that we have deployed MemcacheDB a lot with replication, but for a
customize version. 1.1.0 now has some bugs in replication, I'v already
fixed in the near coming of 1.2.0 version. 1.2.0 will be merged into
trunk these days, and after some testing, we will release it.
Thanks,
--
Steve Chu
http://stvchu.org
Shaojie Liu
Oct 9, 2008, 6:34:52 AM10/9/08
to memca...@googlegroups.com
we r using the Replication feature actually, in version 1.0.3.
we setup one master box and one slave box, taking the responsibilty of write and read respectively. would like to provide more details if necessary, let me know.
and one question if dont mind, i guess memcachedb had already utilized the feature of master election in same BDB group? we have not achieve automatical HA yet, cause it leads application to be a bit complicated by always querying who is the current master.
we setup one master box and one slave box, taking the responsibilty of write and read respectively. would like to provide more details if necessary, let me know.
and one question if dont mind, i guess memcachedb had already utilized the feature of master election in same BDB group? we have not achieve automatical HA yet, cause it leads application to be a bit complicated by always querying who is the current master.
Shaojie Liu
Oct 9, 2008, 6:34:52 AM10/9/08
to memca...@googlegroups.com
we r using the Replication feature actually, in version 1.0.3.
we setup one master box and one slave box, taking the responsibilty of write and read respectively. would like to provide more details if necessary, let me know.
and one question if dont mind, i guess memcachedb had already utilized the feature of master election in same BDB group? we have not achieve automatical HA yet, cause it leads application to be a bit complicated by always querying who is the current master.
we setup one master box and one slave box, taking the responsibilty of write and read respectively. would like to provide more details if necessary, let me know.
and one question if dont mind, i guess memcachedb had already utilized the feature of master election in same BDB group? we have not achieve automatical HA yet, cause it leads application to be a bit complicated by always querying who is the current master.
Reply all
Reply to author
Forward
0 new messages