Re: iSCSI Multipath (Load Balancing) vs RBD Exclusive Lock

["shadow_lin"<shadow_lin@xxxxxxx>]

Hi David,

Thanks for the info.

Could I assume that if use active/passive multipath with rbd exclusive lock  then all targets which support rbd(via block) are safe?

2018-03-08

---

shadow_lin

---

发件人：David Disseldorp <ddiss@xxxxxxx>

发送时间：2018-03-08 08:47

主题：Re: [ceph-users] iSCSI Multipath (Load Balancing) vs RBD Exclusive Lock

收件人："shadow_lin"<shadow_lin@xxxxxxx>

抄送："Mike Christie"<mchristi@xxxxxxxxxx>,"Lazuardi Nasution"<mrxlazuardin@xxxxxxxxx>,"Ceph Users"<ceph-users@xxxxxxxxxxxxxx>

 

Hi shadowlin, 

 

On Wed, 7 Mar 2018 23:24:42 +0800, shadow_lin wrote: 

 

> Is it safe to use active/active multipath If use suse kernel with target_core_rbd? 

> Thanks. 

 

A cross-gateway failover race-condition similar to what Mike described 

is currently possible with active/active target_core_rbd. It's a corner 

case that is dependent on a client assuming that unacknowledged I/O has 

been implicitly terminated and can be resumed via an alternate path, 

while the original gateway at the same time issues the original request 

such that it reaches the Ceph cluster after differing I/O to the same 

region via the alternate path. 

It's not something that we've observed in the wild, but is nevertheless 

a bug that is being worked on, with a resolution that should also be 

usable for active/active tcmu-runner. 

 

Cheers, David 

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com