[Bugs] [Bug 1058300] VMs do not resume after paused state and storage connection to a gluster domain (they will also fail to be manually resumed)

bugzilla at redhat.com bugzilla at redhat.com
Sun Oct 11 07:42:12 UTC 2015


https://bugzilla.redhat.com/show_bug.cgi?id=1058300



--- Comment #50 from Christopher Pereira <kripper at imatronix.com> ---
Hi John,

We dropped gluster replication after finding out that writes are synchronic
(i.e. performance will be limited to storage network bandwith).
We are waiting for async write support in stable gluster branches.

A 1 Gbps network would restrict disk writes to 125 MB/s.
Having multiple VM's means sharing those slow resources.
Besides, gluster doesn't suport SEEK_HOLE reads, giving very bad performance
with raw sparse files.
There were some patches to implement SEEK_HOLE support, but they probably won't
be merged in the short term:
https://bugzilla.redhat.com/show_bug.cgi?id=1220173

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.


More information about the Bugs mailing list