For Giant, we have changed the default librbd caching options to:
rbd cache = true
rbd cache writethrough until flush = true
The second option enables the cache for reads but does writethrough until
we observe a FLUSH command come through, which implies that the guest OS
is issuing barriers.
Hello,
I have a Cluster with 30 OSDs distributed over 3 Storage Servers connected by a
10G cluster link and connected to the Monitor over 1G. I still have a lot to
understand with Ceph. Observing the cluster messages in a "ceph -watch" window
I see a lot of osd "flapping" when it is sitting in a
Based on the ceph watch output my guess is the osd_heartbeat_grace default of
20 is causing my reporting issues. I've seen failures, all of which recover,
from reporting after 22 to ~28 second. I was unable to set osd_heartbeat_grace
using the runtime command - every syntax I tried the cmd faile
this is happen i use *ceph-deploy create ceph01-vm ceph02-vm ceph04-vm *to
create 3 Mons member.
now every 10 hours, one Mon will down. every time have this error, some
time the hardisk have enough space left,such as 30G.
i deployed Ceph before, only create one Mon at first step *ceph-depl
It's really strange! I write a test program according the key ordering
you provided and parse the corresponding value. It's true!
I have no idea now. If free, could you add this debug code to
"src/os/GenericObjectMap.cc" and insert *before* "assert(start <=
header.oid);":
dout(0) << "start: "
I see osd's being failed for heartbeat reporting > default osd_heartbeat_grace
of 20 but the run time config shows that the grace is set to 30. Is there
another variable for the osd or the mon I need to set for the non default
osd_heartbeat_grace of 30 to take effect?
2014-08-23 23:03:08.982590
