Hi guys
I have test memory and cpu hotplug in a CentOS 7 VM.
Memory works fine, but I only can increase the total among of memory but
not decrease.
CPU, although I had enable NUMA, I was unable to change the number of cores.
Perhaps I misunderstood the behavior of hotplug features...
If this the
I have test memory and cpu hotplug in a CentOS 7 VM.
Memory works fine, but I only can increase the total among of memory but not
decrease.
Yes, hotplug only for now. unplug is coming in next qemu release
CPU, although I had enable NUMA, I was unable to change the number of cores.
Perhaps I
and I notice some visual refresh glitch on hardware view.
Seem that default values are display (memory 512mb,...), and half second later
the pending api refresh with corrects values.
(All is working fine, but it's just visually annoying)
Sorry, I am unable to see that.
Please can you
On Fri, 20 Feb 2015 16:21:24 -0500
Eric Blevins ericlb...@gmail.com wrote:
I can confirm this, IPOIB multicast seems to only work intermittently
with pve-kernel-2.6.32-37-pve
According to this thread RHEL 6.6 kernel has introduced a regression
into IPoIB which brakes multicast. The thread
On Fri, 20 Feb 2015 00:29:36 +0100
Michael Rasmussen m...@datanom.net wrote:
I am giving up for now and have reverted back to
pve-kernel-2.6.32-34-pve.
Doing the Same
I forgot to mention that my conclusion is that the infiniband drivers
delivered with pve-kernel-2.6.32-37-pve is broken.
Some questions:
1) if I have an HA cluster, does ZFS get me anything? Or is it really
only useful for standalone systems using local storage?
only useful for local storage
2) to get root-on-zfs, I assume I must reinstall instead of upgrading?
yes
3) since I use CEPH and only have 3 hard
Oh, I forgot the most important part!
It is much better to create pools referencing to disks by id, instead of their
names: it is hard to recover in any situation or just move from one server to
an other.
Like this:
zpool create -f -o ashift=12 rpool mirror
On Fri, 20 Feb 2015 23:35:09 +0100
Michael Rasmussen m...@datanom.net wrote:
According to this thread RHEL 6.6 kernel has introduced a regression
into IPoIB which brakes multicast. The thread referrers to 3 patches
which should solve the issue.
According to this thread RHEL 6.6 kernel has introduced a regression
into IPoIB which brakes multicast. The thread referrers to 3 patches
which should solve the issue.
https://www.mail-archive.com/linux-rdma@vger.kernel.org/msg22511.html
The complete patchset can be found here:
- it use gzip-6 compression by default - as the system supports lz4
compression, I recommend to use this one instead of gzip-X. In general. Of
course, it is possible to change the compression on the fly (for newly written
blocks) but for general usage, lz4 needs less power and provides a
10 matches
Mail list logo