Hello Ray,
One consideration first : You try the 2.7 version which is not the
production one (aka 2.5). From this perspective wether you run 2.7.0 or
2.7.x won't make any big difference, it is the develpment release.
Then if I understand the problem comes from the infiniband driver module
wh
Hi,
Where do you find the 2.7.x-releases? I thought fixes were only released
for the Intel maintenance version?
Regards,
/jon
On 12/04/2015 11:43 AM, jerome.be...@inserm.fr wrote:
Hello Ray,
One consideration first : You try the 2.7 version which is not the
production one (aka 2.5). From
Hi,
I honestly don't know if the compiled versions available here are meant
to be used by everyone but they are publicly browsable on Intel Jenkins
:
https://build.hpdd.intel.com
but as the source is publicly available from the whamcloud git, there
imo might not be any problem
If you are
>
> Here are 4 questions which we cannot find answers in LU-1586:
> 1. According to Andres?s reply, there should some unconsumed
> changelog files on our MDT, and these files have taken all the space (file
> quotas?) Lustre gives to changelog. With Lustre 2.1, these files are under
> OBJECTS
As I mentioned, I am doing a test install to see what I want to run for
deployment. We have run a couple Lustre installs, one 1.8.x based and a
current production one that is 2.3. The Lustre 2.3 server set has been
up for 750 days and has been very solid. This test replaces the old 1.8
setup
Ok,
I am not using IB here but it looks obvious that the max_frag value
differs between the MGS and the client.
Do you use the same lustre version on the MGS/OSS AND the client built
on the same Kernel version ? (ie lustre*-KERNEL_VERSION-LUSTRE_VERSION)
Did you try it with the latest night
Client was rebuilt locally from the source RPM's. I thought I had built
it from the client source from the nightly build but I can see now it
was the 2.7.0 source
lustre-client-2.7.0-2.6.32_504.8.1.el6.x86_64.src.rpm
Client kernel is the OS provided kernel.
At this point I have ripped out al
> On Dec 4, 2015, at 9:48 AM, Ray Muno wrote:
>
> I started with 2.7 since that is what I got pointed to when I went to the
> lustre.org download page. The "Most Recent Release" points me at the 2.7.0
> tree. If I follow the path to download source on that page,
Lustre 2.7 is the most recen
Thanks Rick for this useful summary !
What is not clear to me if maintenance releases will eventually be
available again or if people running production systems will have to
choose among:
1) Buy Intel support contract
2) Move to another filesystem
Thanks, Massimo
On 04/12/2015 17:53, Mohr J
Don't forget option 1.5) Get support from your hardware vendor (DDN, Seagate,
Cray, Terrascala, Dell, etc)
Ed Wahl
OSC
From: lustre-discuss [lustre-discuss-boun...@lists.lustre.org] on behalf of
Massimo Sgaravatto [massimo.sgarava...@pd.infn.it]
Sent:
> On Dec 4, 2015, at 12:41 PM, Massimo Sgaravatto
> wrote:
>
> Thanks Rick for this useful summary !
>
> What is not clear to me if maintenance releases will eventually be available
> again or if people running production systems will have to choose among:
>
> 1) Buy Intel support contract
>
Hi Ray,
I'll throw my 0.02 into the ring:
There are known issues with the Truescale kernel driver in
redhat/centos/scilinux 6.6. You should try kernel 2.6.32-504.23.4 or
newer. Some details of the bug are in LU-6698 and RHSA-2015-1081.
Further, lustre 2.7+ now applies performance tuning para
Ok, but let's take for example Lustre 2.7
RPMs for Lustre 2.7 published on the web are ~ 9 months ago.
I guess that in the meantime several problems were fixed.
And these fixes are available only to Intel customers (I guess there are
some newer rpms available to them) and will never be available
13 matches
Mail list logo
