d module of lustre it auto remove
> entry from grub.conf & doesn't boot with lustre-kernel image.
>
>
>
> If anyone have setup lustre Server on Rocks 6.2 compute node then please
> share setup that I'm missing.
>
>
>
> Than
The only requirement for upgrade ordering in the past was during the 1.8->2.x
upgrade. In that case, 1.8 clients could connect to both 1.8 and 2.x servers,
while 2.x clients can only connect to 2.x servers. Since then there isn't a
hard requirement for upgrade ordering, though we only test a f
I checked the output of dmesg and found
lnet: Unknown parameter `#’
The error was caused due to a comment that was inserted in
/etc/modprobe.d/lustre.conf
Thanks a lot for your time and your help.
> On Jun 22, 2016, at 4:03 PM, Patrick Farrell wrote:
>
> OK, that's not the error we started wi
OK, that's not the error we started with. You were having a problem
loading modules, but aren't now. (probably)
The error output from that command is bad. It asks if the modules are
loaded, but that doesn't mean they aren't.
"No such device" is probably the key here. I'd look at the argume
The error is in the last step when I have to run MD
3. mount -t lustre /dev/vg00/mdt1 /mdt
mount.lustre: mount /dev/mapper/vg00-mdt1 at /mdt failed: No such device
Are the lustre modules loaded?
Check /etc/modprobe.conf and /proc/filesystems
> On Jun 22, 2016, at 3:58 PM, Patrick Farrell wrote:
Ah, now that I look more closely - that warning may not be fatal. Did
you get a failure, or just the warning?
(The guide looks good)
On 06/22/2016 02:52 PM, Sangeetha Banavathi Srinivasa wrote:
Patrick,
I am following the steps on
http://lustre.ornl.gov/lustre101-courses/content/C1/L3/LustreB
Patrick,
I am following the steps on
http://lustre.ornl.gov/lustre101-courses/content/C1/L3/LustreBasicInstall.pdf
Except that I am installing 2.8 instead of 2.7.
Could you recommend any other article/blog that I can follow in order to
successfully install lustre.
> On Jun 22, 2016, at 3:43 PM
Hm. It sounds like your RPMs are not installed correctly, but I vaguely
recall something about a packaging issue with the Lustre 2.8 release RPMs.
Perhaps someone else on the list can help further.
- Patrick
On 06/22/2016 02:40 PM, Sangeetha Banavathi Srinivasa wrote:
I have done that already
I have done that already and re-booted my system.
But I still get the error.
When I did run that command, I got the below error:
WARNING: Can't read module
/lib/modules/2.6.32-573.12.1.el6_lustre.x86_64/weak-updates/kernel/fs/lustre/llog_test.ko:
No such file or directory
egrep:
/lib/modules/2.
Well, that's not patching the kernel. It's making the initrd and and
updating the boot stuff so your system will boot that kernel.
But, yes, that should be what you need to do.
- Patrick
On 06/22/2016 02:36 PM, Sangeetha Banavathi Srinivasa wrote:
When you say install the associated kernel do
When you say install the associated kernel do you mean patching the kernel like
below
sudo /sbin/new-kernel-pkg --package kernel --mkinitrd --dracut --depmod
--install 2.6.32-573.12.1.el6_lustre.x86_64
I have already done rpm -ivh kernel-2.6.32-573.12.1.el6_lustre.x86_64.rpm
> On Jun 22, 2016
Servers upgraded first or with clients. Alex.
On Jun 22, 2016, at 11:01 AM, E.S. Rosenberg
mailto:esr+lus...@mail.hebrew.edu>> wrote:
I always understood the recommendation was to update the clients (and LNET
Routers) before the servers and not the other way around?
___
LLNL has always in the past upgraded the servers first.
Chris
On 06/22/2016 09:01 AM, E.S. Rosenberg wrote:
> I always understood the recommendation was to update the clients (and
> LNET Routers) before the servers and not the other way around?
>
>
> On Tue, Jun 21, 2016 at 1:16 AM, Mohr Jr, Ri
The kernel version is found with uname -r or cat /proc/version.
That needs to match the kernel version in the Lustre RPMs. (The version
string is the 2.6.32-573.12.1.el6_lustre) part.
In fact, it looks like you've got Lustre server RPMs, given the _lustre
in the name of the kernel. That mea
Patrick,
The CentOS version on the machines is 6.8
[lustre@hulk2 ~]$ cat /etc/redhat-release
CentOS release 6.8 (Final)
The kernel version that I have installed is what I found along with the lustre
RPMs
[lustre@amaranth3 ~]$ uname -r
2.6.32-573.12.1.el6_lustre.x86_64
And I
Well, it says to check dmesg... Though that will just list the
symbols. In general, this usually means you've got a mismatch between
the version of the kernel (or OFED) Lustre is built against and your
version of either the kernel or OFED. Given the number and locations of
the mismatches, I'
When I try modprobe lustre, I see the below errors:
[05:44:03] root@hulk2 lustre # modprobe lustre
WARNING: Error inserting obdclass
(/lib/modules/2.6.32-573.12.1.el6_lustre.x86_64/extra/kernel/fs/lustre/obdclass.ko):
Unknown symbol in module, or unknown parameter (see dmesg)
WARNING: Error inser
Hi,
Below are the steps I executed in configuring MDS.
1. mkfs.lustre --fsname lustre --mdt --mgs /dev/vg00/mdt
2. mkdir /mdt
3. mount -t lustre /dev/vg00/mdt /mdt
I see the below error after step 3:
mount.lustre: mount /dev/mapper/vg00-mdt1 at /mdt failed: No such device
Are the lustre module
I always understood the recommendation was to update the clients (and LNET
Routers) before the servers and not the other way around?
On Tue, Jun 21, 2016 at 1:16 AM, Mohr Jr, Richard Frank (Rick Mohr) <
rm...@utk.edu> wrote:
>
> > On Jun 20, 2016, at 5:00 PM, Jessica Otey wrote:
> >
> > All,
>
Hi Yasir.
You must add the lustre packages to the distribution following the rocks
manual:
http://central6.rocksclusters.org/roll-documentation/base/6.2/customization-adding-packages.html
Finally you must add the lustre network and configure the compute
network interfaces that you will use f
Hi
Can anyone help for installing Lustre Server on Rocks 6.2 Compute Node.
I've installed all rpm incl. kernel, lustre server pkg, lustre-module,
lustre-osd-module, lustre-osd-mount, libyaml pkg. After installing kernel it
shows in /boot directory and lustre kernel entry in grub.conf of compute
21 matches
Mail list logo
