hi!
I have a problem again with building lustre on Ubuntu.
After sh autogen.sh and ./configure:
[...]
Making all in lnet
make[5]: Entering directory `/data/src/b2_0/lnet/lnet'
gcc -DHAVE_CONFIG_H -I. -I. -I../.. -D__arch_lib__
-D_LARGEFILE64_SOURCE=1 -include /data/src/b2_0/config.h
Andreas Dilger wrote, On 2009. 11. 28. 0:22:
On 2009-11-27, at 03:13, Papp Tamás wrote:
Craig Prescott wrote, On 2009. 11. 19. 20:42:
We had the same problem with 1.8.x.x.
We set lnet.printk=0 on our OSS nodes and it has helped us
dramatically - we have not seen the 'soft lockup' problem
Craig Prescott wrote, On 2009. 11. 19. 20:42:
We had the same problem with 1.8.x.x.
We set lnet.printk=0 on our OSS nodes and it has helped us
dramatically - we have not seen the 'soft lockup' problem since.
sysctl -w lnet.printk=0
This will turn off all but 'emerg' messages from lnet.
It
Craig Prescott wrote, On 2009. 11. 19. 20:42:
Papp Tamás wrote:
The logs are full with this:
Nov 19 20:03:32 node1 kernel: BUG: soft lockup - CPU#3 stuck for 10s!
[ll_ost_80:4894]
Nov 19 20:03:32 node1 kernel: CPU 3:
snip
Nov 19 20:03:34 node1 kernel: Lustre: Skipped 40339060 previous
liu ning wrote, On 2009. 11. 13. 4:18:
Hi all,
I installed Lustre on a small cluster running CentOS 5.3 and it works
well. But after my installation, the Linux kernel is patched and the
NVIDIA driver can not be used. As a result,the visualization software
can not work properly. So I tried
Papp Tamás wrote, On 2009. 11. 09. 19:34:
Is there any known bug about this?
Is this bug 21221 https://bugzilla.lustre.org/show_bug.cgi?id=21221 ?
10x
tompos
___
Lustre-discuss mailing list
Lustre-discuss@lists.lustre.org
http
Brian J. Murrell wrote, On 2009. 11. 03. 21:57:
No. Because dpkg-buildpackage in the lustre directory doesn't actually
invoke m-a at all. My make debs make target does. That does not
equate to a build dependency for lustre/debian though.
Yes, you're right.
Yes. This looks like a
Brian J. Murrell wrote, On 2009. 11. 02. 21:51:
On Mon, 2009-11-02 at 13:25 -0700, Andreas Dilger wrote:
On 2009-11-02, at 11:41, Brian J. Murrell wrote:
A. In my development tree, I in fact have the following patch:
@@ -4,7 +4,7 @@
Maintainer: Brian J. Murrell
Mag Gam wrote, On 2009. 11. 03. 3:15:
Is it possible to remove a OST permanently on 1.8.x?
Nope, you can only deactivate it.
tamas
___
Lustre-discuss mailing list
Lustre-discuss@lists.lustre.org
Brian J. Murrell wrote, On 2009. 10. 30. 23:07:
On Fri, 2009-10-30 at 16:02 +0100, Papp Tamás wrote:
*Jaunty + b1_8* - failure
Just now, I checked out a b1_8, added the following target to
build/autoMakefile.am.toplevel:
debs:
lversion=$$(sed -ne 's/^#define VERSION
Ralf Utermann wrote, On 2009. 10. 29. 10:03:
check modules/lustre/config.log to see, where it really goes wrong.
Sorry for the late answer.
Actually I almost out of ideas.
Here is, what I did already:
Without deb specific steps. I'm sure, something is missing, or something
I do wrong,
hi All!
I'm trying to make an Ubuntu based patchless client.
I installed Intrepid's kernel 2.6.27 on jaunty, which originally has 2.6.28.
It's booting and working fine.
Also I installed the related headers.
The configure runs OK, but when I compile it:
make[3]: Entering directory
Brian J. Murrell wrote, On 2009. 10. 28. 20:26:
Additionally, b1_8 (aka 1.8.2) also has debian packaging support (on an
unsupported basis at this point) in /debian. Here are the installed
packages, built from the /debian in b1_8 from my client:
ii lustre-client-modules-2.6.28-11-generic
hi All,
I see, there is a page on the wiki, how to build lustre client for
windows. Is there a prebuilt binary available to test it?
Thank you,
tamas
___
Lustre-discuss mailing list
Lustre-discuss@lists.lustre.org
Saber Feki wrote, On 2009. 04. 13. 18:15:
Hi,
I'm currently trying to compile and install the lustre patchless client on
our cluster nodes running OpenSuse 11.0 with the kernel version 2.6.25.
Looking at the mailing list and in the lustre wiki, I was pointed to bug
14250. There, I find a
Ben Evans wrote, On 2009. 03. 25. 14:45:
I remember reading that there is an 8TB limit per OSS in 1.6.6. Is
this correct, and if not what is the limit per OSS?
I think the limit is for OST, if I'm not confused about definitions.
tamas
___
Brian J. Murrell wrote, On 2009. 03. 25. 15:02:
On Wed, 2009-03-25 at 14:41 +0100, Papp Tamas wrote:
hi,
I've tried recently compile B1_8 patchless client support on FC8.
What is the source of your source tree? CVS, something other?
CVS
# make
cd . /bin/sh
Alex Lyashkov wrote, On 2009. 01. 19. 23:42:
On Tue, 2009-01-13 at 18:30 +0100, Papp Tamas wrote:
Is patchless client 2.6.23+ going to be supported by Lustre 1.8?
Patches inspected, but need final testing.
Also, as 1.8.0 rc cycle is started - this huge change will be not
committed
hi!
I see in the changelogs: Recommended e2fsprogs version: 1.40.11-sun1.
But the version 1.40.7 is downloadable (from RH rpms).
Also because 2.6.23+ kernels are still not supported by 1.6.6, could
anyone give us a handy solution like a howto for downloading and
installing the right things.
Deval kulshrestha wrote:
I tried depmod -a but still same error even for modprobe and for original
Error as following
[EMAIL PROTECTED] ~]# !mou
mount -t lustre [EMAIL PROTECTED]:/spfs /mnt/spfs
mount.lustre: mount [EMAIL PROTECTED]:/spfs at /mnt/spfs failed: No such
device
Are the lustre
Deval kulshrestha wrote:
It seems that system does not find luster modules, it gives following error.
[EMAIL PROTECTED] ~]# modprobe lustre
FATAL: Module lustre not found.
RPM's are installed successfully (checked using rpm -qil),
Can you tell me what all rpm do I need to install on
Mag Gam wrote:
I have been getting status -5 messages for a client. What does the
error mean? Running 1.6.5.1
$ grep -hw 5 /usr/include/*/errno*
#define EIO 5 /* I/O error */
tompos
___
Lustre-discuss mailing list
James Braid wrote:
2008/9/30 Papp Tamás [EMAIL PROTECTED]:
What does this mean?
# cat /proc/fs/lustre/mds/storage-MDT/recovery_status
status: INACTIVE
It's normal, it just means recovery is not running (because it's
finished or been aborted or whatever)
This is another
[EMAIL PROTECTED] wrote:
hi all
now , I need to get the clients who are reading or writting data on
one ost. what can i do?
You can set up for a directory which OST should have its datas by lfs
setstripe.
tamas
___
Lustre-discuss mailing list
Papp Tamas wrote:
Andreas Dilger wrote:
On Sep 16, 2008 18:47 +0200, Papp Tam�s wrote:
I run fsck on the node, and remounted it. Fsck found a lot of errors.
After this I see this on the logs again:
Sep 16 18:30:08 node1 kernel: LDISKFS-fs error (device sdb1):
Brian J. Murrell wrote:
No, but you have to understand that we all have tasks that we must
complete in order to meet our goals and that responding to questions
here is done on an as-time-permits basis. Some days there is time and
some there is not. You will just have to bear with us and be
Brian J. Murrell wrote:
On Thu, 2008-09-18 at 20:20 +0200, Papp Tamás wrote:
Actually I tried without feedback:)
A. That's not good. Did you try
http://www.sun.com/software/products/lustre/support.xml?
If I remember well, yes. I think, the problem is that here
Bernd Schubert wrote:
On Tuesday 16 September 2008 18:47:43 Papp Tamás wrote:
hi All,
This morning we see on some client, it cannot connect to one of our node.
I run fsck on the node, and remounted it. Fsck found a lot of errors.
After this I see this on the logs again
Andreas Dilger wrote:
On Sep 17, 2008 09:21 +0200, Papp Tamas wrote:
Andreas Dilger wrote:
On Sep 10, 2008 16:04 +0200, Papp Tam�s wrote:
Cliff White wrote:
We have pre-released versions availble via CVS.
http://wiki.lustre.org/index.php?title=Open_CVS
I don't
Andreas Dilger wrote:
On Sep 12, 2008 11:14 +0200, Papp Tamás wrote:
Andreas Dilger wrote:
You still need to apply patches to allow building Lustre on 2.6.23
kernels. There is a bug open for this in bugzilla.
Where can I get this patches?
In bugzilla... search
Brian J. Murrell wrote:
On Tue, 2008-07-29 at 16:00 +0200, Papp Tamas wrote:
hi,
Is it possible to have 2 meta servers, one of them is x86_64 and the
other one is i386 and of course the data share between the 3 servers
would be made by drbd?
3? I count 2 in your description. In
Kalpak Shah wrote:
Doing a lctl --device OSC_UUID deactivate on the MDS will only stop
new objects from being allocated on that OST. To completely remove the
OST you need to use lctl conf_param lustre-OST.osc.active=0. If
What's the difference between lctl deactivate and 'completely
Papp Tamás wrote:
Bernd Schubert wrote:
Hello Tamás,
On Tuesday 17 June 2008 16:41:55 Papp Tamás wrote:
Dear All,
Is there any reason to not user kernels with version 2.6.22.x above
2.6.22.14 or should it work?
I've just compiled it with 2.6.22.19 and I can mount
Dear All,
I have some more questions.
If I make and lfs df or pure (linux) df, I see some free spaces.
But I know, a node is deactive from write. Why does it show its free
space with the others. I think, it's not a good thing.
Otherwise, is it possible to check from command line which node is
Dear All,
Our small but hardly used cluster was running for weeks without any
problems, but yesterday 3 of 8 nodes froze up in 1 hour.
This is from messages log, there was many of this in it before this last
one:
May 17 22:00:31 node8 kernel: LustreError: dumping log to
Dear All,
What is the expected date when v1.6.5 arrives?
What kernels (patchless clients) will be supported?
Thank you,
tamas
___
Lustre-discuss mailing list
Lustre-discuss@lists.lustre.org
http://lists.lustre.org/mailman/listinfo/lustre-discuss
hi all,
I have still samba problems.
After a while I see D processes, and all samba share get stucked.
In the syslog I see:
Apr 22 15:02:39 samba kernel: Call Trace:
Apr 22 15:02:39 samba kernel: IRQ [8020e007]
__alloc_pages+0x2b8/0x2d1
Apr 22 15:02:39 samba kernel:
Dear All,
Apr 17 16:59:10 node1 kernel: Lustre:
7833:0:(router.c:167:lnet_notify()) Ignoring prediction from
[EMAIL PROTECTED] of [EMAIL PROTECTED] down 542730734650 seconds in the future
What could cause this error message?
I don't find anything really useful searching the web.
My main
David Bogen wrote:
All:
In an effort to test the patches against lustre that might improve NFS
performance I've been attempting to build from source on a RHEL5 system.
I got the lustre (1.6.4.3) and kernel (linux-2.6.18-53.1.13) source
from Sun (that's now a really, um, less-than-ideal
Dear All,
We have a really serious problem.
This weekend we upgraded our mixed 1.6.0.1 + 1.6.4.1 cluster based on
CentOS 4.4 i386 to CentOS 5.1 x86_64.
We reformatted all drbd storages, so recreated the whole cluster and
copy back from backup.
Windows clients reached the cluster through a
Dear All,
What could cause this error?
Kernel: 2.6.9-42.0.10.EL_lustre-1.6.0.1custom-drbd and
2.6.9-55.0.9.EL_lustre.1.6.4.1smp (CentOS 4.4)
After the node freezed up, his failover pair took over the resource, but
it did it too.
I've just looked back in logs and I see, this header corrupted
Dear All,
How can I find about a file, which OST is it on?
I cannot find it in the manual.
Thanks,
tamas
___
Lustre-discuss mailing list
Lustre-discuss@lists.lustre.org
http://lists.lustre.org/mailman/listinfo/lustre-discuss
42 matches
Mail list logo