LUG 2015: View the Agenda & Register
by OpenSFS Administration
<http://opensfs.org/events/lug-2015/>
Register for LUG 2015 & Save with the Early Bird Discount
Full Agenda Now Available
The <http://opensfs.org/events/lug-2015/> 13th annual LustreR User Group
(LUG) conference is just a month away, and we hope to see you there! This
year's event is structured to allow ample time for discussion and
collaboration, while maintaining a strong list of topics.
<http://opensfs.org/events/lug-2015/> Event Overview
LUG 2015 will be held on April 13-15 at the
<http://www.marriott.com/hotels/travel/dendt-denver-marriott-city-center/>
Denver Marriott City Center in Denver, CO. The event will feature 20+
sessions over two and a half days on topics including:
* Deep technical presentations from DataDirect Networks, Intel,
Lawrence Livermore National Laboratory (LLNL), Los Alamos National
Laboratory (LANL), Seagate Technology, NASA, UW Space Science and
Engineering Center, Tokyo Institute of Technology (Japan), and more!
* Wide range of technical poster sessions for easy access at any
point during the event.
<http://opensfs.org/lug-2015-agenda/> >> View the Full LUG 2015 Agenda <<
<http://opensfs.org/events/lustre-developer-day/> Lustre Developer Day
OpenSFS is pleased to host a Lustre Developer Day in Denver, Colorado on
April 16, 2015. The LUG Developer Day is a new event, sponsored by OpenSFS,
to help provide Lustre developers and those interested in doing Lustre
development a chance to meet face to face. This meeting will give newer
developers a chance to listen, learn, and network while giving veteran
developers a chance to present their ideas and share current work.
Visit the Lustre Developer Day website
<http://opensfs.org/events/lustre-developer-day/> to register and view
additional event information.
<http://opensfs.org/events/lug-2015/> Early Bird Registration Ends March
20th!
Act now to take advantage of the special early bird registration discount
for LUG 2015. If you register by March 20th, you will receive a $50 discount
on registration. After March 20th, the LUG registration rate will be $575.
<http://events.r20.constantcontact.com/register/event?oeidk=a07eaip0b0e02ceb
1b5&llr=8irdhctab> >> Register Now <<
<http://www.marriott.com/hotels/travel/dendt-denver-marriott-city-center/>
Book Now and Save
The venue for LUG 2015 is the
<http://www.marriott.com/hotels/travel/dendt-denver-marriott-city-center/>
Denver Marriott City Center located in the heart of downtown Denver. Special
hotel rates are available until March 30th, based on availability.
* General Attendee Room Rate: $169 USD for single occupancy
* Government Attendee Room Rate: $163 USD for single occupancy
If you have any questions, please feel free to contact
<mailto:admin@opensfs.org> admin(a)opensfs.org.
Best regards,
OpenSFS LUG Planning Committee
_________________________
OpenSFS Administration
3855 SW 153rd Drive Beaverton, OR 97003 USA
Phone: +1 503-619-0561 | Fax: +1 503-644-6708
Twitter: <https://twitter.com/opensfs> @OpenSFS
Email: <mailto:admin@opensfs.org> admin(a)opensfs.org | Website:
<http://www.opensfs.org> www.opensfs.org
<http://cdn.opensfs.org/wp-content/uploads/2015/01/LUG_2015_Sponsorship_Oppo
rtunities.pdf> Click here to learn how to become a 2015 LUG Sponsor
Member Benefits at LUG
OpenSFS is pleased to offer one complimentary LUG event pass to each member
company of OpenSFS as a benefit of membership. The discount code has been
sent to the OpenSFS primary contact, but it may be transferred to another
representative at the same company.
Email <mailto:admin@opensfs.com> admin(a)opensfs.org for more information
about using your company pass.
Open Scalable File Systems, Inc. was founded in 2010 to advance Lustre
development, ensuring it remains vendor-neutral, open, and free. Since its
inception, OpenSFS has been responsible for advancing the Lustre file system
and delivering new releases on behalf of the open source community. Through
working groups, events, and ongoing funding initiatives, OpenSFS harnesses
the power of collaborative development to fuel innovation and growth of the
Lustre file system worldwide.
7 years, 2 months
Re: [HPDD-discuss] missing symbol errors when installing lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
by Cory Spitz
It is critical, unless you use ZFS as a backend instead of ldiskfs. In
that case, you don't need it at all.
-Cory
On 3/11/15, 2:46 PM, "Patrick Farrell" <paf(a)cray.com> wrote:
>Ldiskfs is the default on disk file system for Lustre, used to format the
>disk volumes and actually store the data, on the MDT, OSTs, and MGT.
>It's EXT4 derived.
>
>It's possible to mount individual Lustre targets as ldiskfs, and is
>occasionally necessary for troubleshooting. (So, it's a critical server
>component.)
>________________________________________
>From: HPDD-discuss [hpdd-discuss-bounces(a)lists.01.org] on behalf of Kurt
>Strosahl [strosahl(a)jlab.org]
>Sent: Wednesday, March 11, 2015 2:07 PM
>To: hpdd-discuss(a)ml01.01.org
>Subject: Re: [HPDD-discuss] missing symbol errors when installing
>lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
>
>Ahh,
>
> modprobe does work for those, so I suppose it is ok. So what role
>does that package play in the lustre file system? If it isn't necessary
>for the mds/mdt or the ost systems then I'll pull it.
>
>w/r,
>Kurt
>
>
>------------------------------
>
>Message: 2
>Date: Tue, 10 Mar 2015 15:58:48 -0500
>From: Patrick Farrell <paf(a)cray.com>
>To: <hpdd-discuss(a)lists.01.org>
>Subject: Re: [HPDD-discuss] missing symbol errors when installing
> lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
>Message-ID: <54FF5B08.6050004(a)cray.com>
>Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
>
>Kurt,
>
>The ll_* errors, at least, are normal. The install order is such that
>some symbols are undefined at install time for particular KOs, but are
>defined in KOs that are installed later.
>
>I am not familiar with the ldiskfs unknown symbol errors, but I haven't
>installed the exact RPMs you're installing either.
>If you can modprobe ldiskfs and fsfilt-ldiskfs after completing the
>install, then these are probably the same as the ll_* - Undefined at
>install time, but defined by later KOs.
>
>- Patrick Farrell
>_______________________________________________
>HPDD-discuss mailing list
>HPDD-discuss(a)lists.01.org
>https://lists.01.org/mailman/listinfo/hpdd-discuss
>_______________________________________________
>HPDD-discuss mailing list
>HPDD-discuss(a)lists.01.org
>https://lists.01.org/mailman/listinfo/hpdd-discuss
7 years, 2 months
Re: [HPDD-discuss] missing symbol errors when installing lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
by Kurt Strosahl
Right, That is what I had assumed...
I'm just wondering why it is generating such errors from an install of a RPM I downloaded directly from Intel.
w/r,
Kurt
----- Original Message -----
From: "Patrick Farrell" <paf(a)cray.com>
To: "Kurt Strosahl" <strosahl(a)jlab.org>, hpdd-discuss(a)ml01.01.org
Sent: Wednesday, March 11, 2015 3:46:06 PM
Subject: RE: missing symbol errors when installing lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
Ldiskfs is the default on disk file system for Lustre, used to format the disk volumes and actually store the data, on the MDT, OSTs, and MGT. It's EXT4 derived.
It's possible to mount individual Lustre targets as ldiskfs, and is occasionally necessary for troubleshooting. (So, it's a critical server component.)
________________________________________
From: HPDD-discuss [hpdd-discuss-bounces(a)lists.01.org] on behalf of Kurt Strosahl [strosahl(a)jlab.org]
Sent: Wednesday, March 11, 2015 2:07 PM
To: hpdd-discuss(a)ml01.01.org
Subject: Re: [HPDD-discuss] missing symbol errors when installing lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
Ahh,
modprobe does work for those, so I suppose it is ok. So what role does that package play in the lustre file system? If it isn't necessary for the mds/mdt or the ost systems then I'll pull it.
w/r,
Kurt
------------------------------
Message: 2
Date: Tue, 10 Mar 2015 15:58:48 -0500
From: Patrick Farrell <paf(a)cray.com>
To: <hpdd-discuss(a)lists.01.org>
Subject: Re: [HPDD-discuss] missing symbol errors when installing
lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
Message-ID: <54FF5B08.6050004(a)cray.com>
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Kurt,
The ll_* errors, at least, are normal. The install order is such that
some symbols are undefined at install time for particular KOs, but are
defined in KOs that are installed later.
I am not familiar with the ldiskfs unknown symbol errors, but I haven't
installed the exact RPMs you're installing either.
If you can modprobe ldiskfs and fsfilt-ldiskfs after completing the
install, then these are probably the same as the ll_* - Undefined at
install time, but defined by later KOs.
- Patrick Farrell
_______________________________________________
HPDD-discuss mailing list
HPDD-discuss(a)lists.01.org
https://lists.01.org/mailman/listinfo/hpdd-discuss
7 years, 2 months
Re: [HPDD-discuss] missing symbol errors when installing lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
by Kurt Strosahl
Ahh,
modprobe does work for those, so I suppose it is ok. So what role does that package play in the lustre file system? If it isn't necessary for the mds/mdt or the ost systems then I'll pull it.
w/r,
Kurt
------------------------------
Message: 2
Date: Tue, 10 Mar 2015 15:58:48 -0500
From: Patrick Farrell <paf(a)cray.com>
To: <hpdd-discuss(a)lists.01.org>
Subject: Re: [HPDD-discuss] missing symbol errors when installing
lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
Message-ID: <54FF5B08.6050004(a)cray.com>
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Kurt,
The ll_* errors, at least, are normal. The install order is such that
some symbols are undefined at install time for particular KOs, but are
defined in KOs that are installed later.
I am not familiar with the ldiskfs unknown symbol errors, but I haven't
installed the exact RPMs you're installing either.
If you can modprobe ldiskfs and fsfilt-ldiskfs after completing the
install, then these are probably the same as the ll_* - Undefined at
install time, but defined by later KOs.
- Patrick Farrell
7 years, 2 months
missing symbol errors when installing lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64
by Kurt Strosahl
Hello,
I'm getting the below warnings when installing the rpm lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64.rpm, which I downloaded from Intel.
Installing : lustre-osd-ldiskfs-.5.3-2.6.32_431.23.3.el6_lustre.x86_64.x86_64 1/1
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_free_blocks
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_journal_start_sb
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol __ldiskfs_journal_stop
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_ext_search_right
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_discard_preallocations
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_ext_calc_credits_for_insert
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_ext_insert_extent
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_mb_new_blocks
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_ext_search_left
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_ext_walk_space
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/fsfilt_ldiskfs.ko needs unknown symbol ldiskfs_map_inode_page
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/llite_lloop.ko needs unknown symbol ll_iocontrol_register
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/llite_lloop.ko needs unknown symbol ll_direct_rw_pages
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/llite_lloop.ko needs unknown symbol ll_stats_ops_tally
WARNING: /lib/modules/2.6.32-431.el6.x86_64/weak-updates/kernel/fs/lustre/llite_lloop.ko needs unknown symbol ll_iocontrol_unregister
I'm installing it with the other rpms from the el6/servers/RPMS/x86_64 directory.
w/r,
Kurt J. Strosahl
System Administrator
Scientific Computing Group, Thomas Jefferson National Accelerator Facility
7 years, 2 months
An error and some questions
by Kurt Strosahl
Good afternoon,
I'm in the process of standing up a new lustre 2.5.3 file system, and have been getting a strange error when I mount the mds/mdt.
I'm using CentOS 6.5, with the stock infiniband, lustre 2.5.3, and a logical volume (to allow for snapshots and backups). The module for lnet contains the following "options lnet networks=tcp0(em1),o2ib0(bond0)"
Though the mds/mdt seems to mount I see the following error...
~] mount -t lustre /dev/lustrepool/lustrevol /mdt
~] dmesg
LDISKFS-fs (dm-1): mounted filesystem with ordered data mode. quota=on. Opts:
Lustre: lustre-MDT0000: used disk, loading
Lustre: MGS: non-config logname received: params
Lustre: Skipped 1 previous similar message
LustreError: 11-0: lustre-MDT0000-lwp-MDT0000: Communicating with 0@lo, operation mds_connect failed with -11.
~]lctl dl
0 UP osd-ldiskfs lustre-MDT0000-osd lustre-MDT0000-osd_UUID 8
1 UP mgs MGS MGS 5
2 UP mgc MGC<proper ip for ib interface>@o2ib c2f57f70-a066-ea38-0d9b-e54f0f52ae81 5
3 UP mds MDS MDS_uuid 3
4 UP lod lustre-MDT0000-mdtlov lustre-MDT0000-mdtlov_UUID 4
5 UP mdt lustre-MDT0000 lustre-MDT0000_UUID 5
6 UP mdd lustre-MDD0000 lustre-MDD0000_UUID 4
7 UP qmt lustre-QMT0000 lustre-QMT0000_UUID 4
8 UP lwp lustre-MDT0000-lwp-MDT0000 lustre-MDT0000-lwp-MDT0000_UUID 5
Thank you,
Kurt J. Strosahl
System Administrator
Scientific Computing Group, Thomas Jefferson National Accelerator Facility
7 years, 2 months
[PATCH 0/4] staging/lustre: checkpatch cleanup
by Mario J. Rugiero
This patches cleans the following checkpatch issues:
trailing semicolons in macros
__attribute__(format(printf,...)) instead of __printf(...)
__attribute__(aligned(size)) instead of __aligned(size)
__attribute__(packed) instead of __packed
Mario J. Rugiero (4):
staging/lustre: checkpatch cleanup: macros should not use a trailing
semicolon
staging/lustre: checkpatch cleanup: __printf(string-index,
first-to-check) is preferred over __attribute__((format(printf,
string-index, first-to-check)))
staging/lustre: checkpatch cleanup: __aligned(size) is preferred over
__attribute__((aligned(size)))
staging/lustre: checkpatch cleanup: __packed is preferred over
__attribute__((packed))
.../lustre/include/linux/libcfs/libcfs_debug.h | 6 +--
.../include/linux/libcfs/libcfs_kernelcomm.h | 4 +-
.../lustre/include/linux/libcfs/libcfs_private.h | 4 +-
.../staging/lustre/include/linux/lnet/lib-types.h | 2 +-
.../staging/lustre/lustre/include/lprocfs_status.h | 6 +--
drivers/staging/lustre/lustre/include/lu_object.h | 2 +-
.../lustre/lustre/include/lustre/lustre_idl.h | 44 +++++++++++-----------
.../lustre/lustre/include/lustre/lustre_user.h | 14 +++----
drivers/staging/lustre/lustre/include/lustre_dlm.h | 2 +-
drivers/staging/lustre/lustre/include/lustre_net.h | 2 +-
10 files changed, 43 insertions(+), 43 deletions(-)
--
2.3.2
7 years, 2 months
[PATCH 0/4] staging/lustre: checkpatch cleanup
by Mario J. Rugiero
As Dan Carpenter (thank you!) noticed, my original patchset had an extra
from field and lacked the signed-off-by. I'm not sure the first was
fixed, but I would be really confused if it isn't. I checked that the
signed-off-by was generated this time.
As a reminder, this patchset aims to fix the following checkpatch
warnings:
occurences of trailing semicolons in macros
uses of __attribute__(format(printf, ...)) where __printf(...) can be
used
uses of __attribute__(aligned(size)) where __aligned(size) can be used
uses of __attribute__(packed) where __packed can be used
Regards,
Mario.
Mario J. Rugiero (4):
staging/lustre: checkpatch cleanup: macros should not use a trailing
semicolon
staging/lustre: checkpatch cleanup: __printf(string-index,
first-to-check) is preferred over __attribute__((format(printf,
string-index, first-to-check)))
staging/lustre: checkpatch cleanup: __aligned(size) is preferred over
__attribute__((aligned(size)))
staging/lustre: checkpatch cleanup: __packed is preferred over
__attribute__((packed))
.../lustre/include/linux/libcfs/libcfs_debug.h | 6 +--
.../include/linux/libcfs/libcfs_kernelcomm.h | 4 +-
.../lustre/include/linux/libcfs/libcfs_private.h | 4 +-
.../staging/lustre/include/linux/lnet/lib-types.h | 2 +-
.../staging/lustre/lustre/include/lprocfs_status.h | 6 +--
drivers/staging/lustre/lustre/include/lu_object.h | 2 +-
.../lustre/lustre/include/lustre/lustre_idl.h | 44 +++++++++++-----------
.../lustre/lustre/include/lustre/lustre_user.h | 14 +++----
drivers/staging/lustre/lustre/include/lustre_dlm.h | 2 +-
drivers/staging/lustre/lustre/include/lustre_net.h | 2 +-
10 files changed, 43 insertions(+), 43 deletions(-)
--
2.3.2
7 years, 2 months
[PATCH 0/4] staging/lustre: checkpatch cleanup
by mrugiero@gmail.com
From: "Mario J. Rugiero" <mrugiero(a)gmail.com>
Hi, as listed in the todo for this driver is fixing checkpatch
warnings, I wrote this patchset which fixes the following issues on all
of the driver files, one patch per issue:
trailing semicolons in macros
uses of __attribute__((format(printf,...)) instead of __printf(...)
uses of __attribute__(aligned(size)) insted of __aligned(size)
uses of __attribute__((packed)) instead of __packed
Each patch should be pretty much self-contained.
As I'm aware that my communication skills aren't the best, aside from
the obvious criticism of patches, constructive criticism on how to
communicate better with the list is accepted as well. If I broke any
netiquette (I did my best, but something might have slipped), I'm also
open to hearing.
Thanks a lot in advance,
Mario.
Mario J. Rugiero (4):
staging/lustre: checkpatch cleanup: macros should not use a trailing
semicolon
staging/lustre: checkpatch cleanup: __printf(string-index,
first-to-check) is preferred over __attribute__((format(printf,
string-index, first-to-check)))
staging/lustre: checkpatch cleanup: __aligned(size) is preferred over
__attribute__((aligned(size)))
staging/lustre: checkpatch cleanup: __packed is preferred over
__attribute__((packed))
.../lustre/include/linux/libcfs/libcfs_debug.h | 6 +--
.../include/linux/libcfs/libcfs_kernelcomm.h | 4 +-
.../lustre/include/linux/libcfs/libcfs_private.h | 4 +-
.../staging/lustre/include/linux/lnet/lib-types.h | 2 +-
.../staging/lustre/lustre/include/lprocfs_status.h | 6 +--
drivers/staging/lustre/lustre/include/lu_object.h | 2 +-
.../lustre/lustre/include/lustre/lustre_idl.h | 44 +++++++++++-----------
.../lustre/lustre/include/lustre/lustre_user.h | 14 +++----
drivers/staging/lustre/lustre/include/lustre_dlm.h | 2 +-
drivers/staging/lustre/lustre/include/lustre_net.h | 2 +-
10 files changed, 43 insertions(+), 43 deletions(-)
--
2.3.2
7 years, 2 months