April 2015 - HPDD-discuss

Re: [HPDD-discuss] Lustre 1.8.9 wc1 for Kernel 2.6.32-504 - again

by Chris Hunter

We apply patch for LU2800 & LU5237 (b2d09eb6,diff?) to git b1_8 branch. A couple optional patches for LU3067 & LU3596 are available. It has become difficult to build 1.8 clients with recent redhat kernels (they tend backport fixes from the 3.10 kernels). We have not moved to newer client version because of support for legacy OS and (perceived) client performance issues with 2.1/2.4/2.5 versions. IMO the b2_5 client issues appear to be resolved so that's likely our upgrade path. Note it appears the b1_8 branch was closed last month and all open pull requests were abandonded. regards, chris hunter yale hpc group > Hi, > > hmm... there are different versions of the patch and different git > repositories around. This one: > works for the github branch you mentioned: > > patch -p1 < ~/Downloads/2661328b.diff > patching file lustre/llite/dir.c > Hunk #1 succeeded at 1051 with fuzz 1 (offset -126 lines). > Hunk #2 succeeded at 1197 (offset -112 lines). > Hunk #3 succeeded at 1264 (offset -117 lines).

5 years, 6 months

4
5
0 / 0

[PATCH] staging: lustre: declare file operations structure as static

by Gujulan Elango, Hari Prasath (H.)

The fops structure is referenced only locally and hence could be made static Signed-off-by: Hari Prasath Gujulan Elango <hgujulan(a)visteon.com> --- drivers/staging/lustre/lustre/mdc/lproc_mdc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/lustre/lustre/mdc/lproc_mdc.c b/drivers/staging/lustre/lustre/mdc/lproc_mdc.c index acfe08e..fbeddf9 100644 --- a/drivers/staging/lustre/lustre/mdc/lproc_mdc.c +++ b/drivers/staging/lustre/lustre/mdc/lproc_mdc.c @@ -147,7 +147,7 @@ static ssize_t mdc_kuc_write(struct file *file, return count; } -struct file_operations mdc_kuc_fops = { +static const struct file_operations mdc_kuc_fops = { .open = mdc_kuc_open, .write = mdc_kuc_write, .release = single_release, -- 1.9.1

5 years, 10 months

3
2
0 / 0

[PATCH] staging: lustre: fix ifnullfree.cocci warnings

by kbuild test robot

drivers/staging/lustre/lustre/lov/lov_request.c:74:3-8: WARNING: NULL check before freeing functions like kfree, debugfs_remove, debugfs_remove_recursive or usb_free_urb is not needed. Maybe consider reorganizing relevant code to avoid passing NULL values. NULL check before some freeing functions is not needed. Based on checkpatch warning "kfree(NULL) is safe this check is probably not required" and kfreeaddr.cocci by Julia Lawall. Generated by: scripts/coccinelle/free/ifnullfree.cocci CC: Julia Lawall <Julia.Lawall(a)lip6.fr> Signed-off-by: Fengguang Wu <fengguang.wu(a)intel.com> --- lov_request.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) --- a/drivers/staging/lustre/lustre/lov/lov_request.c +++ b/drivers/staging/lustre/lustre/lov/lov_request.c @@ -70,8 +70,7 @@ void lov_finish_set(struct lov_request_s OBDO_FREE(req->rq_oi.oi_oa); if (req->rq_oi.oi_md) OBD_FREE_LARGE(req->rq_oi.oi_md, req->rq_buflen); - if (req->rq_oi.oi_osfs) - kfree(req->rq_oi.oi_osfs); + kfree(req->rq_oi.oi_osfs); kfree(req); }

5 years, 10 months

1
0
0 / 0

[PATCH] staging: lustre: llite: fix ifnullfree.cocci warnings

by kbuild test robot

drivers/staging/lustre/lustre/llite/dir.c:1440:3-8: WARNING: NULL check before freeing functions like kfree, debugfs_remove, debugfs_remove_recursive or usb_free_urb is not needed. Maybe consider reorganizing relevant code to avoid passing NULL values. NULL check before some freeing functions is not needed. Based on checkpatch warning "kfree(NULL) is safe this check is probably not required" and kfreeaddr.cocci by Julia Lawall. Generated by: scripts/coccinelle/free/ifnullfree.cocci CC: Julia Lawall <Julia.Lawall(a)lip6.fr> Signed-off-by: Fengguang Wu <fengguang.wu(a)intel.com> --- dir.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) --- a/drivers/staging/lustre/lustre/llite/dir.c +++ b/drivers/staging/lustre/lustre/llite/dir.c @@ -1436,8 +1436,7 @@ lmv_out_free: goto free_lmv; } free_lmv: - if (tmp) - kfree(tmp); + kfree(tmp); return rc; } case LL_IOC_REMOVE_ENTRY: {

5 years, 10 months

1
0
0 / 0

[PATCH] staging: lustre: llite: fix ifnullfree.cocci warnings

by kbuild test robot

drivers/staging/lustre/lustre/llite/llite_lib.c:989:2-7: WARNING: NULL check before freeing functions like kfree, debugfs_remove, debugfs_remove_recursive or usb_free_urb is not needed. Maybe consider reorganizing relevant code to avoid passing NULL values. drivers/staging/lustre/lustre/llite/llite_lib.c:991:2-7: WARNING: NULL check before freeing functions like kfree, debugfs_remove, debugfs_remove_recursive or usb_free_urb is not needed. Maybe consider reorganizing relevant code to avoid passing NULL values. NULL check before some freeing functions is not needed. Based on checkpatch warning "kfree(NULL) is safe this check is probably not required" and kfreeaddr.cocci by Julia Lawall. Generated by: scripts/coccinelle/free/ifnullfree.cocci CC: Julia Lawall <Julia.Lawall(a)lip6.fr> Signed-off-by: Fengguang Wu <fengguang.wu(a)intel.com> --- llite_lib.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) --- a/drivers/staging/lustre/lustre/llite/llite_lib.c +++ b/drivers/staging/lustre/lustre/llite/llite_lib.c @@ -985,10 +985,8 @@ int ll_fill_super(struct super_block *sb err = client_common_fill_super(sb, md, dt, mnt); out_free: - if (md) - kfree(md); - if (dt) - kfree(dt); + kfree(md); + kfree(dt); if (err) ll_put_super(sb); else if (sbi->ll_flags & LL_SBI_VERBOSE)

5 years, 10 months

1
0
0 / 0

Lustre 2.8 update - April 24th 2015

by Jones, Peter A

http://wiki.opensfs.org/Lustre_2.8_-_April_24th_2015

5 years, 10 months

1
0
0 / 0

Are we supposed to switch to lustre-discuss?

by E.S. Rosenberg

I'm going through some of the slides from LUG, am I correct if I say we're supposed to switch to lustre-discuss? Will the old mailing lists be closed? I find it kind of ridiculous that I'm going to be registered on 3 distinct mailing lists about the same subject that don't get each others' mail.... Regards, Eli

5 years, 10 months

3
4
0 / 0

Lustre Server on Ubuntu 14.04

by Ben Archuleta

Hello All, I have two questions, first does Lustre 2.7.50 (server) support Ubuntu 14.04 running the 3.13.0-24-generic kernel? Secondly if it is supported, are their any good tutorials or documentation in regards to compiling, patching, ect ..... and installing the Lustre server? Our end goal is to run Lustre on Ubuntu and serve out files over InfiniBand. Regards -- Ben J. Archuleta Systems Staff Center for Advanced Research Computing (CARC) University of New Mexico 1601 Central Ave NE Albuquerque, NM 87131 MSC01 1190 PH: 505.382.7911 FX: 505.277.8235 barchu02(a)carc.unm.edu www.carc.unm.edu

5 years, 10 months

2
1
0 / 0

ldlm_poold causing high client load

by Kevin M. Hildebrand

We've been seeing a lot of our compute nodes where something is causing ldlm_poold to start chewing 100% CPU. The nodes are fine for a while, and then at some point something goes wrong, and ldlm_poold goes nuts. Sometimes there are a bunch of ldlm_bl_ processes also consuming large amounts of CPU as well. Once the node gets into this state, it has to be rebooted to fix it. The node won't even shut down cleanly, as umount of /lustre ends up hanging. There are no lustre-related messages in syslog while the erroneous behavior is taking place. The clients and servers have all been running 2.5.3 for the last few months on RedHat 6 (2.6.32-504.3.3.el6.x86_64) I saw the bug report LU-5415 that looks similar, but that report says that 2.5.3 includes a fix for that particular problem. Has anyone else seen this behavior? Thanks, Kevin --- Kevin Hildebrand Division of IT University of Maryland, College Park

5 years, 10 months

1
1
0 / 0

ldiskfs persistent mount options

by Chris Hunter

Hi, In previous lustre versions, when creating ldiskfs OSTs, the default mount options were "errors=remount-ro,extents,mballoc". Somewhere along the way (lustre 2.5.1?) the mount options changed to "errors=remount-ro". I suspect the change is because mballoc & extents are now defaults of the underlying ext4 filesystem. FYI Section 33 of lustre admin guide documents the ldiskfs persistent mount options for commands "mkfs.lustre" & "tunefs.lustre" but the doco in these two places are different. YMMV. chris hunter yale hpc group

5 years, 10 months

2
2
0 / 0

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

HPDD-discuss April 2015