Hi Anna,
We’re about to deploy Lustre/HSM with Phobos for a new large research data
archival system at Stanford (200PB).
https://github.com/phobos-storage
Phobos is open source and a Lustre copytool is available. Archiving policies
can be set up via Robinhood like with other HSMs. Robinhood
, 2023 at 2:07 AM Stephane Thiell
mailto:sthi...@stanford.edu>> wrote:
Hi Robin,
Sorry to hear about your problem.
A few questions…
Why did you run e2fsck?
Did e2fsck fix something?
What version of e2fsprogs are you using?
errno 28 is ENOSPC, what does dumpe2fs say about available spac
Hi Robin,
Sorry to hear about your problem.
A few questions…
Why did you run e2fsck?
Did e2fsck fix something?
What version of e2fsprogs are you using?
errno 28 is ENOSPC, what does dumpe2fs say about available space?
You can check the values of "Free blocks” and "Free inodes” using this
lctl del_ost was added in the upcoming Lustre 2.16 to remove a specific OST
while the filesystem is online. What it does is actually launching the
llog_cancel commands for the specified OST, hence with reduced risk of user
errors (like what happened here). We have used it several times in
Hi Andrew,
kernel: LustreError: 13921:0:(genops.c:478:class_register_device())
astrofs-OST-osc-MDT0001: already exists, won't add
is symptomatic of a llog index issue/mismatch on the MDT vs. MGT. I would check
if the llog backup of MDT0001 (over ldiskfs in CONFIGS) matches the one on the
Salut Diego!
Yes, we have been using NRS TBF by UID on our Oak storage system for months now
with Lustre 2.12. It’s a capacity-oriented, global filesystem, not designed for
heavy workloads (unlike our scratch filesystem) but with many users and as
such, a great candidate for NRS TBF UID. Since
Hi Riccardo,
I would check if the OSTs on this OSS have been registered with the correct
NIDs (o2ib1) on the MGS:
$ lctl --device MGS llog_print -client
and look for the NIDs in setup/add_conn for the OSTs in question.
Best,
Stephane
> On Sep 28, 2021, at 9:52 AM, Riccardo Veraldi
>
Hi Thomas,
Nodemap’s audit_mode is not defaulting to 1 on upgrade to Lustre 2.12. If you
have recently experienced the issue when upgraded a filesystem from 2.10 to
Lustre 2.12, so please check that flag maybe.
Hi Amit,
Your number is very low indeed.
At our site, we're seeing ~100 million files/day during a Robinhood scan with
nb_threads_scan =4 and on hardware using Intel based CPUs:
2020/11/16 07:29:46 [126653/2] STATS | avg. speed (effective): 1207.06
entries/sec (3.31 ms/entry/thread)
Hi Uwe,
We had a similar problem in the past and our conclusion was that the in-kernel
OFED drivers (provided by distribution) likely doesn’t have all the patches
required for SR-IOV to work correctly in VMs (when using VFs on IB HBAs and
KVM). Mellanox OFED is required at least on the VMs
Hi Alastair,
The first thing to do is to upgrade your servers to 2.12.3, as many bugs have
been fixed.
http://wiki.lustre.org/Lustre_2.12.3_Changelog
Stephane
> On Nov 20, 2019, at 7:29 AM, BASDEN, ALASTAIR G.
> wrote:
>
> Hi,
>
> We have a new 2.12.2 system, and are seeing fairly
Hi Americo,
In my experience, you need a proper kmod-mlnx-ofa_kernel RPM installed for the
Lustre build process to find the correct symbols.
To generate the kmod-mlnx-ofa_kernel RPM for the current kernel (in my case,
Lustre patched, server-side), you can try:
$ rpmbuild --rebuild --define
Hi Tung-Han,
Your stack trace looks similar to the one we’ve just seen yesterday on our
2.10.6 system.
I’ve open https://jira.whamcloud.com/browse/LU-12136 to track the issue.
Best,
Stephane
> On Mar 29, 2019, at 8:47 PM, Tung-Han Hsieh
> wrote:
>
> Dear All,
>
> Our system was recently
On one of our filesystem, we add a few new OSTs almost every month with no
downtime, this is very convenient. The only thing that I would recommend is to
avoid doing that during a peak of I/Os on your filesystem (we usually do it as
early as possible in the morning), as the added OSTs will
Hi Steve,
This could be LU-5152 (https://jira.whamcloud.com/browse/LU-5152), which
tentatively tried to fix unprivileged chgrp -R. The patch introduced some kind
of dependency between servers in the quota handling. It has been reverted in
2.10.6, however it’s not clear to me what the plan for
Hi Alex,
I’m no ZFS expert, but for a new project I recently faced some read performance
issues too when doing some zfs 0.7 testing, but not at all as bad as you… so I
feel sorry for you as you seem to have done quite a good work so far… so
perhaps some ideas:
- check that arc is fully
Share your experiences with colleagues. Submit an abstract to present at LUG17.
Abstracts are now due March 10, 2017.
The committee is particularly looking for presentations regarding:
• experiences running the newer community releases - 2.8 and 2.9 - in
production
• experiences
ly
>
>
>
> On Mon, Nov 7, 2016 at 5:59 PM, Stephane Thiell <sthi...@stanford.edu> wrote:
>
> > On Nov 4, 2016, at 11:05 PM, Dilger, Andreas <andreas.dil...@intel.com>
> > wrote:
> >
> > Actually, the nodemap feature will work with any client,
> On Nov 4, 2016, at 11:05 PM, Dilger, Andreas wrote:
>
> Actually, the nodemap feature will work with any client, since it is only
> affecting lookups on the MDS and quota on the OSS.
Great! :-)
>
> It probably would take less time for you to implement the flag
Hi all,
I am currently evaluating Lustre 2.8.60 with UID/GID mapping for a Lustre
storage system that will be connected to multiple clusters. These clusters have
matching UIDs but same groups have different GIDs. Actually GID mapping is
working fine but I can’t find an easy way to avoid
Hi Fernando,
To remove your old and empty OSTs during a maintenance: stop your filesystem,
do the writeconf on all targets and remount your MGS/MDS and then all OSTs
minus the old ones.
With shine, stop your filesystem, comment out the old OST lines in the fs model
file and type “shine
Hi Phil,
I understand that you’re running master on your clients (tag v2_8_56 was
created 4 days ago) and 2.1 on the servers? Running master in production is
already a challenge. Also Lustre has never be good for cross-version
compatibility. For example, it is possible to make 2.1 servers work
22 matches
Mail list logo