On 2015/05/21, 5:34 AM, "Lukas Hejtmanek" wrote:
>some time ago I asked whether it is possible to run lustre server (OSD)
>and client (mount -t lustre) on the same machine. Somewhere close to
>Lustre 1.3.x, the answer was not, such setup is not recommended due to VM
>pressure from client and serve
OK, thanks.
Olaf P. Faaland
LLNL
From: Alexander I Kulyavtsev [a...@fnal.gov]
Sent: Thursday, May 21, 2015 4:25 PM
To: Faaland, Olaf P.
Cc: Alexander I Kulyavtsev; lustre-discuss@lists.lustre.org
Subject: Re: [lustre-discuss] "tag" feature of ltop
As you said in e
As you said in earlier mail, when I sort by IO rate, locks, etc., selected OST
are jumping around or just in different rows as seen below.
Anyway, this is not strong desire in the long term: I may feed cerebro output
to the web page.
Best regards, Alex.
On May 21, 2015, at 5:42 PM, Faaland, Ol
Alexander,
Thanks for your reply.
ltop also lets you sort by OSS, so that the OSTs sharing an OSS are all next to
each other. Do you find tagging more helpful than that?
Olaf P. Faaland
LLNL
From: Alexander I Kulyavtsev [a...@fnal.gov]
Sent: Thursday, May 21, 2
It may have sense to keep tagging.
I marked OSS dslustre15 and then switched to OST view. I have all OSTs on
marked OSS highlighted:
005c F dslustre13 10160 1 0 16503001 95 80
005d F dslustre14 10160 0 0 07472001 95 79
00
And just to add one more item to the list:
If you are running ZFS as the backend storage and have enabled compression, you
need to use an additional flag to du:
du -sh Will give you what is stored on the disk
du --apparent-size will tell you what the actual size of the data is.
I doubt this
I do not think that is the case. Mixing OSTs and clients is often bad
as well. It is generally a question of memory contention.
You can get away with mixing clients with any of the server types if
your demands are modest enough. That is why we sometimes say it is "OK
for testing". But even
I frequently use my hot-spare MDT machine for brief testing with a client.
Usually when we have scheduled outages. I wouldn't rely on it for a day to day
thing, but for testing to see if everything is working properly after patches,
outages, upgrades, etc it seems to work fine. I've used the
Hi,
a few more things which may play a role:
- as you are suspecting, the difference of used blocks vs. used bytes
might be the reason, especially if there are many very small files, but
there are more possible causes:
- some tools use 2^10 bytes and some others use 1000 bytes as kb which
might
I believe it is OK for some simple functional testing, but not production, or
working with large datasets.
OSTs and MDTs can coexist, and I believe OSTs and Clients can coexist
relatively peacefully, but MDTs and Clients are a bad mix.
-Ben Evans
-Original Message-
From: lustre-discus
Hello,
some time ago I asked whether it is possible to run lustre server (OSD) and
client (mount -t lustre) on the same machine. Somewhere close to Lustre 1.3.x,
the answer was not, such setup is not recommended due to VM pressure from
client and server.
Is there anything new with current mainlin
Hi all,
I want to make you aware that a Lustre bug might be the reason for
wrong Lustre quotas, see LU-4345.
With this bug chgrp or chown with only UID or GID specified can create
an arbitrary GID or UID for OST objects on the ldiskfs level.
This causes wrong Lustre quotas because they are compu
12 matches
Mail list logo
