Re: [lustre-discuss] Lustre Client in a container

[E.S. Rosenberg]

On Mon, Jan 1, 2018 at 12:38 AM, Patrick Farrell  wrote:

>
>
> Keeping in mind that the choice of NFS means that you don’t have the POSIX
> guarantees provided by Lustre, so simultaneous access to the same files is
> dicey unless it’s only reading.
>
That just means that during the migration phase the 2.10 filesystem should
not be production yet or alternatively the 1.8 system is out of prodution
and read-only which would anyhow be the goal of a migration or am I
seeing this totally wrong?
Good luck and thanks everyone on lustre-users for being an excelletn
mailing list :)
Eli

> --
> *From:* lustre-discuss  on
> behalf of Daniel Kobras 
> *Sent:* Sunday, December 31, 2017 12:26:49 PM
> *To:* David Cohen
> *Cc:* lustre-discuss@lists.lustre.org
> *Subject:* Re: [lustre-discuss] Lustre Client in a container
>
> Hi David!
>
> Do you require both systems to be available as native Lustre filesystems
> on all clients? Otherwise, reexporting one of the systems via NFS during
> the migration phase will keep all data available but decouple the version
> interdependence between servers and clients. In this situation, it’s
> probably the least experimental option.
>
> Kind regards,
>
> Daniel
>
> > Am 31.12.2017 um 09:50 schrieb David Cohen <
> cda...@physics.technion.ac.il>:
> >
> > Patrick,
> > Thanks for you response.
> > I looking for a way to migrate from 1.8.9 system to 2.10.2, stable
> enough to run the several weeks or more that it might take.
> >
> >
> > David
> >
> > On Sun, Dec 31, 2017 at 12:12 AM, Patrick Farrell  wrote:
> > David,
> >
> > I have no direct experience trying this, but I would imagine not -
> Lustre is a kernel module (actually a set of kernel modules), so unless the
> container tech you're using allows loading multiple different versions of
> *kernel modules*, this is likely impossible.  My limited understanding of
> container tech on Linux suggests that this would be impossible, containers
> allow userspace separation but there is only one kernel/set of
> modules/drivers.
> >
> > I don't know of any way to run multiple client versions on the same node.
> >
> > The other question is *why* do you want to run multiple client versions
> on one node...?  Clients are usually interoperable across a pretty generous
> set of server versions.
> >
> > - Patrick
> >
> >
> > From: lustre-discuss  on
> behalf of David Cohen 
> > Sent: Saturday, December 30, 2017 11:45:15 AM
> > To: lustre-discuss@lists.lustre.org
> > Subject: [lustre-discuss] Lustre Client in a container
> >
> > Hi,
> > Is it possible to run Lustre client in a container?
> > The goal is to run two different client version on the same node, can it
> be done?
> >
> > David
> >
> >
> > ___
> > lustre-discuss mailing list
> > lustre-discuss@lists.lustre.org
> > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
>
> ___
> lustre-discuss mailing list
> lustre-discuss@lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
>
> ___
> lustre-discuss mailing list
> lustre-discuss@lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
>
>
___
lustre-discuss mailing list
lustre-discuss@lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

Re: [lustre-discuss] how to get statistics from lustre clients ?

[Raj]

To add to the list, I like ‘xltop’ which works like a linux ‘top’ command
and integrates with your scheduler.
On Tue, Jan 2, 2018 at 9:25 AM Ben Evans  wrote:

> The OSTs and MDTs have client-specific information in their proc
> hierarchies, so there's no need for client-side monitoring if that's all
> you're looking for.
>
> -Ben Evans
>
> From: Christopher Johnston 
> Date: Tuesday, January 2, 2018 at 10:21 AM
> To: Ben Evans 
> Cc: "Black.S" , "lustre-discuss@lists.lustre.org" <
> lustre-discuss@lists.lustre.org>
> Subject: Re: [lustre-discuss] how to get statistics from lustre clients ?
>
> I use the telgraf plugin for MDS/OSS/OST stats.  I know its not the client
> side that you were after but I find the metrics to be very useful and
> display nicely in grafana.
>
> https://github.com/influxdata/telegraf/tree/master/plugins/inputs/lustre2
>
> On Tue, Jan 2, 2018 at 8:45 AM, Ben Evans  wrote:
>
>> You can get breakdowns of client->OST reads and writes, and combine them
>> into OSS-level info.
>>
>> There is currently no timestamp on it, all the stats files are cumulative
>> since you last cleared.  You can get around this by reading the file
>> regularly and noting the time, and doing the diffs since the last read.
>>
>> There are a few programs out there that do this sort of thing for you
>> already, collectl and collectd come to mind.
>>
>> -Ben Evans
>>
>> From: lustre-discuss  on behalf
>> of "Black.S" 
>> Date: Saturday, December 30, 2017 at 9:18 AM
>> To: "lustre-discuss@lists.lustre.org" 
>> Subject: [lustre-discuss] how to get statistics from lustre clients ?
>>
>> May be anybody know it or get the target where I can search
>>
>> I want to get from each lustre client:
>>
>>-
>>
>>how much lustre client read/write ?
>>-
>>
>>which size blocks ?
>>-
>>
>>with which OSS communicate ?
>>-
>>
>>timestemp of operation ( I mean time then start read/write from
>>lustre client) ?
>>
>> Is I can get that from (or by) linux node with lustre client?
>>
>> ___
>> lustre-discuss mailing list
>> lustre-discuss@lists.lustre.org
>> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
>>
>>
> ___
> lustre-discuss mailing list
> lustre-discuss@lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
>
___
lustre-discuss mailing list
lustre-discuss@lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

Re: [lustre-discuss] how to get statistics from lustre clients ?

[Ben Evans]

The OSTs and MDTs have client-specific information in their proc hierarchies, 
so there's no need for client-side monitoring if that's all you're looking for.

-Ben Evans

From: Christopher Johnston >
Date: Tuesday, January 2, 2018 at 10:21 AM
To: Ben Evans >
Cc: "Black.S" >, 
"lustre-discuss@lists.lustre.org" 
>
Subject: Re: [lustre-discuss] how to get statistics from lustre clients ?

I use the telgraf plugin for MDS/OSS/OST stats.  I know its not the client side 
that you were after but I find the metrics to be very useful and display nicely 
in grafana.

https://github.com/influxdata/telegraf/tree/master/plugins/inputs/lustre2

On Tue, Jan 2, 2018 at 8:45 AM, Ben Evans 
> wrote:
You can get breakdowns of client->OST reads and writes, and combine them into 
OSS-level info.

There is currently no timestamp on it, all the stats files are cumulative since 
you last cleared.  You can get around this by reading the file regularly and 
noting the time, and doing the diffs since the last read.

There are a few programs out there that do this sort of thing for you already, 
collectl and collectd come to mind.

-Ben Evans

From: lustre-discuss 
>
 on behalf of "Black.S" >
Date: Saturday, December 30, 2017 at 9:18 AM
To: "lustre-discuss@lists.lustre.org" 
>
Subject: [lustre-discuss] how to get statistics from lustre clients ?


May be anybody know it or get the target where I can search

I want to get from each lustre client:

  *   how much lustre client read/write ?

  *   which size blocks ?

  *   with which OSS communicate ?

  *   timestemp of operation ( I mean time then start read/write from lustre 
client) ?

Is I can get that from (or by) linux node with lustre client?

___
lustre-discuss mailing list
lustre-discuss@lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org


___
lustre-discuss mailing list
lustre-discuss@lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

Re: [lustre-discuss] how to get statistics from lustre clients ?

[Christopher Johnston]

I use the telgraf plugin for MDS/OSS/OST stats.  I know its not the client
side that you were after but I find the metrics to be very useful and
display nicely in grafana.

https://github.com/influxdata/telegraf/tree/master/plugins/inputs/lustre2

On Tue, Jan 2, 2018 at 8:45 AM, Ben Evans  wrote:

> You can get breakdowns of client->OST reads and writes, and combine them
> into OSS-level info.
>
> There is currently no timestamp on it, all the stats files are cumulative
> since you last cleared.  You can get around this by reading the file
> regularly and noting the time, and doing the diffs since the last read.
>
> There are a few programs out there that do this sort of thing for you
> already, collectl and collectd come to mind.
>
> -Ben Evans
>
> From: lustre-discuss  on behalf
> of "Black.S" 
> Date: Saturday, December 30, 2017 at 9:18 AM
> To: "lustre-discuss@lists.lustre.org" 
> Subject: [lustre-discuss] how to get statistics from lustre clients ?
>
> May be anybody know it or get the target where I can search
>
> I want to get from each lustre client:
>
>-
>
>how much lustre client read/write ?
>-
>
>which size blocks ?
>-
>
>with which OSS communicate ?
>-
>
>timestemp of operation ( I mean time then start read/write from lustre
>client) ?
>
> Is I can get that from (or by) linux node with lustre client?
>
> ___
> lustre-discuss mailing list
> lustre-discuss@lists.lustre.org
> http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
>
>
___
lustre-discuss mailing list
lustre-discuss@lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

Re: [lustre-discuss] how to get statistics from lustre clients ?

[Ben Evans]

You can get breakdowns of client->OST reads and writes, and combine them into 
OSS-level info.

There is currently no timestamp on it, all the stats files are cumulative since 
you last cleared.  You can get around this by reading the file regularly and 
noting the time, and doing the diffs since the last read.

There are a few programs out there that do this sort of thing for you already, 
collectl and collectd come to mind.

-Ben Evans

From: lustre-discuss 
>
 on behalf of "Black.S" >
Date: Saturday, December 30, 2017 at 9:18 AM
To: "lustre-discuss@lists.lustre.org" 
>
Subject: [lustre-discuss] how to get statistics from lustre clients ?


May be anybody know it or get the target where I can search

I want to get from each lustre client:

  *   how much lustre client read/write ?

  *   which size blocks ?

  *   with which OSS communicate ?

  *   timestemp of operation ( I mean time then start read/write from lustre 
client) ?

Is I can get that from (or by) linux node with lustre client?
___
lustre-discuss mailing list
lustre-discuss@lists.lustre.org
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org