Joshua Baker-LePain wrote:
I currently backup a bit more than 16TB to an LTO3 library and don't
find it that painful. I use AMANDA and break the data down into
bite-size chunks. AMANDA handles spreading these chunks out over the
whole backup cycle, so that each night's backup is about the s
On Fri, 10 Apr 2009, Robert G. Brown wrote:
I'll look there, but as I'm typing this on a Dell M1530 that even as of
Fedora 10 still has non-functional hardware components and marginal
hardware components, forgive me if I'm a bit cynical. I'll believe that
Dell is properly supporting linux when:
Joshua Baker-LePain wrote:
> I currently backup a bit more than 16TB to an LTO3 library and don't find
> it that painful. I use AMANDA and break the data down into bite-size
> chunks. AMANDA handles spreading these chunks out over the whole backup
> cycle, so that each night's backup is abou
On Fri, 10 Apr 2009, Mark Hahn wrote:
otoh, I can't think of why yum couldn't trigger rebuilds itself as part
of resolving dependencies. maybe it does already...
This would be the best way. vmware (among its other current virtues)
initiates an automated rebuild after a kernel update the firs
On Fri, 10 Apr 2009 at 1:15pm, David Mathog wrote
Thankfully I don't have to do this myself, not having data anywhere near
that size to cope with, but it seems to me that backing up a nearly full
16TB RAID is likely to be a painful, expensive, exercise.
Going with tape first...
The fastest tap
Billy Crook wrote:
> As a very,
> very, general rule, you might put no more than 8TB in a raid5, and no
> more than 16TB in a raid6, including what's used for parity, and
> assuming magnetic, enterprise/raid drives. YMMV, Test all new drives,
> keep good backups, etc...
Thankfully I don't have
2009/4/10 Robert G. Brown :
>
> anything. I use my laptop install primarily for access to a Cisco VPN
> (because Cisco sucks and still doesn't support linux worth a damn) and
> to be able to run/debug a Windows-only EMR client.
You're right there Bob. I have a shiny Sun/Opteron workstation at
hom
On Fri, 10 Apr 2009, Mark Hahn wrote:
HP has its own distro, but is still trying to use a traditional approach to
making patches patches available. (ie, ftp patch files
that unpack to rpm(s), install script and docs). it seems pretty obvious
that yum repos are the way to go (is there any _tec
in other words, with a normal all-precompiled distro, dkms still seems
redundant.
True, but when you add something like the NVidia proprietary drivers
which have a kernel component and aren't a part any distro due to
licensing/distribution restrictions, DKMS is a fantastic.
you missed me there
On Fri, Apr 10, 2009 at 12:27, Joe Landman
wrote:>
> We have 1 customer using 24 drives (22 for data with 2 hot spares) as an md
> raid6 on DeltaV. Normally we'd suggest something smaller (collections of
> RAID6 and then striping across them to form RAID60's).
>
> With late model kernels, mdadm,
Michael Will wrote:
raid6 is also new code with new bugs that can lead to dataloss as well,
regardless of its nice 'can survive
two drive failures' feature. I have seen it happen.
All code (anywhere) can have bugs. Arguing that raid6 module has bugs a
non-sequitur. It is well tested, and
Mark Hahn wrote:
>
> in other words, with a normal all-precompiled distro, dkms still seems
> redundant.
True, but when you add something like the NVidia proprietary drivers
which have a kernel component and aren't a part any distro due to
licensing/distribution restrictions, DKMS is a fantastic.
raid6 is also new code with new bugs that can lead to dataloss as well,
regardless of its nice 'can survive
two drive failures' feature. I have seen it happen.
Michael
-Original Message-
From: beowulf-boun...@beowulf.org on behalf of Orion Poplawski
Sent: Fri 4/10/2009 10:23 AM
To: land
Orion Poplawski wrote:
Joe Landman wrote:
Stuart Midgley wrote:
Good work Stuart!
What are the lessons learnt? Well with software raid Linux is both your
1) Use RAID6. It is your friend. RAID5 is unashamedly your enemy.
So, what's a reasonably limit on the number of disks in a RAID6 arr
Joe Landman wrote:
Stuart Midgley wrote:
Good work Stuart!
What are the lessons learnt? Well with software raid Linux is both your
1) Use RAID6. It is your friend. RAID5 is unashamedly your enemy.
So, what's a reasonably limit on the number of disks in a RAID6 array
using 1.5TB disks?
On Fri, 10 Apr 2009, Kilian CAVALOTTI wrote:
They do already. Well, except maybe for the dumb-dumb GUI and the
games. But pretty much all wha'is required to make your servers run
(including their OpenManage suite, drivers and firmwares) is available
in RPM repositories. I understand it's not ver
Bill Broadley wrote:
Guy Coates wrote:
Yikes, epic recovery.
What are the lessons learnt?
You forgot the obvious one.
I suggest ditching silly old centos/redhat kernels and run something new
enough to allow for scrubbing. So that all your disks don't silently start
collecting errors waitin
new kernel until you try - it might have dependencies on a symbol that
gets removed from the kernel update, for instance.
Right, although distributions like RHEL do a good job of keeping the
kernel unchanged from an API perspective within a given release.
well, I wasn't thinking about the semi
On Fri, 10 Apr 2009 at 12:51pm, Mark Hahn wrote
Erm, I just (as in, this week) upgraded the BIOS of a batch of DL160 G5s
from within Linux.
nice. for a lot of models, they make available a dos/win exe
that produces a bootable floppy image. kind of a pain.
Indeed.
being able to check and
2009/4/10 Mark Hahn :
>
> I'm not so sure about that - why would VMed windows be more secure?
> my understanding is that the thing that makes windows vulnerable is the
> hooks that make windows integration work. and it's the integration
> that people expect, no?
One reason you might want to do V
Mark Hahn wrote:
I'd like to add that Dell's DKMS (Dynamics Kernel Management System) is
great:
http://linux.dell.com/projects.shtml#dkms
really? I've never much seen the point, since when I want a kernel
update, it's almost never for drivers, but more fundamental parts of the
kernel, often
HP distributes linux-based firmware updates, but only for add-in devices
(scsi controllers, BMC, etc), not the bios itself...
Erm, I just (as in, this week) upgraded the BIOS of a batch of DL160 G5s from
within Linux.
nice. for a lot of models, they make available a dos/win exe
that produces
Mark Hahn wrote:
>> It's useful because it will automatically build and install existing
>> kernel modules for newly-installed kernels. Many vendors ship drivers as
>> RPMs separate from the kernel, so they won't get updated when the kernel
>> is updated unless you use something like dkms.
>
> inte
It's useful because it will automatically build and install existing
kernel modules for newly-installed kernels. Many vendors ship drivers as
RPMs separate from the kernel, so they won't get updated when the kernel
is updated unless you use something like dkms.
interesting. the distro-based app
On Fri, 10 Apr 2009 at 12:26pm, Mark Hahn wrote
HP distributes linux-based firmware updates, but only for add-in devices
(scsi controllers, BMC, etc), not the bios itself...
Erm, I just (as in, this week) upgraded the BIOS of a batch of DL160 G5s
from within Linux.
--
Joshua Baker-LePain
QB
easier (DKMS). Upgrading a BIOS version from a Linux shell is not
something you want to give up after you tried it.
if a vendor wanted to do a good job, I think that opening/documenting
the bios would be a huge win. for instance, giving us a way to poke
(or at least peek) the actual bios settin
Mark Hahn wrote:
>>> I'd like to add that Dell's DKMS (Dynamics Kernel Management System) is
>>> great:
>>>
>>> http://linux.dell.com/projects.shtml#dkms
>
> really? I've never much seen the point, since when I want a kernel
> update, it's almost never for drivers, but more fundamental parts of
>
I'd like to add that Dell's DKMS (Dynamics Kernel Management System) is
great:
http://linux.dell.com/projects.shtml#dkms
really? I've never much seen the point, since when I want a kernel
update, it's almost never for drivers, but more fundamental parts
of the kernel, often not even modules.
On Fri, Apr 10, 2009 at 5:39 PM, Kilian CAVALOTTI
wrote:
>>> http://linux.dell.com/projects.shtml#dkms
>>>
>>> I understand Jeff is relatively new to Dell. Maybe in the future we'll
>>> see his influence improve Dell's clustering situation.
>>
>> I'm sure that he will! And Linux in general. This
On Fri, Apr 10, 2009 at 4:53 PM, Robert G. Brown wrote:
> On Fri, 10 Apr 2009, Prentice Bisbal wrote:
>
>> While we're all repenting for our sins against Dell on this Good Friday,
>> I'd like to add that Dell's DKMS (Dynamics Kernel Management System) is
>> great:
>>
>> http://linux.dell.com/proje
On Fri, 10 Apr 2009, Prentice Bisbal wrote:
While we're all repenting for our sins against Dell on this Good Friday,
I'd like to add that Dell's DKMS (Dynamics Kernel Management System) is
great:
http://linux.dell.com/projects.shtml#dkms
I understand Jeff is relatively new to Dell. Maybe in th
On Fri, 10 Apr 2009, Lux, James P wrote:
...I don't know that they're harmonics, per se, but just the
phenomenon of aerodynamic flutter, which was very poorly understood and
is basically a resonance thing. (the nonlinearities in the state
Um, "basically a resonance thing" = "harmonics" in a li
Jim,
Thanks for clearing that up. I somehow knew you'd be the one to provide
the answer. ;)
Lux, James P wrote:
> ...On Thu, 9 Apr 2009, Prentice Bisbal wrote:
>
>> I disagree with the sonic barrier wall analaogy. Is it that clearly
>> technical barrier the slowed down jet research, or did the n
On Thu, 9 Apr 2009, Peter St. John wrote:
Yes, supersonic travel over land is restricted in the U.S.; so for example
the SST was good for NY to London but not NY to LA.
We're no where near the limits to information per cubic centimter; but I
don't know how to define information **processing** p
While we're all repenting for our sins against Dell on this Good Friday,
I'd like to add that Dell's DKMS (Dynamics Kernel Management System) is
great:
http://linux.dell.com/projects.shtml#dkms
I understand Jeff is relatively new to Dell. Maybe in the future we'll
see his influence improve Dell'
Stuart Midgley wrote:
Good work Stuart!
What are the lessons learnt? Well with software raid Linux is both your
1) Use RAID6. It is your friend. RAID5 is unashamedly your enemy.
2) Scrub early, scrub often. We cron this ~1/week on Delta-V's (sounds
similar to your box).
3) pay attentio
Guy Coates wrote:
> Yikes, epic recovery.
>
>> What are the lessons learnt?
>
> You forgot the obvious one.
I suggest ditching silly old centos/redhat kernels and run something new
enough to allow for scrubbing. So that all your disks don't silently start
collecting errors waiting to cascade in
One of the things I like about VIM is that I can install it everywhere. I
use it on VMS, as well as unices and MSWin. That vivivi is the Editor of the
Beast just adds flavor :-) vim.org has MSWin self-extracting executable.
Peter
On 4/6/09, Geoff Galitz wrote:
>
>
>
>
>
> Can anyone recommend a g
Yikes, epic recovery.
> What are the lessons learnt?
You forgot the obvious one. As has been discussed here before, raid5
and disks > 1TB are a bad combination, and raid6 is your friend.
Cheers,
Guy
--
Dr. Guy Coates, Informatics System Group
The Wellcome Trust Sanger Institute, Hinxton, Ca
From: beowulf-boun...@beowulf.org [beowulf-boun...@beowulf.org] On Behalf Of
Robert G. Brown [...@phy.duke.edu]
Sent: Thursday, April 09, 2009 2:31 PM
To: Prentice Bisbal
Cc: Beowulf Mailing List
Subject: Re: [Beowulf] Moores Law is dying
...On Thu, 9 Apr
40 matches
Mail list logo
