| 1 |
Donnie Berkholz writes: |
| 2 |
> Bryan Green wrote: |
| 3 |
> > I am looking for something of a survey of examples of Gentoo-driven cluster |
| 4 |
> s |
| 5 |
> > out there. If such a survey has been done, perhaps someone point me to it. |
| 6 |
> |
| 7 |
> http://www.gentoo.org/proj/en/cluster/#doc_chap2 |
| 8 |
|
| 9 |
Whoa, I don't know how I missed that page. Thanks! |
| 10 |
|
| 11 |
> |
| 12 |
> Joel Martin has previously posted Lustre ebuilds to the list (for both |
| 13 |
> client and server, I thinkg). You may be interested. We'll want to get |
| 14 |
> them into portage at some point, so there's no requirement that you use |
| 15 |
> Suse server-side. |
| 16 |
|
| 17 |
Yes, I actually used those ebuilds to test Lustre on our "mini" 3x3 |
| 18 |
hyperwall which runs Gentoo. I was able to get it working, but over here |
| 19 |
they want the supported, released version, whereas those ebuilds are for the |
| 20 |
beta. I tried to install the released version, but eventually ran into |
| 21 |
problems. Also, since getting support from CFS is a requirement, that |
| 22 |
restricts the OS choice to specific versions of Suse or Redhat. |
| 23 |
|
| 24 |
The beta is supposed to become stable and supported by what was January, but |
| 25 |
now has apparently been pushed out to March. :( The only chance of putting |
| 26 |
Gentoo on the nodes of this cluster is if we can decide to go with the |
| 27 |
version that is still currently in beta. This is because the beta, version |
| 28 |
1.6, has a "patchless client", and so CFS is agnostic about OS on the client |
| 29 |
side. For the server side, support=Suse as far as anyone I've talked to is |
| 30 |
concerned. |
| 31 |
|
| 32 |
> |
| 33 |
> > I'd be grateful for any feedback I get from others on the list about the |
| 34 |
> > clusters they maintain or use, and perhaps some comments about the efficacy |
| 35 |
> > of Gentoo in an environment where stability is very important, and how |
| 36 |
> > system administration compares to administration of a Suse or Redhat cluste |
| 37 |
> r. |
| 38 |
> |
| 39 |
> The main difference is that, since we're "live," you need to consider |
| 40 |
> how you want to deal with upgrades. You may wish to pick a static |
| 41 |
> portage tree, import it into some sort of version control, and |
| 42 |
> selectively import changes you want (probably just security bumps, which |
| 43 |
> you can find using the wonderful glsa-check tool from gentoolkit). |
| 44 |
> |
| 45 |
> I've got a glsa-check wrapper that I use to make things a little easier, |
| 46 |
> which shows and optionally applies applicable updates. I attached it. |
| 47 |
> |
| 48 |
|
| 49 |
I'd very interested in the different approaches here. I had thought about a |
| 50 |
static portage tree, but that left the problem of getting needed updates, |
| 51 |
especially GLSA's. Your suggested approach sounds very interesting. |
| 52 |
How big of an extra administrative burden does that create? Maintaining our |
| 53 |
own version controlled portage tree might be a hard sell. Thanks for the |
| 54 |
script - I'll take a look at it. Is there any documentation out there about |
| 55 |
a static portage tree? |
| 56 |
|
| 57 |
-bryan |
| 58 |
|
| 59 |
P.S.: I checked out SiCortex at SC06, and talked to one of the guys there. |
| 60 |
Its definitely Gentoo. It sounds like they are a bunch of Gentoo |
| 61 |
enthusiasts, actually. |
| 62 |
|
| 63 |
-- |
| 64 |
gentoo-cluster@g.o mailing list |
Archives