Ronald Pottol's questions

So we have some dell blades and chassis (blades are M600's, chassis M1000's) and other systems (R710 with MD3000 array). The R710 exports a source tree via nfs for the blades to build and test with.

The problem is the blades loose the nfs mounts. Blades in the same chassis, with what seem like identical configurations, have their connections hang, they cannot even ping the server. They eventually come back.

It is mostly Dell, in fact, we have a cable running from the r710 to a switch in one of the chassis, and another to a switch and from there to the chassis, both can have issues.

We are running Centos5 or Fedora Core release 5 (Bordeaux). The nfs server is running CentOS release 5.4 (Final).

Any thoughts? troubleshooting tips?

These are all to the same host, but via different routes:

Through a switch

[root@b053 ~]# ping svnwatch-data
PING storage.rack1.rinera.int (10.1.1.54) 56(84) bytes of data.

--- storage.rack1.rinera.int ping statistics ---
9 packets transmitted, 0 received, 100% packet loss, time 7999ms

Routed through another host:

[root@b053 ~]# ping svnwatch-data2
PING storage2.rack1.rinera.int (172.16.100.25) 56(84) bytes of data.
64 bytes from 172.16.100.25: icmp_seq=1 ttl=64 time=0.260 ms
64 bytes from 172.16.100.25: icmp_seq=2 ttl=64 time=0.217 ms
64 bytes from 172.16.100.25: icmp_seq=3 ttl=64 time=0.201 ms
64 bytes from 172.16.100.25: icmp_seq=4 ttl=64 time=0.264 ms

--- storage2.rack1.rinera.int ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 2999ms
rtt min/avg/max/mdev = 0.201/0.235/0.264/0.031 ms

With the host connected to a different chassis's switch (they are daisy chained)

[root@b053 ~]# ping svnwatch-data-eth2
PING svnwatch-data-eth2.rack1.rinera.int (10.1.1.56) 56(84) bytes of data.
64 bytes from 10.1.1.56: icmp_seq=1 ttl=64 time=0.598 ms
64 bytes from 10.1.1.56: icmp_seq=2 ttl=64 time=0.096 ms
64 bytes from 10.1.1.56: icmp_seq=3 ttl=64 time=0.168 ms

--- svnwatch-data-eth2.rack1.rinera.int ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2000ms
rtt min/avg/max/mdev = 0.096/0.287/0.598/0.222 ms
[root@b053 ~]#

They are a mix of Red Hat and CentOS 5.x, I'd like to have something on one host that can check basics like cpu/mem/disk hardware, Model, HW RAID controllers, etc would be nice.

I have key based ssh authentication set up, but I do not want to have to install anything on these hosts. Worst case, I could scp something to /tmp, but no install please.

Near as I can see, either I need to install something, or it is something really basic that I would run on the machine, and then manually aggregate.

Am I stuck writing some Perl to run and process the output of dmidecode, lspci, etc?

If I do, I'll put it up someplace, and link to it.

Related questions that do not have the answer:
Get server hardware details for ubuntu server
How to find out details about hardware on the Linux machine? this one had a nice bit of shell script, but it errors.

I am tired of worrying about data loss at home. My wife is a semi pro photographer, and essentially all of our family memories are digital (and we ought to convert the ones that are not). I am planning on setting up two systems to host the disks, and run some sort of cluster file system (and then back that up to external drives, which rotate off site weekly) via iSCSI on the disks in the two systems. I run Linux, and need to support Apple OS X, and a little Windows (Windows can be ignored, as it will be used for either VMware experimentation with its own LUNs, or it can live with SMB). So it would be great if there was a Mac native client, but I can live with NFS.

The environment will also have a Mac laptop, a Linux laptop, and a large desktop, which will come and go as far as being attached to the main file system, but they will only be users, they will not contribute storage.

Hardware wise, I am fairly open, low power would be good, as these will likely be on 24/7 (unless I set up my Linksys router to send a wake on LAN packet when someone looks for them?), looking for probably a dual core Atom with an nVidia chipset, so I can have 4GB of RAM, with 2-4 SATA ports, and dual gigabit Ethernet. I plan on buying 5 disks to start (gotta figure one will die, so getting the spare, which will be tested of course, first), and it would be good to have a port (and drive slot in the case) or two free. I will have a desktop and two reasonably beefy laptops for playing with VMs and other intensive work, so I don't think a desktop CPU makes sense (tempting though). Being able to boot from a CF card would be nice.

Rather not deal with proprietary drivers, but as these will be headless (probably), it will not be an issue. I have had issues with ATI drivers causing issues with user switching and powersave/suspend (do you want Planet Penguin Racer to fly, or to not have to log out to let you wife or kids use the computer?).

I would like to reuse cases and power supplies I already have, but they are ATX (old, but not so old as to be AT), will there be issues with small MB (cases I worry about, I expect the MB will take a standard 24 pin ATX plug)?

Hardware I am considering:
ZOTAC IONITX-F-E Atom N330 1.6GHz Dual-Core Mini ITX Intel Motherboard - has 16x pci express
ZOTAC IONITX-A-U Atom N330 1.6GHz Dual-Core Mini ITX Intel Motherboard - comes with powersupply
Asus AT3N7A-I Intel Atom 330/Nvidia ION/ Mini ITX Motherboard - has a PCI slot

All have just one gigabit Ethernet port. I'd prefer the one with its own PS, the Zotac's at least have WiFi (gives some redundancy, with horrible performance).

Summary: FS that might work: Gluster seems good (anyone used it personally? How about laptops that come and go?), pNFS might work too, anything else THAT WILL JUST WORK ONCE IT IS SET UP? (to similar questions, too many people posting links to stuff that is a long way from production ready, or that doesn't have any kind of fault tolerance.)

No: Luster really wants a BIG install. xtreemfs - no redundancy, pvfs - I just don't like stuff that sits on your existing fs, GFS2 and OCFS2 - a pain from what people say here.

cool, but not there - POHMELFS, CEPH, CRFS,