Re: Something similar to Soekris boards, for server applications

[Christopher LILJENSTOLPE]

Greetings Sime,


How many of these boards are you considering?  There are some 
wicked-dense
low-power cpu cluster solutions out there...

Chris
On 30Nov2011, at 08.27, Sime Ramov wrote:

> Hello, I am looking for something in the spirit of Soekris boards, but
> more suited for server applications, e.g. for hosting Django apps.
>
> Current net6501 is maxed out at 2 GB of RAM and 1.6 Ghz *single-core*
> (two threads) atom.
>
> The reason I am considering Soekris is because dedicated servers are
> often underused and idling. Few GB of memory, anemic processor and SSD
> gets one a surprisingly long way, especially with properly chosen stack
> and caching.
>
> So the general idea is: one Django app = one Soekris board. This is much
> better than virtualization (bare metal forever) or putting more apps on
> a big server.
>
> Some apps would run great on this, but a more powerful CPU and more
> memory would be needed for more demanding workloads.
>
> Any recommendations for similar, but a bit more powerful and versatile
> hardware (think one app = one hardware device)? Thanks.
>

--
ff/g?
Check my PGP key here: https://www.asgaard.org/~cdl/cdl.asc
Current vCard here: https://www.asgaard.org/~cdl/cdl.vcf

[demime 1.01d removed an attachment of type application/pgp-signature which had 
a name of signature.asc]

Re: mSATA failure on 6501 w/ OpenBSD 5.0

[Christopher LILJENSTOLPE]

Greetings guys,

I'm the original reporter, comments in line...

One other comment, when I got back to the console and rebooted, the entire
drive was scrodded.  I haven't seen an fsck like that in quite some time.
Unable to really come up as an operational system.  Basically re-pxe'd boot.rd
and re-installled.  I scribbled over the drive before the install without any
issues.  The drive checked.

On 29Nov2011, at 04.10, Remco wrote:

> Chris Cappuccio wrote:
>
>> here is the key error message. it means your whole ahci disk has
>> disappeared (and anything you can still run is happening from cache.)
>>
>> --
>> ahci0: stopping the port, softreset slot 31 was still active.
>> ahci0: failed to reset port during timeout handling, disabling it
>> --
>>
>> likely a reboot will fix it. this is a known problem with ahci driver and
>> intel ahci controllers.
>
> I am not so sure this is a driver problem.
>
> I think I accidentilly "emulated" this problem the other day on my desktop
> system (not a 6501):
> Nov 28 16:38:44 ws0001 /bsd: ahci1: stopping the port, softreset slot 31
was
> still active.
> Nov 28 16:38:44 ws0001 /bsd: ahci1: failed to reset port during timeout
> handling, disabling it
>
> I have this external drive bay connected through e-SATA. After unmounting
> the drive I switched off the external drive's power. Running disklabel on
> the drive resulted in the above failures, which I guess makes sense, after
> all, I made the drive "disappear".

The drive is a transcend 16 GB mSATA that's installed on the motherboard - not
really some way for it to "go away"

>
>>
>> the "failed to reset port" and "softreset slot was still active" problems
>> become really obvious once you start maxing out disks on an ahci
>> controller with a softraid array. they rarely present problems in normal
>> use! but, the SSD sata drive may evoke different behavior for some reason.
>> i think continuous runs of iogen over a RAID1 array might bring out
>> similar issues all by itself, even with regular hard disks
>>
>
> Maxing out disks sounds like having more activity on the disks, possibly
> making them draw more power. Could these errors relate to bad power cabling
> or insufficient power supply ?
>
> If multiple disks with an insufficiently powered system are the problem,
one
> solution might be a power supply that can deliver more power, another
> possible solution might be using external drive bays, each having their own
> power supply.
>
> For stuff purely SSD related, a motherboard BIOS update and/or SSD firmware
> update may help as well.

The power supply is running about 40% of max rated - and the drive is SSD a
small SSD, so I don't think that it could be the power supply.

Current BIOS on the 6501 (an update is coming, but current for right now).
SSD is brand new, and has the latest transcend firmware (as far as I can
tell).


Chris

>

--
ff/g?
Check my PGP key here: https://www.asgaard.org/~cdl/cdl.asc
Current vCard here: https://www.asgaard.org/~cdl/cdl.vcf

Re: [Soekris] Fwd: mSATA failure on 6501 w/ OpenBSD 5.0

[Christopher LILJENSTOLPE]

Greetings,

Any thoughts as to how to get around this - it's only been up for a few 
days.
Rebooting my home router every 24 hours is not spouse endearing behavior :)

Chris

On 28Nov2011, at 14.30, Chris Cappuccio wrote:

> here is the key error message. it means your whole ahci disk has disappeared
(and anything you can still run is happening from cache.)
>
> --
> ahci0: stopping the port, softreset slot 31 was still active.
> ahci0: failed to reset port during timeout handling, disabling it
> --
>
> likely a reboot will fix it. this is a known problem with ahci driver and
intel ahci controllers.
>
> the "failed to reset port" and "softreset slot was still active" problems
become really obvious once you start maxing out disks on an ahci controller
with a softraid array. they rarely present problems in normal use! but, the
SSD sata drive may evoke different behavior for some reason. i think
continuous runs of iogen over a RAID1 array might bring out similar issues all
by itself, even with regular hard disks
>
> dragonflybsd's port of openbsd's ahci driver has incorporated several of
workarounds for problems directly related to this. (reset this when that
happens, etc..) that might be a good place to start looking, if you can easily
reproduce the problem then you would know quickly when a ported fix from their
driver has helped.

--
ff/g?
Check my PGP key here: https://www.asgaard.org/~cdl/cdl.asc
Current vCard here: https://www.asgaard.org/~cdl/cdl.vcf