Topic: [ANN] Spondoolies-Tech - carrier grade, data center ready mining rigs - page 340. (Read 1260363 times)

on a side note i was messing around on the net and this sounds interesting for you folks in hot areas

http://www.gogreensolar.com/pages/battery-for-backup-power-and-off-grid-energy

the key point is they accept bitcoin so you BTC whales out there can really take advantage of that

I'm on Eligius .. and 2.3.44 firmware. I put in place a port-forward setup for SSH and HTTP. The stratum goes out via NAT/MASQ.

If I understand the pools using stratum protocol right, the pool requires work restart if there are no shares posted back over a given amount of time and they reallocate the work to another worker. If you still submit results after a given time, they get rejected Therefore, the miner is notified by the pool to restart the work with the new work allocation.  
This may be due to bad luck (i.e. your machine could not find shares over the required pool difficulty for some time) or traffic shaping (if your DC does that to conserve bandwidth, and your share results do not make it in time back to the pool).

On another note, I found out that I can't use the "switchpool" command in cgminer API, on SP30. It was originally programmed in (i.e. "promotePool" JS function), but commented out. What happens, is that when the 2nd pool gets "promoted" in cgminer, the watchdog monitors connectivity with the first pool. It restarts cgminer, which goes back to the first pool after restart Funny. For some time I was browsing the code for the MinePeon donation code, which was originally switching pools for the configured amount of time you want to donate your hashes to MinePeon (default 0).  The watchdog though, will restart cgminer on the "0" pool as soon as it has connectivity to it. Therefore, the "donation" thingie does not work with the default setup even if you want to.

Ok, because on my miners with static IP I was not able to get them working... always got:

Stratum connection to pool 0 interrupted.
Lost X shares due to stratum disconnect on pool 0
Stratum from pool 0 requested work restart

I tried different pools ... as soon the machines were DHCP all fine. Which firmware are you running?
thx

I have static IP's on all of mine on Eligius with no issues.

Does the unit works well with static IP addresses? which pool are you using?

Thank you Soros!

You might be able to use a small nettop and configure it as a NAT/DHCP device and hang it in the side of the cabinet beside your SP30.

Yeah, I also use data centers meant for internet hosting and they give you a static IP block with a gateway IP address. This doesn't work well with Spondoolies and KnC default setups. However, I have space for a NAT/DHCP 1U device.

It's just an image I made to update the cgminer. No need to upgrade if your unit works fine for you - one will not get a better hash rate.

There is a limitation of 1250 on 208 volt because of PSU limitation. There is no calibration for 208V. To go more then 1250 one needs 220V.

We currently have about 230 TH/s mostly comprising SP30s on p2pool. They seem to work with p2pool just fine.

It is only the self-reporting that changed, the hash rate in pool should not have changed.

Can you overclock it a bit ? Some 20GH it's normal variance for 1.4TH, I might say.

I run my SP30 with a static IP (internal 10.10.0.100) because it was the one the unit leased for the first time and I liked it . DHCP is better though for hosted environments and multiple machines.
Moreover I am the one fixing things if something goes bad with the network here  .. one small 1u server in a corner rack drives the whole setup, I won't complain about space, usually you run out of power before you run out of space.
 

I went to 1.5.6 on my two SP10's and hashing speeds dropped from a solid 1.4TH to a 1.38TH.

Please open SSH HTTP ports to 1-2 of bad units and send me email with details how to connect.
P.S. - I disabled SW upgrade of SP10s for now just in case.

Blasphemy! Can't work on a holiday!

And by the way, anyone who bought an SP3x or SP2x you'll never break even. Sorry, I had to say it. I can't help myself. I'm an addict.

Hi guys.

Please, before opening a support call to handle low-rate or not-mining, open HTTP and SSH ports and add connect details to the request.
Without that I can not help you.

Regards

There is a unit in your hosting facility that provides DHCP addresses to hosted devices. I don't think your hosting facility just gives you your server IP and lets you choose any static address from range.

Working with static is possible, but history teaches that there are a lot of things that can go wrong with network configuration - so we do not provide support it because we don't have enough support power for that.

They always post this shit on the last day.  Be happy.  KnC delivers two month late.

Does anyone is running SP10/30 machine with static IP configuration?

I am wondering if I am the only person with this problem .. I am asking these because I am having hard time with several datacenters ....of course you could add an equipment to do the DHCP but it will use 1u of space and that will increase the hosting costs.

I lose patience, we are already on September 26. Each day passes my confidence drops at the sight of October  .