It is not much, but I also added a new 741 to the pool yesterday.

It turns out that even though I have plenty of capacity left on my mining breakers, the new 741 (combined with everything else also on the subpanel) was just enough extra to trip the breaker feeding the entire subpanel after a few hours.  I adjusted down some older gear to compensate (I hope). 

You were not the only one.  When I looked into it, mine was reporting that the file system was mounted read only (and thus wouldn't save changes).  In my case, I had edited /etc/config/network (to reconfigure as DHCP), /etc/config/cgminer (for pools), and a couple other things on the card from my desktop before installing it into the Pi.  It ran perfectly but I couldn't save any changes because of the messed up mounting.  I've gotten away with those edits before (editing /etc/config/network for DHCP prior to first boot being especially useful), but I assume it was my own meddling that messed it up this time-- or maybe it is just coincidence and I also had a bad image.

My 7.8TH is back!  My A6s have been moved to the winter mining hole feeding the HVAC system.  I'm mining again in the evenings for waste heat when the temps are cooler outside.

Well y'all, I just shut down my A6s, which means I have no miners here at the moment.  They just are not profitable anymore on 0.09/kWh power.  I'll bring them back online for winter heat when it gets cold enough outside.  Until then I'm waiting for news on the A7.

That didn't take long after asking!    If I had know all it took was to ask then I would have done so much earlier!

That one would be from a rental.  I don't know the share diff since it was a rental and not my own hardware, but if I did the math correct on the block info (no guarantees on that), it was around 673G.

Now I'm asking for one from my own hardware 

Ok-- I'm here asking now too!  I'm well past when I should have found a block here!

I missed a pricing opportunity to "burst" my rentals to 2x the amount I usually rent in order to bring my N avg back up from the poor rental prices most of the previous days.  It isn't a total loss though since I did catch a little here and there and there yesterday and also had my own hardware.  My payouts will be about half of what they would have been.  The poor rental prices kept me from over-renting during the blockless time, so I was really looking forward to the luck aligning in my favor for a change.  It ended up being a wash-- I didn't catch the bad luck, but I mostly missed the good luck too thanks to the bot going down.  I sure would have liked catching the burst last night to get all the gains without the losses.  Rentals are very low margin (I'm at +3% over 3 months) so every missed opportunity is expensive.

It was nice to wake up to so many blocks.  

It was not nice to wake up to the the VM that runs my node and rental bot being nearly down and completely useless.  It was completely out of memory.  It was up but unusable, yet it was not so unusable as to set off my alarm during the night.  Although it has worked perfectly for months now, starting last night a grep command in a script I had monitoring the node started thinking that debug.log was a binary file instead of a text file (throwing a "Binary file (standard input) matches" error) and somehow that led to a memory leak down the line (I still don't know exactly how).  Rebooting and adding a -a switch to the grep command fixed it, but along with the node my rental bot was down much of the night.  

Then, just as I fixed that the power went out for about an hour!  When the power came back on none of my A6 would mine.  I had to reflash the SD cards and reconfigure to get each of the rPi to boot.

Crazy morning!

+1.1 = tlhIlwI

All of my hardware had failed over to solo mining and my rentals are showing as dead too.

Best share from a few days solo mining my SP20s was 801,248,735.  It was fun trying.  They are now profit switching between NiceHash and Kano.is until the rental rates calm down, then they go into retirement (likely sometime later today).

It was nice to see the pool hash rate stay stable through the halving.  My A6s and rentals are remaining here, but I will be turning off my SP20s later today.  I got 19 profitable months out of those!

I moved some of my SP20s from Kano.is to here for the last couple days before halving.  After 19 profitable months with them (far more than I ever expected), the SP20s will be powered down at halving.  The odds of actually hitting a solo block with them are near zero, but I may as well have a couple days of fun with them before their end!

My own notes on the same, from the perspective of the revision 2 A6 (I do not have any of the first model):

• Voltage- I do overvolt to 12.2V in the UI.  On batch 2 this means 12.15V PSU and provides about 3.9TH (this is a much lower voltage requirement than the batch 1 voltage).  Even during hot summer days I've had no difficulty with this overvolt.
• RPI Image- I'm running a 2016-02-05 image.  Any version is quite stable, but it can be noted that later images try to keep the temps to 68C; the original BlockC image targets 75C.  The later images that target 68C do result in some unusual fan ramping action that didn't happen with the original BlockC image, but it is easily fixed (see next point)
• Fan Speed- I've never tried the AVA4_OPTIONS for a fixed fan speed, but I do use min 40 in the Cgminer Configuration tab.  Using the default speed results in the miner getting stuck in a fan ramp up/down cycle and underperforming at only 3.2TH.  Min 40 fixes the issue.  This doesn't mean that faster is better though-- if the miner gets too cold then performance starts to drop (e.g. I tried min 70 once and was very disappointed with the 3.6TH result-- 40 seems to be the best setting for my case and provides the expected 3.9TH).  I'm still running the original A6 fans-- I have the Sanyo fans here but I haven't switched them out yet; I do not know if the Sanyo fans change this issue.

Wow!  Maybe I will get lucky on my over-renting (thanks to the new NH decrease restriction).  Those were some nice quick blocks!  

I've been trying to tweak my bot to get the rates down as fast as possible, but there can be dozens of steps down to below LRP.  I'm about to give up on keeping twice as many orders posted as desired and letting the bot switch between NH and WH sides for the cheapest price.  That worked really well in the past, but now it takes way too long to turn off orders when the whole lot of them fire up.  I may just give up on the best price switching thing and just post half of the orders on each side so that I don't over-rent and hope for the best (accepting that one side will go above my max rate and be turned off from time to time).

I now pass on the title of highest prize won in these picks (I hit the BTC0.3 one last year) to DanDan.

Congrats DanDan!

I too have been enjoying the 0.0024 rates.  0.0029 is as high as my bot will chase the price right now, so except for my own 12TH rig I'm out at the moment.  Today's blocks bring me to a 4% loss since starting up the rental bot.  I keep hoping to reach a small gain before June, but it isn't looking likely. 

+0.6 = tlhIlwI

I stayed away from rentals for months while occasionally mentioning the withholding risk, but I never saw any evidence of it happening.  Despite the amount of rental activity this pool was getting, our long term averages just didn't show an issue.  The math made/makes the risk from rental withholding low (e.g. if 10% of the pool were rentals and 10% of the rental service was bad hash, then the impact on the pool would be 1% -- I'm talking typical rental activity here, not the block busting bursts).  With no evidence of withholding actually happening, the 1% risk wasn't too bad.  So, I gave (and still am) rentals a shot.  At the best point I was up over 1BTC.  The bad luck came and I kept renting steady through the bad luck.  I lost 2BTC and found myself down 1BTC.  Now, just as I'm beginning to recover (maybe) with an upturn in luck, is when a campaign starts to ban rentals from the pool...

Don't get me wrong, if rentals are provably causing the pool harm then something should be done about them (myself included), but let's wait until Kano looks through the data before jumping to (and acting on) conclusions. 

Regardless, a lot of the rental problem may go away on its own on 6/1 anyway.  NH is switching to PH units but keeping the 0.0001 scale (so, to decrease what is now a single step from 0.0026 to 0.0025 will require going 2.6000, 2.5999, 2.5998, etc. one step at a time with minutes of delay between each tiny step).  That is probably going to make it impossible to get the rental price back down to anything decent whenever the mysterious highly overpriced infinity hash rate orders flash up for an hour and go away.  Anyone who runs a bot to maintain a steady N avg (like myself) is going to find this a nuisance-- possibly enough to give up.

We're now past 500% (415 plus the 87.7 orphan).  Not nice!