I can’t fix your pool

I received all of my purchased mining rigs and they are all happily hashing. It’s been a while since I’ve posted an update. The last time I posted an update I was waiting on hashing power. Orders took a bit longer to arrive and finalize than I’d planned. When rigs arrived at the data center there were a few technical issues. The technical issue I most remember is that the Zeus Lightning I added to my collection will not mine with Minera, so it’s running on the Hashra platform for now. But for some reason I cannot access the url the data center designated for that rig. Thankfully I have not needed to, as I can make all the pool switches I want via Mining Rig Rentals. Also, I have not needed to restart the miner, though that would be impossible without being able to access the platform.

The data center technician who specializes in mining rigs tried everything possible to get Minera to work, but the rig just doesn’t like Minera for some reason. So far I’ve just let my inability to access its Hashra platform go for now. I could also have the technician connect the rig to a simple Linux-based mining program like the one I originally used. This program is not particularly user friendly, but it does allow me to do all the basic things I would need to be able to do to control a rig remotely. I imagine this will become a top priority the first time the rig quits on me. It’s a pretty good rig, because it has not quit on me yet, and it has been weeks.

My other rigs are working out OK. I now have a fleet of three Furies and two Black Widows in addition to the Zeus Lightning. All of them can be rented on Mining Rig Rentals and as you can see they have been rented out numerously. My only complaint is that the price for Scrypt hashing power is ridiculously low. I have bills to pay at the data center and it’s always a bit dicey whether or not I will even break even each month, let alone ROI. As long as there is a profitable new scrypt coin I can mine each month I can do OK. The multipools won’t hack it, though. And it’s getting close to the rental fees not hacking it either, especially with the price of Bitcoin in the gutter (compared to how it was earlier this year).

Yes, I have complaints about the prices, but as far as the actual hashing goes, my rigs are top notch. You point them to a pool and they will mine it. At least that has been my experience with them. I’ve rarely set them on a pool that didn’t work.

Many of my renters, however, cannot say the same thing. As a rig owner I do not see the pools chosen by the renters. I believe the MRR platform is set up that way so that a miner who discovers an overlooked profitable pool to mine does not have to divulge his secret when he decides to rent some additional hashing power. This is great as long as the pool is truly a good one. However, apparently there are plenty of problematic pools out there. I have repeatedly experienced watching my rigs hashing away just fine, making a picture perfect hashing graph just above the advertised hash rate, only to completely fail as soon as a renter took it over. Pool failures are a common occurrence.

Sometimes I’ll get a rude message from the renter, often written in broken English and including some expletives. The renter assumes there is something wrong with the rig. The fact that the hashing graph was perfect right up until the time he rented means nothing to him. The rig was waiting to break down just for him. It’s a conspiracy.

MRR does not issue refunds for pool failures, so I don’t either. In the beginning I would compensate the renter by adding some time onto their rental period. Not anymore. Margins are just too tight. The ultimate pool failure is when my rig can no longer connect to the MRR servers and it starts mining in my backup multipool. If the renter is not paying attention I basically get to double dip, collecting both the entire rental fee and the proceeds from my backup multipool. If I catch it, I notify the renter that his pool failed and that he needs to choose another one. Usually they figure it out themselves… eventually. Amazingly enough, once the pool is switched, the rig goes back to hashing like a champ for the renter.

But there are other ways the pool can fail, leaving the rig connected to it, but doing nothing beneficial for anyone. The shares get rejected, all of them. The shares are mismatched (whatever that means). The shares are stale, and therefore useless.  Sometimes the worker (meaning my rig) gets banned outright by the pool. I’m not sure all the reasons for a pool to ban a worker, but it seems to happen most when the difficulty is set too low and one worker is submitting a disproportionate number of shares. Since most of the pools I mine in use Vardiff, I’ve never had that problem as I let the pool determine the difficulty level for each rig. Some renters manually set the difficulty to a very low value and then get themselves banned. Finally, I once encountered a renter who was mining in a pool which only allowed a set amount of hashing power per account. The renter complained that all his other rented rigs worked just fine, so I put in a support request asking what was going on. The reply message indicated it wasn’t my fault and that the pool was known to be problematic because of this hash power limitation. I think the renter eventually figured something out.

Usually when a renter’s pool has failed for any reason, they make the necessary adjustment and the rig gets back to hashing as expected. Occasionally, though, the renter feels compelled to remain in the failed pool. The most recent renter who did that claimed he was renting on behalf of a friend and that friend didn’t want to mine in any other pool. So the rig just stayed offline for the duration of his rental period. It was frustrating to me because the pool didn’t fail outright so I wasn’t able to mine either. Towards the end of the rental period the renter sent me a message urging me to enjoy the free Bitcoin. To which I retorted that it wasn’t free Bitcoin to me because my rig was still tied up the entire time it was stuck in his problematic pool.

The only type of pool failure I have personally experienced is the kind where the rig simply can’t maintain the expected hash rate due to a high percentage of rejected shares. This happened to me today, as a matter of fact. I was trying to mine Camorra Coin in the Supernova pool. My error/reject rate approached 50 percent. I am used to seeing a bunch of rejected shares at the very beginning when I first point a rig to a pool, but once the first block has been found that settles out and the reject rate stays well below five percent. But not this time. The pool had a note instructing us to switch to a different port if we were getting higher than expected reject rates. I switched to the different port. It seemed to work just fine for an hour or so. I gave up on the pool (at least for that particular rig) when the pool dashboard showed me that the Black Widow was only hashing at 2.6 MH/s.

Who knows why the reject rate was so high? Perhaps there is an issue with the coin’s code, or maybe there is something wrong with the pool. I don’t know and I don’t have time to trouble shoot it thoroughly because every minute that my rig is tied up in a pool that is causing it to underperform is a minute that I’m losing potential earnings. I’ll put up with it for a while to see if it will resolve itself (a lot of initial pool issues do), and I may try a few easy fixes, but at a certain point I give up and find a pool that will work properly. Or, sometimes my rig gets rescued by a new renter.

All those pool issues I described are issues which have nothing to do with the condition or performance of the rig itself, and as such are not anything that a rig owner can fix for a renter. If you ever rent from me and your pool doesn’t work, know that I can’t fix your pool. Mining Rig Rentals also knows this so they won’t refund your money if their records show that the problem was with the pool.

Occasionally a rig does go down, and occasionally a rig just needs to be restarted. I have taken requests from renters to restart rigs and I have also restarted them without being asked if it seemed that was needed. Although it takes less than three minutes to go through the restart sequence I’m usually happy to add six minutes to a renter’s time in those cases. Mining rigs are basically computers and as such they do need to be rebooted every once in a while. Usually I can see it coming when I can’t get my rig to maintain its advertised hash rate in a pool which normally works very well. One pool which can always be counted on to work is the Litecoin Pool. If I have any doubts about my rig’s hash rate I will switch it to the Litecoin Pool. If the hash rate immediately bounces back then I know the other pool was the problem. If the rig is sluggish even in the Litecoin Pool then I reboot it and that solves 99 percent of the problems. I recommend all renters register an account on the Litecoin Pool and set up a worker for their rented rigs. If the pool they chose doesn’t work, then switch to the Litecoin Pool for a few minutes and see if that solves the problem. I don’t recommend staying in the Litecoin Pool as the earnings will usually be less than the rental fee. But it’s handy for troubleshooting.

I can’t speak for all renters but I know that the only reason I ever rent hashing power is because there is a hot new scrypt coin I want to mine which I believe will make me a profit even after I’ve subtracted the rental cost. The multipools generally set the rental fees because no rig owner in his right mind is going to rent out for less than what he can make with the same rig mining a stable multipool. The main reason the scrypt rental rates have been so low lately is because the multipools have had low yields. In short, I wouldn’t rent hashing power just to mine in a multipool. I rent hashing power because I expect I will be able to sell the coin for a much better return than that.

Those hot new coins are of varying quality and their pools vary widely as well. Most of the pools I’ve mined in work just fine, but sometimes they don’t. When they don’t work, your best bet is to switch to another pool that does work. Usually a new coin will have more than one pool, so you can try one of the other ones. If it doesn’t, then you have to choose a different coin to mine instead. You need to have a backup plan. Before you rent, have your primary pool and backup pool chosen and make sure the backup pool is one you can feel good about mining. If you are not sure a pool will work, then rent a small amount of hashing power for the shortest possible time period and just see if it will work in the pool. Once you know it works then you can always rent more hashing power. MRR allows for a minimum rental time of three hours per rig. It is the minimum I have set on all my rigs including my Fury Fleet. It costs next to nothing to rent 3.8 MH/s for three hours, but that will be long enough to find out if the pool works. If my Fury Fleet is unavailable, you can always look for another rig with a small amount of hashing power. There are a few single Furies and even miners which hash at less than one MH/s which you can rent as a tester. A tiny bit spent on renting a small rig for three hours can save you tons of Bitcoins of rented hashing power later if it turns out the pool doesn’t work. Since I own rigs I always try mining a pool with one of my own rigs before I even consider renting.

If you are going to rent hashing power, please take some time to test the pool you plan to mine in before giving it all you’ve got. Either point one of your own rigs to it or rent a small rig for the minimum amount of time to test it. If the rig you are renting does not perform as advertised, you should first assume the problem lies with your pool, especially if the rig was hashing just fine before you rented it. You did take a look at the rig graph first, right? If you suspect the rig to be at fault, then point it to the Litecoin Pool for ten minutes and see if it does any better. Most of the time that my rigs aren’t working the issue is with the pool and that is something I can do nothing about as the rig owner. If you do suspect the rig is at fault and you’ve ruled out the pool as being the culprit, feel free to message me with your concerns and/or a request to reboot it. If the rig truly doesn’t work and I don’t respond–I rent to people in all time zones, and sometimes I am away from my computer–by all means put in a refund request with MRR. I also am more than willing to compensate for time if the rig itself went down for any reason. I can fix the rig, but I can’t fix the pool.

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *