We're under a tight DoS attack. Been analyzing it, but not sure what the cause is yet.
The cause is you dumbass,
Lets see how you handle this pool from now on.
Topic: [OLD] Eligius: ASIC, no registration, no fee CPPSRB BTC + 105% PPS NMC, 877 # - page 157. (Read 458370 times)
January 18, 2012, 08:22:36 PM
The cause is you dumbass,
Lets see how you handle this pool from now on.
January 18, 2012, 08:20:52 PM
I guess this is the new type of "dos attack"
Under the old dos attack, if speed drops, then there should be some share submitting problems with cgminer,
and cgminer will automatically switch to other pools.
This new "dos attack" does not. It EAT shares, Strange
Any shares accepted should be represented as usual. If that's not the case, please double check your miner isn't dropping the shares (eg, with Wireshark) and get in touch with me about it in our IRC channel. Under the old dos attack, if speed drops, then there should be some share submitting problems with cgminer,
and cgminer will automatically switch to other pools.
This new "dos attack" does not. It EAT shares, Strange
January 18, 2012, 07:51:31 PM
I guess this is the new type of "dos attack"
Under the old dos attack, if speed drops, then there should be some share submitting problems with cgminer,
and cgminer will automatically switch to other pools.
This new "dos attack" does not. It EAT shares, Strange
Under the old dos attack, if speed drops, then there should be some share submitting problems with cgminer,
and cgminer will automatically switch to other pools.
This new "dos attack" does not. It EAT shares, Strange
January 18, 2012, 03:25:47 PM
We're under a tight DoS attack. Been analyzing it, but not sure what the cause is yet.
January 18, 2012, 02:30:16 PM
The hashing speed drops, but cgminer shows there's no problem submitting shares, and cgminer don't switch to other pool.
Anyone noticed this?
Anyone noticed this?
Yes me, usually I mine at 700M/s but now I'm at 400M/s. No problem in the logs.
January 18, 2012, 01:55:08 PM
The hashing speed drops, but cgminer shows there's no problem submitting shares, and cgminer don't switch to other pool.
Anyone noticed this?
Anyone noticed this?
January 17, 2012, 11:36:38 PM
I'm more concerned about the current state of "buffer empty". You mentioned that it is acting as if "proportional" now and the per-share price varies. How is this affected by pool-hopping? Since the share price is varying would this make the pool a target now?
The share price isn't itself varying; the currency is. Since there aren't enough Bitcoins to go around, the pool issues its own fiat "currency" called "extra credit"; this has no value itself, but the pool tries to convert it to Bitcoins every block until it's gone.Specific to pool hopping, the whole premise of hopping is being able to get more than the fair PPS price for shares. This is never possible with SMPPS.
Also, I'd like to note that we're gaining and will probably have a buffer again pretty soon. However, the law of large numbers means that in the long run, the buffer should get closer and closer to zero.
January 17, 2012, 11:22:40 PM
I'm more concerned about the current state of "buffer empty". You mentioned that it is acting as if "proportional" now and the per-share price varies. How is this affected by pool-hopping? Since the share price is varying would this make the pool a target now?
January 17, 2012, 09:13:29 PM
While I can clean out old shares from the core database, the webserver still requires a full history of shares for active servers.
That sounds like a pretty bad design flaw. So I have to assume it's more a case of using a product to perform a task it was never designed to do or even using it in a manner never envisioned by the developer. January 17, 2012, 08:57:45 PM
While I can clean out old shares from the core database, the webserver still requires a full history of shares for active servers.
That sounds like a pretty bad design flaw. So I have to assume it's more a case of using a product to perform a task it was never designed to do or even using it in a manner never envisioned by the developer.
The buffer was exhausted. Due to timing, I ended up with 0.20 BTC leftover. I don't consider this amount worth the time trying to figure out how to split it up.
Enjoy your $1.32
January 17, 2012, 08:48:11 PM
Ok, thanks for the answers.
January 17, 2012, 06:04:57 PM
Why did you need to switch over?
While I can clean out old shares from the core database, the webserver still requires a full history of shares for active servers. This copy of the shares had grown to be 400 GB, and disk space doesn't grow forever. I figured it prudent to start over while we had the opportunity.What is the buffer? The SMPPS buffer?
Yes, that buffer.Now that we've switched to Ra, did that happen because people moved to 'generous' and was the buffer donated?
The buffer was exhausted. Due to timing, I ended up with 0.20 BTC leftover. I don't consider this amount worth the time trying to figure out how to split it up. January 17, 2012, 05:58:22 PM
Eligius-Ra, the new server, is online the same host and everything, ready for the moment we get to a switchover point. However, Eligius-Su has been finding short blocks, taking it further away. If miners wish to donate the rest of the buffer to me, they can switch their miners to use generous.mining.eligius.st to begin using Eligius-Ra early. When/if Eligius-Su's hashrate drops off to under 100 GH or so due to people moving to Eligius-Ra, I will take that as a vote to donate the buffer, and shutdown Eligius-Su, replacing it with Eligius-Ra. Either way, don't forget to update your Artefact2 stats links to use /7/ instead of /5/ 
Can you explain a bit more what this is about? Why did you need to switch over? What is the buffer? The SMPPS buffer? Now that we've switched to Ra, did that happen because people moved to 'generous' and was the buffer donated?
January 16, 2012, 07:48:57 AM
So why is the reward per share bouncing all over the place now? Am I mistaken in thinking it's supposed to be a fixed rate per share?
http://eligius.st/wiki/index.php/Shared_Maximum_PPSLong story short, if the pool doesn't have enough money for PPS, it's basically proportional, but will try to make up the difference in the future short blocks. Long-run, it will average out to normal PPS prices.
January 16, 2012, 05:18:20 AM
So why is the reward per share bouncing all over the place now? Am I mistaken in thinking it's supposed to be a fixed rate per share?
January 13, 2012, 07:21:57 PM
Eligius-Ra, the new server, is online the same host and everything, ready for the moment we get to a switchover point. However, Eligius-Su has been finding short blocks, taking it further away. If miners wish to donate the rest of the buffer to me, they can switch their miners to use generous.mining.eligius.st to begin using Eligius-Ra early. When/if Eligius-Su's hashrate drops off to under 100 GH or so due to people moving to Eligius-Ra, I will take that as a vote to donate the buffer, and shutdown Eligius-Su, replacing it with Eligius-Ra. Either way, don't forget to update your Artefact2 stats links to use /7/ instead of /5/
January 09, 2012, 10:30:29 AM
That means they had no value to begin with. If BTC dropped to $0 every time I sold some, I wouldn't bother with BTC to begin with.
Sure it would drop to zero if you obtained all the coins from the profits of having a Massive percentage of the mining power, and dumped them all at once.Yeah, but whats the difference between him doing it all at once or all of us doing it all at once? Do we even know how much he got? If it was less than 10BTC, hell, that just paid for the server for a month or two.
January 09, 2012, 10:28:50 AM
That means they had no value to begin with. If BTC dropped to $0 every time I sold some, I wouldn't bother with BTC to begin with.
Sure it would drop to zero if you obtained all the coins from the profits of having a Massive percentage of the mining power, and dumped them all at once. January 09, 2012, 10:26:58 AM
But not to the extent that is being claimed. The way merged mining works is when I give 400 mhash to the pool, that 400 mhash goes to Bitcoin AND namecoin AND whatever else without diminishing the original 400 that went to Bitcoin. He can merge mine 9000 other alt chains, and I still get my 400 mhash worth of Bitcoin shares.
But are you getting your 400mhash worth of altcoins? No, he is keeping them. Lame.
That implies said altcoins actually have value, of course. He could setup signed message shit for the rest of the alt chains just like he did for nmc, but the other chains simply have no value.
That means they had no value to begin with. If BTC dropped to $0 every time I sold some, I wouldn't bother with BTC to begin with.
January 09, 2012, 10:23:43 AM
But not to the extent that is being claimed. The way merged mining works is when I give 400 mhash to the pool, that 400 mhash goes to Bitcoin AND namecoin AND whatever else without diminishing the original 400 that went to Bitcoin. He can merge mine 9000 other alt chains, and I still get my 400 mhash worth of Bitcoin shares.
But are you getting your 400mhash worth of altcoins? No, he is keeping them. Lame.
That implies said altcoins actually have value, of course. He could setup signed message shit for the rest of the alt chains just like he did for nmc, but the other chains simply have no value.
Jump to: