@Luke-Jr
I see very high reject-rates with my ztex board on your pool.
Ztex-Quad (BTCMiner or cgminer): 4.53%
GPU (cgminer): 1.8%
To compare:
Deepbit:
Ztex-Quad: 0.18%
GPU: 0.21%
ABCPool:
Ztex-Quad: 0.22%
GPU: 0.29%
Any idea why?
Topic: [OLD] Eligius: ASIC, no registration, no fee CPPSRB BTC + 105% PPS NMC, 877 # - page 151. (Read 458255 times)
May 09, 2012, 05:22:42 AM
May 06, 2012, 10:11:41 AM
Eloipool locked up while I was AFK and nobody bothered to call - I guess with continuous uptime for 2 months we all just assumed it was stable 
What happened in detail was, epoll returned that a socket was writable, but then actually trying to write to it blocked. This was a (miner-side) getmemorypool request, so I'm guessing maybe Python blocks until all the data is written, rather than just trying a single send(). To fix it, I set all the connection sockets to non-blocking, so they should never block under any circumstances.
While I was in that code, I also added an attempt to write immediately when Eloipool decides to send data (rather than waiting to check if the socket is readable). There's a chance this may improve longpoll response times slightly.
What happened in detail was, epoll returned that a socket was writable, but then actually trying to write to it blocked. This was a (miner-side) getmemorypool request, so I'm guessing maybe Python blocks until all the data is written, rather than just trying a single send(). To fix it, I set all the connection sockets to non-blocking, so they should never block under any circumstances.
While I was in that code, I also added an attempt to write immediately when Eloipool decides to send data (rather than waiting to check if the socket is readable). There's a chance this may improve longpoll response times slightly.
My stales were slightly higher after the patch (0.6~ - 1.41~) though. I'm not sure if it's related to this but it occurred after the patch.
May 06, 2012, 10:06:48 AM
Eloipool locked up while I was AFK and nobody bothered to call - I guess with continuous uptime for 2 months we all just assumed it was stable
What happened in detail was, epoll returned that a socket was writable, but then actually trying to write to it blocked. This was a (miner-side) getmemorypool request, so I'm guessing maybe Python blocks until all the data is written, rather than just trying a single send(). To fix it, I set all the connection sockets to non-blocking, so they should never block under any circumstances.
While I was in that code, I also added an attempt to write immediately when Eloipool decides to send data (rather than waiting to check if the socket is readable). There's a chance this may improve longpoll response times slightly.
What happened in detail was, epoll returned that a socket was writable, but then actually trying to write to it blocked. This was a (miner-side) getmemorypool request, so I'm guessing maybe Python blocks until all the data is written, rather than just trying a single send(). To fix it, I set all the connection sockets to non-blocking, so they should never block under any circumstances.
While I was in that code, I also added an attempt to write immediately when Eloipool decides to send data (rather than waiting to check if the socket is readable). There's a chance this may improve longpoll response times slightly.
May 05, 2012, 11:48:53 PM
Pool down?
Status Address Latency Uptime (API)
Network problem (timeout, unreachable) ra.mining.eligius.st:8337 N/A Downtime : 8 minutes
Status Address Latency Uptime (API)
Network problem (timeout, unreachable) ra.mining.eligius.st:8337 N/A Downtime : 8 minutes
April 17, 2012, 10:11:36 AM
Hi Luke-Jr,
something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
Everything looks fine to me...something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
Hi,
once again. Still having problems with Eligius pool.
If I jump in (with 5GH) stats are not going up. Not after 15min, 30min... The 15min stat remains 0 or at maximux 50MH with 10 or 20 shares submitted.
Therefore I started mining with a new bitcoin address. I've submitted around 1000 shares. One day later the address is still unknow. The bitcoin address is correct. Just send 0.01 BTC to it from another PPS pool.
So whats going on... for me nothing looks fine.
Here's the link to Eligius with the new bitcoin address: http://eligius.st/~artefact2/7/1P7jUfXUEgk4oqXXxsQ2RAJsa9uBp1ChSA
Fixed and working
Thanks a lot. April 17, 2012, 09:28:50 AM
Hi Luke-Jr,
something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
Everything looks fine to me...something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
Hi,
once again. Still having problems with Eligius pool.
If I jump in (with 5GH) stats are not going up. Not after 15min, 30min... The 15min stat remains 0 or at maximux 50MH with 10 or 20 shares submitted.
Therefore I started mining with a new bitcoin address. I've submitted around 1000 shares. One day later the address is still unknow. The bitcoin address is correct. Just send 0.01 BTC to it from another PPS pool.
So whats going on... for me nothing looks fine.
Here's the link to Eligius with the new bitcoin address: http://eligius.st/~artefact2/7/1P7jUfXUEgk4oqXXxsQ2RAJsa9uBp1ChSA
April 17, 2012, 09:24:42 AM
Hi Luke-Jr,
something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
Everything looks fine to me...something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
Hi,
once again. Still having problems with Eligius pool.
If I jump in (with 5GH) stats are not going up. Not after 15min, 30min... The 15min stat remains 0 or at maximux 50MH with 10 or 20 shares submitted.
Therefore I started mining with a new bitcoin address. I've submitted around 1000 shares. One day later the address is still unknow. The bitcoin address is correct. Just send 0.01 BTC to it from another PPS pool.
So whats going on... for me nothing looks fine.
Here's the link to Eligius with the new bitcoin address: http://eligius.st/~artefact2/7/1P7jUfXUEgk4oqXXxsQ2RAJsa9uBp1ChSA
April 12, 2012, 05:41:57 PM
How much does the pool owe its miners then, and what do you call that?
The payout queue can be viewed at http://eligius.st/~twmz/Thank you, Luke-jr.
April 12, 2012, 05:36:44 PM
If 1 bitcent and 1 anti-bitcent are destroyed in mutual annihilation how much energy is produced?
That's how MintChipCents are produced! Just add blood, sweat and tears to create even more! 
April 12, 2012, 05:17:28 PM
I'm surprised I'm the only one who wants to know, but what caused the charts to mess up yesterday?
Due to a minor glitch in the code I use to do manual payouts (sendmany), some addresses were over-paid. The safeguards in place in the coinbaser code kicked in and complained that those addresses' balances were taken negative, and refused to run. To get the coinbaser running again, I had to make a temporary copy of the database, and run it without that safeguard in place, updating the cache beyond the point where they were taken negative. However, when doing so, the updates created the JSON block statistics in the copied directory. Since this was missing, Artefact2's web graphs were missing the "total ever paid" information, and put the estimates at the zero-level instead. I moved the relevant file, so it now is drawing current data correctly, but unfortunately there is no easy way to modify the historical record it keeps. After that ~hour falls off the left side of the history, it should return completely to normal.How much does the pool owe its miners then, and what do you call that?
The payout queue can be viewed at http://eligius.st/~twmz/ April 12, 2012, 05:12:56 PM
Are we talking about the same thing? The SMPPS buffer? Which can be negative, and unless you're using a different definition of 'balance' is as likely to be negative as positive?
Yes, and as the creator of SMPPS, I know that the buffer in fact cannot be negative by definition... Either there are funds in it, or not. There can't be anti-funds in the pool's wallet any more than there can be anti-funds in your own. April 12, 2012, 04:13:46 PM
I'm surprised I'm the only one who wants to know, but what caused the charts to mess up yesterday? I happened to be looking when the lines were down really low (perhaps because all previous payout information was missing?), and the charts pretty much didn't show anything. Even now it's hard to see detail on them and presumably will be for the rest of a week until the messed up part scrolls off the left side. I'm just as curious regarding whether or not data is missing and/or accurate as I am regarding whether or not the charts will be fixed/normal before that part scrolls off.
ETA: I forgot the charts could be zoomed on. I'm now far more concerned about the integrity of the stats than the issue in the charts. Not that I'm even that concerned about the stats (mostly curious).
ETA: I forgot the charts could be zoomed on. I'm now far more concerned about the integrity of the stats than the issue in the charts. Not that I'm even that concerned about the stats (mostly curious).
April 12, 2012, 04:02:50 PM
If 1 bitcent and 1 anti-bitcent are destroyed in mutual annihilation how much energy is produced?
April 12, 2012, 11:11:37 AM
Are we talking about the same thing? The SMPPS buffer? Which can be negative, and unless you're using a different definition of 'balance' is as likely to be negative as positive?
Yes, and as the creator of SMPPS, I know that the buffer in fact cannot be negative by definition... Either there are funds in it, or not. There can't be anti-funds in the pool's wallet any more than there can be anti-funds in your own. April 12, 2012, 10:52:00 AM
Are we talking about the same thing? The SMPPS buffer? Which can be negative, and unless you're using a different definition of 'balance' is as likely to be negative as positive?
You should talk to my bank manager some day. He has much to say on the topic
How do you have a negative balance?
You should talk to my bank manager some day. He has much to say on the topic
April 12, 2012, 10:47:39 AM
Luke-jr, shouldn't the pool be in a negative buffer at the moment? Your stats page is showing a positive buffer. Come to think of it, it shows that you've never had a negative buffer. Bug?
There is no such thing as a negative buffer. How do you have a negative balance? April 12, 2012, 10:43:20 AM
Luke-jr, shouldn't the pool be in a negative buffer at the moment? Your stats page is showing a positive buffer. Come to think of it, it shows that you've never had a negative buffer. Bug?
April 06, 2012, 12:01:02 PM
FWIW, I just published an old copy of Eligius's coinbaser script.
April 05, 2012, 02:43:23 PM
Hi Luke-Jr,
something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
Everything looks fine to me... something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
April 05, 2012, 05:02:21 AM
Hi Luke-Jr,
something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
something is wrong with the stats at Eligius. Since Apr 2 it's completely wired.
Also the "submitted shares" count looks wrong. Does Eligius loose shares, or it's just a problem with the stats?
Jump to: