Author

Topic: OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.0 - page 434. (Read 5805983 times)

legendary
Activity: 952
Merit: 1000
Ya my first several shares are usually rejects. I am mining over wifi, with about a dozen+ other computers on the network, so DCs are a fact of life. Still, I can live with 0.3% for now. Wink
Don't go further. This is the issue.
I'm not complaining. I know it's a problem, but one I sorta have to live with. Tongue
sr. member
Activity: 383
Merit: 250
Yep I'm on CGMiner 2.9.5, with stratum on Ozcoin. I set the diff to 2 with 1.5GH/s, and today I'm just under 0.3% rejects, which is actually pretty high. I'm usually around 0.1-0.2%. Cheesy
Those rejects are 99% likely to be either during connect while the difficulty corrects itself (directly related to the problem being discussed) or due to reconnects (i.e. if you lose pool connection)

The fix for the 2nd one (reconnects) is simply an enhancement ... that no doubt there will be few arguments about (when it gets in vogue Tongue) since it is just an enhancement - no compatibility issue with the current implementation.
Ya my first several shares are usually rejects. I am mining over wifi, with about a dozen+ other computers on the network, so DCs are a fact of life. Still, I can live with 0.3% for now. Wink

LOL @ WIFI
hero member
Activity: 938
Merit: 501
Yep I'm on CGMiner 2.9.5, with stratum on Ozcoin. I set the diff to 2 with 1.5GH/s, and today I'm just under 0.3% rejects, which is actually pretty high. I'm usually around 0.1-0.2%. Cheesy
Those rejects are 99% likely to be either during connect while the difficulty corrects itself (directly related to the problem being discussed) or due to reconnects (i.e. if you lose pool connection)

The fix for the 2nd one (reconnects) is simply an enhancement ... that no doubt there will be few arguments about (when it gets in vogue Tongue) since it is just an enhancement - no compatibility issue with the current implementation.
Ya my first several shares are usually rejects. I am mining over wifi, with about a dozen+ other computers on the network, so DCs are a fact of life. Still, I can live with 0.3% for now. Wink

Don't go further. This is the issue.
legendary
Activity: 952
Merit: 1000
Yep I'm on CGMiner 2.9.5, with stratum on Ozcoin. I set the diff to 2 with 1.5GH/s, and today I'm just under 0.3% rejects, which is actually pretty high. I'm usually around 0.1-0.2%. Cheesy
Those rejects are 99% likely to be either during connect while the difficulty corrects itself (directly related to the problem being discussed) or due to reconnects (i.e. if you lose pool connection)

The fix for the 2nd one (reconnects) is simply an enhancement ... that no doubt there will be few arguments about (when it gets in vogue Tongue) since it is just an enhancement - no compatibility issue with the current implementation.
Ya my first several shares are usually rejects. I am mining over wifi, with about a dozen+ other computers on the network, so DCs are a fact of life. Still, I can live with 0.3% for now. Wink
legendary
Activity: 4634
Merit: 1851
Linux since 1997 RedHat 4
Yep I'm on CGMiner 2.9.5, with stratum on Ozcoin. I set the diff to 2 with 1.5GH/s, and today I'm just under 0.3% rejects, which is actually pretty high. I'm usually around 0.1-0.2%. Cheesy
Those rejects are 99% likely to be either during connect while the difficulty corrects itself (directly related to the problem being discussed) or due to reconnects (i.e. if you lose pool connection)

The fix for the 2nd one (reconnects) is simply an enhancement ... that no doubt there will be few arguments about (when it gets in vogue Tongue) since it is just an enhancement - no compatibility issue with the current implementation.
legendary
Activity: 952
Merit: 1000
Yep I'm on CGMiner 2.9.5, with stratum on Ozcoin. I set the diff to 2 with 1.5GH/s, and today I'm just under 0.3% rejects, which is actually pretty high. I'm usually around 0.1-0.2%. Cheesy
legendary
Activity: 4634
Merit: 1851
Linux since 1997 RedHat 4
Please just ignore Luke-Jr.
P.S. Kano is simply a liar, as usual.
I feel like I've read this thread before. Maybe I'm psychic?!

Either way, it's been rough trying to keep up with all of you on the work being done to stratum and CGMiner, but keep it up!
Stratum and CGMiner is fine.
The minor issues with Stratum are that shares can be lost.

However, that number is already WAY lower than using GetWork or Moron's piece of shit.
Yes it is important to understand that.

I've been going on about the Stratum issue for quite a while.
(ckolivas actually mentioned it to them first, but I didn't know that until shortly after I brought it up)

I'd like to get no Rejects per day.
At the moment with Stratum on OzCoin with a fixed 8 difficulty and ignoring reconnects (that I force each day at midnight) I get 1 reject each day with 1.6GH/s
Yes I'm bitching about getting 1 reject each day Tongue
I want 0 Smiley

This I guess is part of the reason why slush has ignored my request ... but in the end he's trying to implement a fix on the mistake he's made rather than just simply fixing it and ignoring the so called necessary backwards compatibility Sad
It's not been around long enough to care about backwards compatibility with a mistake.
legendary
Activity: 952
Merit: 1000
Please just ignore Luke-Jr.
P.S. Kano is simply a liar, as usual.
I feel like I've read this thread before. Maybe I'm psychic?!

Either way, it's been rough trying to keep up with all of you on the work being done to stratum and CGMiner, but keep it up!
legendary
Activity: 2576
Merit: 1186
ok so
96+ 23 = 119
114.8+5.2=119
Was my point. Maybe it's the maths...
You're not supposed to stop at 5.2 seconds. Stratum jobs can be used infinitely, until the block is invalid or the pool expires the job. Unlike getwork, there is no 232 nonce limit.

This left me wondering why you said 120 seconds to start with and only had math to make 119. Possibly to account for round trip delays or the share validation. You didn't really say. So either your math was off, or you left out part of your math.
I subtracted a second for the time between the pool sending the job and you receiving it.

P.S. Kano is simply a liar, as usual.
legendary
Activity: 4634
Merit: 1851
Linux since 1997 RedHat 4
Throwing out work with every notify message on stratum, which comes on average every 30 seconds, would have the same effect as a longpoll every 30 seconds. That's a lot of work to discard midstream. Why on earth would there even be a clean_work method with stratum if every notify implied we throw out the work?
You're not supposed to throw out the shares on LPs. Restarting work has no cost except on BFLs (where you have a bug that loses valid shares).

So Junior is saying my reject ~9minutes into the block is because CGMiner didn't change to new work fast enough?
That sounds highly unlikely, and unrelated. What kind of a reject?

Top post on the page you are asking on. LOOK UP!
Ok, I missed that somehow. "unknown-work" can only occur if one of these 3 cases:
  • a block was found (which you ruled out)
  • the miner submitted it after it expired (120 seconds on EMC; new work is delivered at least every 96 seconds)
  • the user extranonce1 was lost (only possible if you reconnected)
I'm pretty sure cgminer discards all work/shares in the last case (reconnection), so that means it either:
  • was working on a job 23 seconds after it had been replaced (this is sufficient for as low as 200 Mh/s)
  • was working on a job 21 seconds after it had been replaced AND took over 2 seconds to submit it to the pool
This makes sense, since cgminer is failing to move on to new jobs as they come in.

So a BFL single running 864mhash bitstream in your example would need to work on the work for 114.8 seconds to go over. Interesting to know.
I didn't change scan time for stratum but the settings of scan time at 115 had worked fine on get work.

I will bump it down but since not every 120 seconds has this happening I would assume it is something else. The ping to Eclipse is 76ms right now. It can vary but assuming I got my work at 96 seconds add .5 for significant delays, 5.16s for single to process work I get 101.66 seconds. Leaving me 18.34 seconds to submit.....
Please just ignore Luke-Jr.
Most of what he says is FUD and even if it's not all FUD - having to waste time wading through it all is just that - a waste of time.
His answer will simply be - you must use his miner coz it does it properly ... lulz
hero member
Activity: 981
Merit: 500
DIV - Your "Virtual Life" Secured and Decentralize
ok so
96+ 23 = 119
114.8+5.2=119
Was my point. Maybe it's the maths...

Looking at it another way. I have 119 seconds or your 23 seconds should be 24 seconds to hash the work unit, find shares, and return those shares for money.

The 120 seconds is only to add either a second for submission or isn't in existance.

This left me wondering why you said 120 seconds to start with and only had math to make 119. Possibly to account for round trip delays or the share validation. You didn't really say. So either your math was off, or you left out part of your math.

Expiry still at 120
scan time adjusted to 95
Same Unknown Work
legendary
Activity: 2576
Merit: 1186
Throwing out work with every notify message on stratum, which comes on average every 30 seconds, would have the same effect as a longpoll every 30 seconds. That's a lot of work to discard midstream. Why on earth would there even be a clean_work method with stratum if every notify implied we throw out the work?
You're not supposed to throw out the shares on LPs. Restarting work has no cost except on BFLs (where you have a bug that loses valid shares).

So Junior is saying my reject ~9minutes into the block is because CGMiner didn't change to new work fast enough?
That sounds highly unlikely, and unrelated. What kind of a reject?

Top post on the page you are asking on. LOOK UP!
Ok, I missed that somehow. "unknown-work" can only occur if one of these 3 cases:
  • a block was found (which you ruled out)
  • the miner submitted it after it expired (120 seconds on EMC; new work is delivered at least every 96 seconds)
  • the user extranonce1 was lost (only possible if you reconnected)
I'm pretty sure cgminer discards all work/shares in the last case (reconnection), so that means it either:
  • was working on a job 23 seconds after it had been replaced (this is sufficient for as low as 200 Mh/s)
  • was working on a job 21 seconds after it had been replaced AND took over 2 seconds to submit it to the pool
This makes sense, since cgminer is failing to move on to new jobs as they come in.

So a BFL single running 864mhash bitstream in your example would need to work on the work for 114.8 seconds to go over. Interesting to know.
No, that 21-23 seconds is real time. No matter what hashrate. 200 Mh/s is the hashrate where if you are any slower, you might still be working on the old work normally.
hero member
Activity: 981
Merit: 500
DIV - Your "Virtual Life" Secured and Decentralize
Throwing out work with every notify message on stratum, which comes on average every 30 seconds, would have the same effect as a longpoll every 30 seconds. That's a lot of work to discard midstream. Why on earth would there even be a clean_work method with stratum if every notify implied we throw out the work?
You're not supposed to throw out the shares on LPs. Restarting work has no cost except on BFLs (where you have a bug that loses valid shares).

So Junior is saying my reject ~9minutes into the block is because CGMiner didn't change to new work fast enough?
That sounds highly unlikely, and unrelated. What kind of a reject?

Top post on the page you are asking on. LOOK UP!
Ok, I missed that somehow. "unknown-work" can only occur if one of these 3 cases:
  • a block was found (which you ruled out)
  • the miner submitted it after it expired (120 seconds on EMC; new work is delivered at least every 96 seconds)
  • the user extranonce1 was lost (only possible if you reconnected)
I'm pretty sure cgminer discards all work/shares in the last case (reconnection), so that means it either:
  • was working on a job 23 seconds after it had been replaced (this is sufficient for as low as 200 Mh/s)
  • was working on a job 21 seconds after it had been replaced AND took over 2 seconds to submit it to the pool
This makes sense, since cgminer is failing to move on to new jobs as they come in.

So a BFL single running 864mhash bitstream in your example would need to work on the work for 114.8 seconds to go over. Interesting to know.
I didn't change scan time for stratum but the settings of scan time at 115 had worked fine on get work.

I will bump it down but since not every 120 seconds has this happening I would assume it is something else. The ping to Eclipse is 76ms right now. It can vary but assuming I got my work at 96 seconds add .5 for significant delays, 5.16s for single to process work I get 101.66 seconds. Leaving me 18.34 seconds to submit.....
legendary
Activity: 1540
Merit: 1001
Just adding my 0.0016 BTC worth here..

Since p2pool has gone berserk, I have half my miners pointing at EMC, and half at Oz.

Oz is using stratum.
EMC is using GBT.

After about 12 hours, there were noticeably more rejects on EMC than on Oz.  That's with using GBT, not stratum, on Oz.

Not sure if it means anything.

M
legendary
Activity: 2576
Merit: 1186
Throwing out work with every notify message on stratum, which comes on average every 30 seconds, would have the same effect as a longpoll every 30 seconds. That's a lot of work to discard midstream. Why on earth would there even be a clean_work method with stratum if every notify implied we throw out the work?
You're not supposed to throw out the shares on LPs. Restarting work has no cost except on BFLs (where you have a bug that loses valid shares).

So Junior is saying my reject ~9minutes into the block is because CGMiner didn't change to new work fast enough?
That sounds highly unlikely, and unrelated. What kind of a reject?

Top post on the page you are asking on. LOOK UP!
Ok, I missed that somehow. "unknown-work" can only occur if one of these 3 cases:
  • a block was found (which you ruled out)
  • the miner submitted it after it expired (120 seconds on EMC; new work is delivered at least every 96 seconds)
  • the user extranonce1 was lost (only possible if you reconnected)
I'm pretty sure cgminer discards all work/shares in the last case (reconnection), so that means it either:
  • was working on a job 23 seconds after it had been replaced (this is sufficient for as low as 200 Mh/s)
  • was working on a job 21 seconds after it had been replaced AND took over 2 seconds to submit it to the pool
This makes sense, since cgminer is failing to move on to new jobs as they come in.
legendary
Activity: 4634
Merit: 1851
Linux since 1997 RedHat 4
Throwing out work with every notify message on stratum, which comes on average every 30 seconds, would have the same effect as a longpoll every 30 seconds. That's a lot of work to discard midstream. Why on earth would there even be a clean_work method with stratum if every notify implied we throw out the work?
You're not supposed to throw out the shares on LPs. Restarting work has no cost except on BFLs (where you have a bug that loses valid shares).

So Junior is saying my reject ~9minutes into the block is because CGMiner didn't change to new work fast enough?
That sounds highly unlikely, and unrelated. What kind of a reject?

Top post on the page you are asking on. LOOK UP!
Luke-Jr thank you for yet again pointing out why people who use your miner get less shares Tongue (and shouldn't be using it)
Yeah processing previous block work after an LP ... right ... ... ...
hero member
Activity: 981
Merit: 500
DIV - Your "Virtual Life" Secured and Decentralize
Throwing out work with every notify message on stratum, which comes on average every 30 seconds, would have the same effect as a longpoll every 30 seconds. That's a lot of work to discard midstream. Why on earth would there even be a clean_work method with stratum if every notify implied we throw out the work?
You're not supposed to throw out the shares on LPs. Restarting work has no cost except on BFLs (where you have a bug that loses valid shares).

So Junior is saying my reject ~9minutes into the block is because CGMiner didn't change to new work fast enough?
That sounds highly unlikely, and unrelated. What kind of a reject?

Top post on the page you are asking on. LOOK UP!
legendary
Activity: 1795
Merit: 1208
This is not OK.
hero member
Activity: 807
Merit: 500

That mean you have to go to IHOP.
1) I don't like hopping
2) I don't like pancakes
3) Prayer takes so much time
4) I assume I detected your humor, but if I did not, or if IHOP means something else as well, feel free to tell.  Tongue
legendary
Activity: 1795
Merit: 1208
This is not OK.
Debug build would be helpful. If you're running fedora that suggests you're building it yourself. If that's the case, add "-g" to your CFLAGS, rebuild your cgminer 2.9.5 (without stripping the file), enable coredumps with "ulimit -c unlimited" and then run whatever it is that's crashing. Once you get a "core dumped" message at the end of running it, then run:
gdb cgminer core
bt full

And post the information from that please.
I'm getting "No stack."  I assume that makes it obvious I've never done this before...

That means you have to go to IHOP.
Jump to: