[DEAD] DeepBit.net PPS+Prop,instant payouts, we pay for INVALID BLOCKS too - page 99.

slush

legendary

Activity: 1386

Merit: 1097

Quote from: jedi95 on December 31, 2011, 09:15:16 PM

Phoenix never supported nTime rolling, so this isn't possible.

Not sure why, but some miners which were reporting themselves as "phoenix" already performed some ghetto-style ntime rolling. They were using one job for many minutes, only extending the ntime, which is corrupted behaviour (miner should ask for new job even when there's no new bitcoin block, to refresh his merkle tree). Not sure if it was done by unofficial patch floating on the Internet, but I implemented few workarounds especially because of such issues with phoenix.

DeepBit

donator

Activity: 532

Merit: 501

We have cookies

Quote from: Hawkix on January 01, 2012, 10:50:54 AM

The only wrong number is the last 10 minutes average hashrate, which is about 10% lower than it used to be for months. Surprisingly, when I counted number of submitted shares (reported on deepbit) and used that to calculate my hashrate, I got the correct, good-old number.

To me, it looks like someone changed the code for estimating hashrate.

That's strange. Yes, I noticed this 10% drop with my pool monitoring routines, but it only lasted for a few hours.
I just checked the rate estimation code, but there is nothing that can be affected by new year or absolute time and there were no changes recently.
PM me your login name please so I can check your displayed hashrate.

At least if your submitted shares number is the same, your income won't be affected.

Hawkix

hero member

Activity: 531

Merit: 505

I restarted all workers, but hashrate is still reported as low.

What is interesting, i cannot see any problems in the clients, no reports about disconnecting, empty work queue, etc.

The only wrong number is the last 10 minutes average hashrate, which is about 10% lower than it used to be for months. Surprisingly, when I counted number of submitted shares (reported on deepbit) and used that to calculate my hashrate, I got the correct, good-old number.

To me, it looks like someone changed the code for estimating hashrate.

DeepBit

donator

Activity: 532

Merit: 501

We have cookies

Quote from: kitich on January 01, 2012, 04:59:24 AM

Hello DeepBit. HNY )
What happened at 00:00? =)

Thanks.

Check this post: https://bitcointalksearch.org/topic/m.669081
Since another pool on the same DC reported similar hashrate drop I think that it was caused by some peering link suffered from attack or newyearing, but this lasted only a few hours and now it's normal again. Do you still experience this issue ?
Can you try reconnecting one of your miners and check if it's normal again or not?
Do you see any packet loss to pit.deepbit.net ?

Hawkix

hero member

Activity: 531

Merit: 505

I must confirm that I also experience about 10% reported hashrate drop since new year. Not sure if it is real hashrate drop (aka slower shares submissions), or just different formula for computing the number. Can you please check it?

kitich

newbie

Activity: 44

Merit: 0

Hello DeepBit. HNY )

What happened at 00:00? =)

Daily graph:
http://dl.dropbox.com/u/4553930/ff_hashrate_day.png

Weekly graph:
http://dl.dropbox.com/u/4553930/ff_hashrate_week.png

My local hashrate is at the same level than before 00:00. About 9.5Ghs. All workers are up.

DeepBit

donator

Activity: 532

Merit: 501

We have cookies

Quote from: jedi95 on December 31, 2011, 09:15:16 PM

Phoenix never supported nTime rolling, so this isn't possible. I added support for it to the backend in 1.7.2, but the miner core doesn't do anything with it yet.

Sorry, I got this info from slush and had no time to check myself yet.

jedi95

full member

Activity: 219

Merit: 120

Quote from: DeepBit on December 30, 2011, 04:55:44 PM

UPDATE: Looks like I found one possible reason. There may be a bug in the Phoenix miner that causes it not to check the allowed nTime rolling range, so on slow hardware this may cause some shares to be... well, almost valid.
This needs additional checking.

Phoenix never supported nTime rolling, so this isn't possible. I added support for it to the backend in 1.7.2, but the miner core doesn't do anything with it yet.

DeepBit

donator

Activity: 532

Merit: 501

We have cookies

Quote from: LightRider on December 31, 2011, 06:04:55 PM

Getting a lot of errors recently. Problems with the site?

If you are talking about last ~4 hours then I noticed some hashrate drop, possibly caused by connection problems between DCs.
~2% of users were affected, but I don't know for sure yet what is the cause. Looking into this issue.
What kind of errors are you getting ?

UPDATE: Looks like the issue is resolved.

LightRider

legendary

Activity: 1500

Merit: 1022

I advocate the Zeitgeist Movement & Venus Project.

Getting a lot of errors recently. Problems with the site?

DeepBit

donator

Activity: 532

Merit: 501

We have cookies

Quote from: RobertRibbeck on December 30, 2011, 03:00:17 PM

hello !!!!!!!!!!!!!!!!!!!!!!

I already sent you four PMs today, no need to add that many exclamation marks :)

Since you are experiencing this issue only with two PCs and only with phoenix, but not poclbm and I don't see any errors in your submissions, I can assume that it may be something caused my combination of the phoenix miner and your internet connection (or hardware).
I'll check my logs again tomorrow, may be I'll ask you to post sample share that was not accounted for (but not today, not ready to check it yet).

UPDATE: Looks like I found one possible reason. There may be a bug in the Phoenix miner that causes it not to check the allowed nTime rolling range, so on slow hardware this may cause some shares to be... well, almost valid.
This needs additional checking.

Currently the best miner is poclbm or poclbm-based GUIminer. Other miners may be sometimes a bit faster, but may contain some bugs.

RobertRibbeck

full member

Activity: 221

Merit: 100

I just watched several shares sent and show accepted by Phoenix 1.7.1
pool didn't count at least 3 of the shares

how can a accepted be sent back to the miner but pool total never counts them Huh

I'm seeing at least 5% loss per day for the last month alone

I reported this behavior over a month ago
NO REPLY - NO ANSWER

hello !!!!!!!!!!!!!!!!!!!!!!

Eveofwar

sr. member

Activity: 406

Merit: 250

Quote from: os2sam on December 28, 2011, 03:35:00 PM

Where did the shares per block averages go on the stats page? We're no where near a difficulty change.
Sam

Stats are delayed by one hour
Average shares per block: 1269163 (+9.4%)
Average in last 24 hours: 1364050 (+17.6%)

?

os2sam

legendary

Activity: 3583

Merit: 1094

Think for yourself

Quote from: DeepBit on December 28, 2011, 03:45:01 PM

Quote from: os2sam on December 28, 2011, 03:35:00 PM

Where did the shares per block averages go on the stats page? We're no where near a difficulty change.

Sorry, it was caused by site update. Already enabled again.

No problem.
Thanks,
Sam

DeepBit

donator

Activity: 532

Merit: 501

We have cookies

Quote from: os2sam on December 28, 2011, 03:35:00 PM

Where did the shares per block averages go on the stats page? We're no where near a difficulty change.

Sorry, it was caused by site update. Already enabled again.

os2sam

legendary

Activity: 3583

Merit: 1094

Think for yourself

Where did the shares per block averages go on the stats page? We're no where near a difficulty change.
Sam

Fiyasko

legendary

Activity: 1428

Merit: 1001

Okey Dokey Lokey

Quote from: DeepBit on December 25, 2011, 01:49:25 AM

Quote from: SlaveInDebt on December 25, 2011, 01:38:37 AM

Merry X-dos Grin

What do you mean ?
May be "X-mas"?

no kidding If anything went down, I never saw a hiccup Cheesy

Deepbit's been running great for awhile now

DeepBit

donator

Activity: 532

Merit: 501

We have cookies

Quote from: SlaveInDebt on December 25, 2011, 01:38:37 AM

Merry X-dos ;D

What do you mean ?
May be "X-mas"? :)

SlaveInDebt

hero member

Activity: 699

Merit: 500

Your Minion

Merry X-dos Grin

Fiyasko

legendary

Activity: 1428

Merit: 1001

Okey Dokey Lokey

Quote from: DeepBit on December 22, 2011, 02:31:18 PM

Quote from: Fiyasko on December 22, 2011, 02:15:34 PM

Hey, Three of my miners went down at the same time, And the "auto alert by email" system did not send me an email about the failure to have submitted a peice of work untill Four Hours later when i logged into deepbit and saw the RedText of my failed miners, About 4mins after seeing the red, and downned miners, I got the email.... HMMMM NOT HAPPY
I have the Threshold set to three minutes (now two)

PM me your login name so I can check when the message was really sent. May be it was delayed somewhere in transit.

Also, looks like you didn't noticed how e-mail alert works: https://deepbit.net/help/2#2
Blue/red text depends on received shares, but e-mail alert checks getwork requests to prevent false positives, so changing threshold doesn't affects this.

Correct.
The threshold value i stated because it felt like "one minute after seeing that your miners were down: we began to count to 3min, And when we did, you got the email"

PM'd you my info

---Also, My miners (with a 2min threshold [i understand the getwork Y/N=Alive/Dead/Why argument]) Constantly flickered into red "failure" (as i expected and intended) But i still have not recieved an email of "failed mining"

Topic: [DEAD] DeepBit.net PPS+Prop,instant payouts, we pay for INVALID BLOCKS too - page 99. (Read 1601412 times)