Bitcoin Forum
June 28, 2017, 05:51:39 PM *
News: Latest stable version of Bitcoin Core: 0.14.2  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 ... 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 [354] 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 ... 777 »
  Print  
Author Topic: [1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool  (Read 2326863 times)
nreal
Full Member
***
Offline Offline

Activity: 186


View Profile
November 30, 2013, 06:50:06 AM
 #7061

Can anyone tell me some tricks to minimize bitcoind GetBlockTemplate latency?  Mine seems to run very consistently at ~1s, but I see many public nodes with less than 0.3s.

What is the source of this latency and what can be done to address it?


blockmaxsize= bigger the slower, minrelaytxfee= smaller is slower , bitcoind on ssd or ramdisk is fast. Bitcoin blocks are stored very  stupid way, because blocks arent in folders, so you cant mount --bind some of them to ramdisk, I think all the blocks must be in ram or ssd to get any improvement.. 15,4gt now..



If my node is configured with 1mbit blockmaxsize and minrelaytxfee=0.00001 latency is 0.4-0.9s. That makes miners go allmost asleep - i dont get much shares with these settings alltough it doesnt affect efficiency rate.

blockmaxsize=100 000 minrelaytxfee=0.0005 latency is +-0.03 and butterflylabs sc:s start to make noise.

Ordered 32gb ram to make 24gb ramdisk for bitcoin, cant wait to see what is the effect with that.

Gentoo makes it east to make a ramdisk

file /etc/fstab

tmpfs      /var/tmp/portage   tmpfs     noatime,nodev,nosuid      0 0

kernel three or two  settings

CONFIG_BLK_DEV_RAM=y
CONFIG_BLK_DEV_RAM_COUNT=16
CONFIG_BLK_DEV_RAM_SIZE=24576 > for 24gb ramdisk

1498672299
Hero Member
*
Offline Offline

Posts: 1498672299

View Profile Personal Message (Offline)

Ignore
1498672299
Reply with quote  #2

1498672299
Report to moderator
1498672299
Hero Member
*
Offline Offline

Posts: 1498672299

View Profile Personal Message (Offline)

Ignore
1498672299
Reply with quote  #2

1498672299
Report to moderator
POLONIEX TRADING SIGNALS
+50% Profit and more via TELEGRAM
ALTCOINTRADER.CO
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1498672299
Hero Member
*
Offline Offline

Posts: 1498672299

View Profile Personal Message (Offline)

Ignore
1498672299
Reply with quote  #2

1498672299
Report to moderator
1498672299
Hero Member
*
Offline Offline

Posts: 1498672299

View Profile Personal Message (Offline)

Ignore
1498672299
Reply with quote  #2

1498672299
Report to moderator
1498672299
Hero Member
*
Offline Offline

Posts: 1498672299

View Profile Personal Message (Offline)

Ignore
1498672299
Reply with quote  #2

1498672299
Report to moderator
Krak
Hero Member
*****
Offline Offline

Activity: 591



View Profile WWW
November 30, 2013, 07:13:02 AM
 #7062

If my node is configured with 1mbit blockmaxsize and minrelaytxfee=0.00001 latency is 0.4-0.9s. That makes miners go allmost asleep - i dont get much shares with these settings alltough it doesnt affect efficiency rate.

blockmaxsize=100 000 minrelaytxfee=0.0005 latency is +-0.03 and butterflylabs sc:s start to make noise.
Changing these settings doesn't affect the share rate whatsoever.

BTC: 1KrakenLFEFg33A4f6xpwgv3UUoxrLPuGn
nreal
Full Member
***
Offline Offline

Activity: 186


View Profile
November 30, 2013, 09:37:01 AM
 #7063

If my node is configured with 1mbit blockmaxsize and minrelaytxfee=0.00001 latency is 0.4-0.9s. That makes miners go allmost asleep - i dont get much shares with these settings alltough it doesnt affect efficiency rate.

blockmaxsize=100 000 minrelaytxfee=0.0005 latency is +-0.03 and butterflylabs sc:s start to make noise.
Changing these settings doesn't affect the share rate whatsoever.

Well I can say that youre wrong, very much indeed. Going from 0.03s to 0.4s latency you will see alot less shares with asics, because of the lag ofcourse. Immediately you can hear the difference when your equipment runs with work all the time, fans start to really blow..

Try it, and tell us how it went.

one week with 1000 000 blocksize and 0.0001 fees, then the next week drop it to 50 000 and 0.005 fees. If it doesnt make you more shares then the first week im suprised.

But if you have bitcoind on ramdisk difference might not be that big then, im going to try that when my memorys arrive.
nreal
Full Member
***
Offline Offline

Activity: 186


View Profile
November 30, 2013, 10:02:04 AM
 #7064

If my node is configured with 1mbit blockmaxsize and minrelaytxfee=0.00001 latency is 0.4-0.9s. That makes miners go allmost asleep - i dont get much shares with these settings alltough it doesnt affect efficiency rate.

blockmaxsize=100 000 minrelaytxfee=0.0005 latency is +-0.03 and butterflylabs sc:s start to make noise.
Changing these settings doesn't affect the share rate whatsoever.

Well I can say that youre wrong, very much indeed. Going from 0.03s to 0.4s latency you will see alot less shares with asics, because of the lag ofcourse. Immediately you can hear the difference when your equipment runs with work all the time, fans start to really blow..

Try it, and tell us how it went.

one week with 1000 000 blocksize and 0.0001 fees, then the next week drop it to 50 000 and 0.005 fees. If it doesnt make you more shares then the first week im suprised.

But if you have bitcoind on ramdisk difference might not be that big then, im going to try that when my memorys arrive.

Theres one thing if you run your node on linux kernel that can be tweaked too, infact two

Preemptible Kernel (Low-Latency Desktop) (PREEMPT)

CONFIG_PREEMPT:

This option reduces the latency of the kernel by making
all kernel code (that is not executing in a critical section)
preemptible. This allows reaction to interactive events by
permitting a low priority process to be preempted involuntarily
even if it is in kernel mode executing a system call and would
otherwise not be about to reach a natural preemption point.
This allows applications to run more 'smoothly' even when the
system is under load, at the cost of slightly lower throughput
and a slight runtime overhead to kernel code.

Select this if you are building a kernel for a desktop or
embedded system with latency requirements in the milliseconds
range.

Nr2

Timer frequency

Allows the configuration of the timer frequency. It is customary
to have the timer interrupt run at 1000 Hz but 100 Hz may be more
beneficial for servers and NUMA systems that do not need to have
a fast response for user interaction and that may experience bus
contention and cacheline bounces as a result of timer interrupts.
Note that the timer interrupt occurs on each processor in an SMP
environment leading to NR_CPUS * HZ number of timer interrupts
per second.

Prompt: Timer frequency
Location:
-> Processor type and features
Defined at kernel/Kconfig.hz:5
Selected by: m

1000 HZ (HZ_1000)

CONFIG_HZ_1000:

1000 Hz is the preferred choice for desktop systems and other
systems requiring fast interactive responses to events.

Symbol: HZ_1000 [=y]
Type : boolean
Prompt: 1000 HZ
Location:
-> Processor type and features
-> Timer frequency (<choice> [=y])
Defined at kernel/Kconfig.hz:42
Depends on: <choice>
lanfanblue
Jr. Member
*
Offline Offline

Activity: 31


View Profile
November 30, 2013, 10:32:44 AM
 #7065

If my node is configured with 1mbit blockmaxsize and minrelaytxfee=0.00001 latency is 0.4-0.9s. That makes miners go allmost asleep - i dont get much shares with these settings alltough it doesnt affect efficiency rate.

blockmaxsize=100 000 minrelaytxfee=0.0005 latency is +-0.03 and butterflylabs sc:s start to make noise.
Changing these settings doesn't affect the share rate whatsoever.

Well I can say that youre wrong, very much indeed. Going from 0.03s to 0.4s latency you will see alot less shares with asics, because of the lag ofcourse. Immediately you can hear the difference when your equipment runs with work all the time, fans start to really blow..

Try it, and tell us how it went.

one week with 1000 000 blocksize and 0.0001 fees, then the next week drop it to 50 000 and 0.005 fees. If it doesnt make you more shares then the first week im suprised.

But if you have bitcoind on ramdisk difference might not be that big then, im going to try that when my memorys arrive.

In my experience, the block template latency has minor effects on the share rate.
I use Avalon miner. I tuned this latency from <0.1 to ~0.5 by changing maxblocksize from 50k to 700k. In both scenarios, I got almost the same share rate.
You can check log of bitcoind, p2pool asks for a new blocktemplate approximately every 15s. This means, the asic miner works on different block template every 15s. I don't think it's a big deal whether that latency is 0.05s or 0.5s compared to 15s working time period. (0.5/15=3% DOA, but my avalon has 5% DOA due to its long latency. So it won't hurt me a lot. But it might be an issue if it's >1s.)

But I found that the blocksize might be related to orphan rate. When I was using the default block size 250k, I got lots of orphans. When I tuned it to other values, no matter larger or smaller, the orphan rate dropped. But this might be wring, because my Internet connection is really unstable.
nreal
Full Member
***
Offline Offline

Activity: 186


View Profile
November 30, 2013, 10:54:02 AM
 #7066

So its different here, because with 1000 000 blocksize and latency 0.4-1s my hashreate drops every now and then from 120ghash to even 80ghash. with 0.03 latency it keeps goin at 110 - 135 ghash.

Bitcoind uses sometimes 1% cputime on core i7 3770k ,ssd disk, 16gb ram

astutiumRob
Full Member
***
Offline Offline

Activity: 197



View Profile WWW
November 30, 2013, 01:31:04 PM
 #7067

Is anyone mining October Jupiters on P2Pool with cgminer successfully
12 days (since last firmware reconfig), no crashes, no problems with mine.

www.astutium.com - domains | hosting | vps | servers | cloud - proud to accept bitcoins. UK colocation for BFL and KNC ASICs in Tier3+ DC
Register Domains with BTC
Want to make some bitcoins ? Miner on ebay | Buy GH/s
aamarket
Sr. Member
****
Offline Offline

Activity: 258


View Profile WWW
November 30, 2013, 05:05:17 PM
 #7068

I have a public p2pool node running for some time,
the last week on cheap tvstick mk888b hardware.
It seems running fine, probably latency is a bit worse
than with Intel i3-3220 before.

If anybody wants to test, use your btc address as username,
any password(different from username) and go to

http://aamarket.eu:9332/

feedback welcome.

IMPORTANT:http://bitcointalk.org/index.php?topic=177133.0,Tips welcome BTC:1AAMARKETmJvfjDwEFmhyYYwfre7ZFVseP  RIC:RGnX6LcJrsVEuYeySDDxkmH7AjRqoprcKt
nreal
Full Member
***
Offline Offline

Activity: 186


View Profile
December 01, 2013, 08:09:50 AM
 #7069

2013-12-01 10:05:22.933224 GOT SHARE! 1ghdfrxzhjghjk---- 250f99fa prev ba9ac418 age 14.43s
2013-12-01 10:05:22.956239 > Error while processing Event callbacks:
2013-12-01 10:05:22.956296 > Traceback (most recent call last):
2013-12-01 10:05:22.956323 >   File "/mnt/p2pool/p2pool/bitcoin/worker_interface.py", line 136, in <lambda>
2013-12-01 10:05:22.956349 >     lambda header, user, coinbase_nonce: handler(header, user, pack.IntType(self._my_bits).pack(nonce) + coinbase_nonce),
2013-12-01 10:05:22.956374 >   File "/mnt/p2pool/p2pool/work.py", line 403, in got_response
2013-12-01 10:05:22.956397 >     self.node.set_best_share()
2013-12-01 10:05:22.956421 >   File "/mnt/p2pool/p2pool/node.py", line 297, in set_best_share
2013-12-01 10:05:22.956457 >     self.best_share_var.set(best)
2013-12-01 10:05:22.956487 >   File "/mnt/p2pool/p2pool/util/variable.py", line 74, in set
2013-12-01 10:05:22.956510 >     self.changed.happened(value)
2013-12-01 10:05:22.956532 > --- <exception caught here> ---
2013-12-01 10:05:22.956554 >   File "/mnt/p2pool/p2pool/util/variable.py", line 42, in happened
2013-12-01 10:05:22.956577 >     func(*event)
2013-12-01 10:05:22.956599 >   File "/mnt/p2pool/p2pool/node.py", line 96, in broadcast_share
2013-12-01 10:05:22.956622 >     peer.sendShares([share for share in shares if share.peer_addr != peer.addr], self.node.tracker, self.node.known_txs_var.value, include_txs_with=[share_hash])
2013-12-01 10:05:22.956648 >   File "/mnt/p2pool/p2pool/p2p.py", line 291, in sendShares
2013-12-01 10:05:22.956671 >     assert tx_hash in known_txs, 'tried to broadcast share without knowing all its new transactions'
2013-12-01 10:05:22.956694 > exceptions.AssertionError: tried to broadcast share without knowing all its new transactions


2013-12-01 10:07:28.634340 P2Pool: 17342 shares in chain (17346 verified/17346 total) Peers: 20 (3 incoming)
2013-12-01 10:07:28.634406  Local: 123GH/s in last 10.0 minutes Local dead on arrival: ~4.7% (3-7%) Expected time to share: 2.3 hours
2013-12-01 10:07:28.634436  Shares: 1 (1 orphan, 0 dead) Stale rate: ~100.0% (20-100%) Efficiency: ~0.0% (0-97%) Current payout: 0.0000 BTC
xgtele
Sr. Member
****
Offline Offline

Activity: 282



View Profile
December 01, 2013, 08:27:01 AM
 #7070

I have a public p2pool node running for some time,
the last week on cheap tvstick mk888b hardware.
It seems running fine, probably latency is a bit worse
than with Intel i3-3220 before.

If anybody wants to test, use your btc address as username,
any password(different from username) and go to

http://aamarket.eu:9332/

feedback welcome.


Looks interesting. Which OS do you installed on it?

wlz2011
Member
**
Offline Offline

Activity: 71


View Profile
December 03, 2013, 04:21:57 AM
 #7071


1. DOA變100%的問題我也有注意到,目前重開挖礦程式就會好,似乎和運算能力沒關係,比較像是隨機出現。

2.

我目前沒有大於4.5T的環境可以測試,而且最近在我礦池上都沒發生null。

比對之前有出現短暫null時,整體礦池也沒有超過4.5Thash/s過。

是否可以試著換挖礦程式 cgminer 或 bfgminer 試試。

3. taiwan pool的配置無法從外面連線,我有先將你的IP加入我連線的清單,伺服器下次重啟就會連上你的礦池。

如果有任何問題歡迎pm跟我討論 Smiley

183.136.216.39:9332 服务器停机维护

2013.12.03  我们开始测试矿池 http://mine.yuyi.tw:9332

矿池算力上到4.5T,算力溢出很明显。如下图



http://i44.tinypic.com/jj71hg.jpg

大概有300G的算力被溢出,并且抢走了2个SHARE。 Grin

这仅仅是个开始,相信会有足够的证据留在图表中。
yuyi
Newbie
*
Offline Offline

Activity: 6


View Profile
December 03, 2013, 05:06:16 PM
 #7072

It is not a problem, but it had been solved in my pool. ( http://mine.yuyi.tw:9332 )

I already sent you a pm in detail.  Smiley



183.136.216.39:9332 服务器停机维护

2013.12.03  我们开始测试矿池 http://mine.yuyi.tw:9332

矿池算力上到4.5T,算力溢出很明显。如下图

http://i44.tinypic.com/jj71hg.jpg

http://i44.tinypic.com/jj71hg.jpg

大概有300G的算力被溢出,并且抢走了2个SHARE。 Grin

这仅仅是个开始,相信会有足够的证据留在图表中。
forrestv
Hero Member
*****
Offline Offline

Activity: 515


View Profile
December 03, 2013, 06:48:45 PM
 #7073

P2Pool release 13.4 - commit hash: 8e7cb8223a38bdd4ff55971d2945f5ccb571e063

Windows binary: http://u.forre.st/u/wftncvpn/p2pool_win32_13.4.zip
Windows binary signature: http://u.forre.st/u/eoietnxc/p2pool_win32_13.4.zip.sig
Source zipball: https://github.com/forrestv/p2pool/zipball/13.4
Source tarball: https://github.com/forrestv/p2pool/tarball/13.4

Changes:
* Require Litecoin >=0.8.5.1
* Warn about old versions of Bitcoin and Terracoin
* Many improvements to web interface and graphs
* Fixed error messages being created in bitcoind's debug.log due to invalid message being sent

1J1zegkNSbwX4smvTdoHSanUfwvXFeuV23
Raulnsh
Newbie
*
Offline Offline

Activity: 22


View Profile
December 04, 2013, 04:25:31 AM
 #7074

Hi,
I have some strange problem. I've discovered that my node of version 13.3 kept litecoin log from 2013-08-21 (when I've installed it) to 2013-08-22. No log since 2013-08-23; yesterday it have stopped payment to default address for found shares. I've restarted it, and now I see log for 2013-12-04.
Why this can happen?
forrestv
Hero Member
*****
Offline Offline

Activity: 515


View Profile
December 04, 2013, 04:37:20 AM
 #7075

Hi,
I have some strange problem. I've discovered that my node of version 13.3 kept litecoin log from 2013-08-21 (when I've installed it) to 2013-08-22. No log since 2013-08-23; yesterday it have stopped payment to default address for found shares. I've restarted it, and now I see log for 2013-12-04.
Why this can happen?

If your hard drive was full or some other error happened while writing to the log file, it just stops writing to the log file but continues to work.

1J1zegkNSbwX4smvTdoHSanUfwvXFeuV23
oteoztsoffn
Newbie
*
Offline Offline

Activity: 14


View Profile
December 04, 2013, 05:19:48 AM
 #7076

Why not state clearly in this thread that p2pool.org and p2pool.com are not affiliated with p2pool?
forrestv
Hero Member
*****
Offline Offline

Activity: 515


View Profile
December 04, 2013, 05:26:15 AM
 #7077

Why not state clearly in this thread that p2pool.org and p2pool.com are not affiliated with p2pool?

I added some clarification about p2pool.org to the first post. I think p2pool.com is pretty obviously not.

In other related news, using http://p2pool.info/ as the new homepage for P2Pool is on the horizon since I recently took up maintenance of it and as it's already well-known.

1J1zegkNSbwX4smvTdoHSanUfwvXFeuV23
K1773R
Legendary
*
Offline Offline

Activity: 1582


/dev/null


View Profile
December 04, 2013, 07:01:29 AM
 #7078

Why not state clearly in this thread that p2pool.org and p2pool.com are not affiliated with p2pool?

I added some clarification about p2pool.org to the first post. I think p2pool.com is pretty obviously not.

In other related news, using http://p2pool.info/ as the new homepage for P2Pool is on the horizon since I recently took up maintenance of it and as it's already well-known.
good, about time to handle with the guy @ p2pool.org

[GPG Public Key]  [Devcoin Builds]  [BBQCoin Builds]  [Multichain Blockexplorer]  [Multichain Blockexplorer - PoS Coins]  [Ufasoft Miner Linux Builds]
BTC/DVC/TRC/FRC: 1K1773RbXRZVRQSSXe9N6N2MUFERvrdu6y ANC/XPM AK1773RTmRKtvbKBCrUu95UQg5iegrqyeA NMC: NK1773Rzv8b4ugmCgX789PbjewA9fL9Dy1 LTC: LKi773RBuPepQH8E6Zb1ponoCvgbU7hHmd EMC: EK1773RxUes1HX1YAGMZ1xVYBBRUCqfDoF BQC: bK1773R1APJz4yTgRkmdKQhjhiMyQpJgfN
nreal
Full Member
***
Offline Offline

Activity: 186


View Profile
December 04, 2013, 07:35:20 AM
 #7079

I think there should be a graph for transactions too, it would help to compare nodes and for fun too.
Or is it a bad idea?
oteoztsoffn
Newbie
*
Offline Offline

Activity: 14


View Profile
December 04, 2013, 12:03:27 PM
 #7080

I think there should be a graph for transactions too, it would help to compare nodes and for fun too.
Or is it a bad idea?

You mean p2pool.info or what?
Pages: « 1 ... 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 [354] 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 ... 777 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!