[1500 TH] p2pool: Decentralized, DoS-resistant, Hop-Proof pool

[jonnybravo0311]

mdude77, there's some discussion in the S2 hardware thread about the lower-than-expected hash rates.  People are blaming a few things:

• Under-powered PSU with which S2 ships
• Crap soldering of chips to the boards

I've been debating between the S2 and the SP-10 Dawson as my next hardware purchase.  The price difference is pretty significant - at this point with S2 coupons (or 112bit's group buy) I could almost get 2 of them for the price of a single Dawson - even if I were to jump on RoadStress' group buy of the Dawson.  Unfortunately I don't know if I've got enough available circuits in my house to handle 2 more miners - and paying someone outlandish rates to host in a data center isn't going to happen, either.  I might be calling in the electrician to see if I can get a new 240/30 circuit and install my own rack/cabinet at home.

Anyhow... that was way off topic, sorry about that

[mdude77]

mdude77, there's some discussion in the S2 hardware thread about the lower-than-expected hash rates.  People are blaming a few things:

• Under-powered PSU with which S2 ships
• Crap soldering of chips to the boards

I've been debating between the S2 and the SP-10 Dawson as my next hardware purchase.  The price difference is pretty significant - at this point with S2 coupons (or 112bit's group buy) I could almost get 2 of them for the price of a single Dawson - even if I were to jump on RoadStress' group buy of the Dawson.  Unfortunately I don't know if I've got enough available circuits in my house to handle 2 more miners - and paying someone outlandish rates to host in a data center isn't going to happen, either.  I might be calling in the electrician to see if I can get a new 240/30 circuit and install my own rack/cabinet at home.

Anyhow... that was way off topic, sorry about that

Can you give me the link you are referring to?  As I said it only happens with p2pool.  And I replaced the PSU.

M

[jonnybravo0311]

mdude77, there's some discussion in the S2 hardware thread about the lower-than-expected hash rates.  People are blaming a few things:

• Under-powered PSU with which S2 ships
• Crap soldering of chips to the boards

I've been debating between the S2 and the SP-10 Dawson as my next hardware purchase.  The price difference is pretty significant - at this point with S2 coupons (or 112bit's group buy) I could almost get 2 of them for the price of a single Dawson - even if I were to jump on RoadStress' group buy of the Dawson.  Unfortunately I don't know if I've got enough available circuits in my house to handle 2 more miners - and paying someone outlandish rates to host in a data center isn't going to happen, either.  I might be calling in the electrician to see if I can get a new 240/30 circuit and install my own rack/cabinet at home.

Anyhow... that was way off topic, sorry about that

Can you give me the link you are referring to?  As I said it only happens with p2pool.  And I replaced the PSU.

M

I saw your other post as well.  I missed the part where you stated this was solely a p2pool issue, and that Eligius reports at 990GH/s.  I checked the graphs on your node, and the attached miners seem to have the proper mean hashing speed (assuming of course your S2s are on your node) .

What really strikes me as odd is the S2 itself reporting the lower-than-expected hash rate exclusively when you point it to p2pool.  I wonder if the version of cgminer the S2 ships with might be the culprit?  I know kano had released a new binary of cgminer for the S1.  Maybe he's got one for the S2 as well?

Wish I could be of more help.  In your other thread you mentioned there is one chain not reporting temps, but only a single X in the list, so I'm not sure that would be the problem.  Again, I keep falling back to your statement that it works properly on Eligius, but not on p2pool.  That really seems to rule out a hardware problem.  Since you've already changed out the PSU with no effect and made sure the boards are properly seated, I'm out of other ideas.

EDIT: p2pool itself has been acting up recently.  For example, there are a number of threads wondering why the getBlockTemplate latency is spiking across the board.  I'm a victim of it as well.  It's upwards of 1s now.  Between 3am and 8am this morning the latency was normal, but before and after it sits at least 2.5 times higher than normal.  I saw the same trend on the graphs on your node.  Another thing I've noticed is that my local DOA percentage has climbed.  Initially I was running about 1% or so... now I'm averaging about 4-6%.  You can see the steady progression upwards on the monthly graphs on my node.

[mdude77]

mdude77, there's some discussion in the S2 hardware thread about the lower-than-expected hash rates.  People are blaming a few things:

• Under-powered PSU with which S2 ships
• Crap soldering of chips to the boards

I've been debating between the S2 and the SP-10 Dawson as my next hardware purchase.  The price difference is pretty significant - at this point with S2 coupons (or 112bit's group buy) I could almost get 2 of them for the price of a single Dawson - even if I were to jump on RoadStress' group buy of the Dawson.  Unfortunately I don't know if I've got enough available circuits in my house to handle 2 more miners - and paying someone outlandish rates to host in a data center isn't going to happen, either.  I might be calling in the electrician to see if I can get a new 240/30 circuit and install my own rack/cabinet at home.

Anyhow... that was way off topic, sorry about that

Can you give me the link you are referring to?  As I said it only happens with p2pool.  And I replaced the PSU.

M

I saw your other post as well.  I missed the part where you stated this was solely a p2pool issue, and that Eligius reports at 990GH/s.  I checked the graphs on your node, and the attached miners seem to have the proper mean hashing speed (assuming of course your S2s are on your node) .

What really strikes me as odd is the S2 itself reporting the lower-than-expected hash rate exclusively when you point it to p2pool.  I wonder if the version of cgminer the S2 ships with might be the culprit?  I know kano had released a new binary of cgminer for the S1.  Maybe he's got one for the S2 as well?

Wish I could be of more help.  In your other thread you mentioned there is one chain not reporting temps, but only a single X in the list, so I'm not sure that would be the problem.  Again, I keep falling back to your statement that it works properly on Eligius, but not on p2pool.  That really seems to rule out a hardware problem.  Since you've already changed out the PSU with no effect and made sure the boards are properly seated, I'm out of other ideas.

EDIT: p2pool itself has been acting up recently.  For example, there are a number of threads wondering why the getBlockTemplate latency is spiking across the board.  I'm a victim of it as well.  It's upwards of 1s now.  Between 3am and 8am this morning the latency was normal, but before and after it sits at least 2.5 times higher than normal.  I saw the same trend on the graphs on your node.  Another thing I've noticed is that my local DOA percentage has climbed.  Initially I was running about 1% or so... now I'm averaging about 4-6%.  You can see the steady progression upwards on the monthly graphs on my node.

I only have one S2 at the moment.  I saw a spike in the latency as well on my p2pool node.  This low hashrate on p2pool isn't a recent phenomena, I've been playing with it for about 5 days now trying different things.  Only recently did it occur to me it's limited to p2pool only.

M

[ovidus]

Just got my node up and going at http://96.38.175.165:9332 it's setup as a public pool should be able to handle a few users, hosted out of Reno nv

[emjeey]

Hi!

I was minig at an public p2pool nod for a while and get paybents for every bloch founf - however, for yesterday's two I got no paymernt - how can I found out what went wrong (I still have my address at the activ users and predicted payment list.

Thanks for answer in advance!
MJ

[mdude77]

I've come to the conclusion my S2 doesn't work properly with p2pool because:

1 - p2pool can't feed it work fast enough.  Not likely because of vardiff, setting it to the max it seems to support (1024) doesn't change things
2 - the 10 second restarts
3 - something with my hardware

I'm thinking it's #2.

M

[phillipsjk]

EDIT: p2pool itself has been acting up recently.  For example, there are a number of threads wondering why the getBlockTemplate latency is spiking across the board.  I'm a victim of it as well.  It's upwards of 1s now.  Between 3am and 8am this morning the latency was normal, but before and after it sits at least 2.5 times higher than normal. 

Bitcoind version 0.9 changed the default blocksize from 500kB to 750kB. That may explain it. You can test by telling Bitcoin to make 500000 byte blocks again:

Code:

#Block creation options
blockminsize=0          #(Sizes are in bytes) default: 0
blockmaxsize=500000     #default: 750000
blockprioritysize=196000        # high-priority/low-fee transactions d: 50000

The blockprioritysize (number of free transactions included) was lowered as well (I set it to ~1/3)

[jonnybravo0311]

EDIT: p2pool itself has been acting up recently.  For example, there are a number of threads wondering why the getBlockTemplate latency is spiking across the board.  I'm a victim of it as well.  It's upwards of 1s now.  Between 3am and 8am this morning the latency was normal, but before and after it sits at least 2.5 times higher than normal. 

Bitcoind version 0.9 changed the default blocksize from 500kB to 750kB. That may explain it. You can test by telling Bitcoin to make 500000 byte blocks again:

Code:

#Block creation options
blockminsize=0          #(Sizes are in bytes) default: 0
blockmaxsize=500000     #default: 750000
blockprioritysize=196000        # high-priority/low-fee transactions d: 50000

The blockprioritysize (number of free transactions included) was lowered as well (I set it to ~1/3)

Hi phillipsjk,

I'm aware of the block size change in the 0.9.* clients; however, I'm not sure that would explain the latency spikes that are currently happening.  I've had 0.9.1 installed and running since release and it is only within the past few days that the latency has gone haywire.  My normal latency is ~0.4s, which by itself isn't that good since guys running the entire thing in a RAM drive get ~.02s.  In the past 2 days, the latency has spent many hours of ~1s latency.  It isn't only my node, either.  Nothing has changed on my end in terms of hardware/software/load/network traffic/etc.  Even if something had changed on my end, it wouldn't be reflected across multiple nodes on the network.

I've come to the conclusion my S2 doesn't work properly with p2pool because:

1 - p2pool can't feed it work fast enough.  Not likely because of vardiff, setting it to the max it seems to support (1024) doesn't change things
2 - the 10 second restarts
3 - something with my hardware

I'm thinking it's #2.

M

Curious... what 10 second restarts?  Anyone with an S2 on p2pool also experiencing this?  How about one of the SP-10 Dawsons?  Dragon 1T?

[Krak]

I've come to the conclusion my S2 doesn't work properly with p2pool because:

1 - p2pool can't feed it work fast enough.  Not likely because of vardiff, setting it to the max it seems to support (1024) doesn't change things
2 - the 10 second restarts
3 - something with my hardware

I'm thinking it's #2.

M

Curious... what 10 second restarts?  Anyone with an S2 on p2pool also experiencing this?  How about one of the SP-10 Dawsons?  Dragon 1T?

He's talking about the time between shares, but it's actually 30 seconds now.

[Polyatomic]

I've come to the conclusion my S2 doesn't work properly with p2pool because:

1 - p2pool can't feed it work fast enough.  Not likely because of vardiff, setting it to the max it seems to support (1024) doesn't change things
2 - the 10 second restarts
3 - something with my hardware

I'm thinking it's #2.

M

Its the host controlling your S2 man , probably terribly underpowered.

[mdude77]

I've come to the conclusion my S2 doesn't work properly with p2pool because:

1 - p2pool can't feed it work fast enough.  Not likely because of vardiff, setting it to the max it seems to support (1024) doesn't change things
2 - the 10 second restarts
3 - something with my hardware

I'm thinking it's #2.

M

Its the host controlling your S2 man , probably terribly underpowered.

Yet it works fine with a conventional pool, like Eligius?  That would also mean everyone else with S2s would have the same problem?

M

[mdude77]

I've come to the conclusion my S2 doesn't work properly with p2pool because:

1 - p2pool can't feed it work fast enough.  Not likely because of vardiff, setting it to the max it seems to support (1024) doesn't change things
2 - the 10 second restarts
3 - something with my hardware

I'm thinking it's #2.

M

I don't think it's #2.  I have it running through a proxy pointed to Eligius, and Eligius restarts every 30 seconds or so as well, yet I don't have the problem.

M

[ceslick]

Shares seem to be getting fewer and fewer at the moment..

[JoeGruff]

So I swapped my hdd for a ssd.  A used three year old ssd.

And it's great!

The blockchains take seconds to verify and load now.  my computer is dead silent.  And it uses about 10 less watts.

[windpath]

Hi phillipsjk,

I'm aware of the block size change in the 0.9.* clients; however, I'm not sure that would explain the latency spikes that are currently happening.  I've had 0.9.1 installed and running since release and it is only within the past few days that the latency has gone haywire.  My normal latency is ~0.4s, which by itself isn't that good since guys running the entire thing in a RAM drive get ~.02s.  In the past 2 days, the latency has spent many hours of ~1s latency.  It isn't only my node, either.  Nothing has changed on my end in terms of hardware/software/load/network traffic/etc.  Even if something had changed on my end, it wouldn't be reflected across multiple nodes on the network.

I wish... I'm 100% SSD, have the same latency issue as the rest of us...

[norgan]

quick question. when compiling how do you get the right version reported? I've compiled mine and it shows version f0eeb48. It should show 13.4-f0eeb48 or similar right?

[andmax]

I have just received my new Ant S1's trying to connect them to my node on the local network  I have setup the following 3 pools

Pool 1 192.168.1.20:9332
Pool 2 100.166.178.107:9332
pool 3 us2.eclipsemc.com:3333

but when the S1 connects is goes straight to pool 3 why won't it connect to my node on the local network, under the mining status tab the first 2 pools aren't in the list only pool 3 as alive, is there a setting I have missed.

[mdude77]

I have just received my new Ant S1's trying to connect them to my node on the local network  I have setup the following 3 pools

Pool 1 192.168.1.20:9332
Pool 2 100.166.178.107:9332
pool 3 us2.eclipsemc.com:3333

but when the S1 connects is goes straight to pool 3 why won't it connect to my node on the local network, under the mining status tab the first 2 pools aren't in the list only pool 3 as alive, is there a setting I have missed.

I use:

http://192.168.0.194:9332

so, try putting the http:// in front of it.

M

[andmax]

I have just received my new Ant S1's trying to connect them to my node on the local network  I have setup the following 3 pools

Pool 1 192.168.1.20:9332
Pool 2 100.166.178.107:9332
pool 3 us2.eclipsemc.com:3333

but when the S1 connects is goes straight to pool 3 why won't it connect to my node on the local network, under the mining status tab the first 2 pools aren't in the list only pool 3 as alive, is there a setting I have missed.

I use:

http://192.168.0.194:9332

so, try putting the http:// in front of it.

M

No still not working on my node, the pool 2 is now showing up in the Miner Status as Dead which is correct as it is my mates pool and it is down at the minute, but my node isn't even showing up. The only settings I changed is the IP the gateway and DNS ip's to point to my router under network/ WAN I haven't changed anything under the LAN tab that is just left on default which is protocol DHCP client