Question: My accounts stats often show variations in hash rate and lower than expected hash rates. According to Phoenix output on my miner I'm getting consistent, steady 295+310=605 MHash/s. Why does this vary so much on the account stats? Sometimes I see 515 MHash/s when Phoenix is telling me the same steady values at 605. Isn't this value just reported based on how quickly I complete shares and shouldn't that be a direct result of my rate, not affected by probability or other user's rates?
Note: my stales is always about 0.2% so it's not that.
PS. The ads on the site seem to be for real-world mining equipment a lot of the time. You'd get a lot more click-thrus if you could select keywords/ads suitable for tech people, eg. web hosting, amazon gpu deals etc.
Topic: [~1000 GH/sec] BTC Guild - 0% Fee Pool, LP, SSL, Full Precision, and More - page 15. (Read 379076 times)
August 20, 2011, 08:20:14 PM
August 20, 2011, 11:13:59 AM
A few people probably saw funny looking stats in the last 6 hours. Yesterday I put up a 4th pool instance on US West since it has been getting a bit more load than Central. 4th pool worked fine, except I missed the rpcallowip in bitcoin.conf. The pool found a few blocks, but the master server couldn't poll bitcoind them to get the txid/fees included in the block.
This has been resolved. Due to the way the servers handle rounds, no round share data was lost, simply the rounds for found blocks are slightly jumbled in their confirmation counts.
This has been resolved. Due to the way the servers handle rounds, no round share data was lost, simply the rounds for found blocks are slightly jumbled in their confirmation counts.
August 19, 2011, 09:46:55 PM
Had an approximately 2 second block solve, US West didn't catch it in time. I updated the aggression for server sync'ing already, but it still has a 5 second pause if a sync fails. Since the round was so short, US West ended up sync'ing after the round was already over. As usual, if a server shows 0 shares during a round, the calculation script stops for manual inspection in case something really bad happened. When I woke up, I forced it to calculate the round and it caught up with the backlog after a few minutes.
I think this happened again!
I was having the same issue with shares counting backwards, which is why I thought it was a similar issue.
August 19, 2011, 07:11:41 PM
... and it is easy to work out 
Follow my link in my sig: http://tradebtc.net/bitprob.php
Enter the pool hash rate (currently 2068.14 for west)
Then enter a number of hours and it will tell you the chance of failing to find a block for that long.
4 hours gives: 1.5230%
i.e. one in about 65 blocks ...
5 hours gives: 0.5350%
i.e. one in about 187 blocks
Follow my link in my sig: http://tradebtc.net/bitprob.php
Enter the pool hash rate (currently 2068.14 for west)
Then enter a number of hours and it will tell you the chance of failing to find a block for that long.
4 hours gives: 1.5230%
i.e. one in about 65 blocks ...
5 hours gives: 0.5350%
i.e. one in about 187 blocks
August 19, 2011, 06:44:51 PM
Had an approximately 2 second block solve, US West didn't catch it in time. I updated the aggression for server sync'ing already, but it still has a 5 second pause if a sync fails. Since the round was so short, US West ended up sync'ing after the round was already over. As usual, if a server shows 0 shares during a round, the calculation script stops for manual inspection in case something really bad happened. When I woke up, I forced it to calculate the round and it caught up with the backlog after a few minutes.
I think this happened again!
i am not so sure. a 5 hour round is not that unusual when you consider the pool speed and the difficulty.
deepbit is more then double the size and they have 1:30ish rounds which means for us to have some 3 hour
rounds is normal on occasion and a 5 hour round an usual but totally expected event.
August 19, 2011, 06:08:48 PM
Had an approximately 2 second block solve, US West didn't catch it in time. I updated the aggression for server sync'ing already, but it still has a 5 second pause if a sync fails. Since the round was so short, US West ended up sync'ing after the round was already over. As usual, if a server shows 0 shares during a round, the calculation script stops for manual inspection in case something really bad happened. When I woke up, I forced it to calculate the round and it caught up with the backlog after a few minutes.
I think this happened again!
August 19, 2011, 06:06:28 PM
everything looks good here except this monster round we are trying to solve. it really eats into
the 24 hour rewards for a day.
the 24 hour rewards for a day.
August 19, 2011, 03:32:35 PM
ddos? hash rate down ~25-30% across the board
August 19, 2011, 02:51:37 PM
my stales did also go up to 1.9% using uscentral ... dont know what happend the pool.
August 19, 2011, 03:31:04 AM
Been running off of BTC Guild for the last few days, really like the pool, even with my lowly 400Mh/s 
August 18, 2011, 11:53:47 PM
yea was trying to mask location, but not very well. ran ping -t and only got 1 unreachable over 15 mins. trying to think what has changed from when i was getting sub 1% stales. frustrating...
August 18, 2011, 11:32:52 PM
I highly doubt his IP is what you think it is. His Nat router shows up as 1.1. the 2nd hop should be his default gateway at the ISP.
The 3rd hop is that default gateway's next hop.
The 3rd hop is that default gateway's next hop.
That's true, I wasn't paying enough attention, it does reveal geographic proximity though that I believe they were trying to hide by adding xxxx to the fqdn
August 18, 2011, 11:30:37 PM
OK, but whatever you do, don't type "host 76.216.52.2" 
August 18, 2011, 11:27:07 PM
Just did a tracert to uswest and this came back... Doesn't look good to me, but not really sure if the timeouts are causing all of my stales or not (I did xxxx a couple things out). I am on ATT Uverse at 30+ mbps. Could ATT have some network issues that are causing my stales?
Code:
1 <1 ms <1 ms <1 ms 192.168.1.1
2 * * * Request timed out.
3 21 ms 21 ms 22 ms 76-216-52-2.lightspeed.xxxxx.sbcglobal.net [xx.16.52.2]
4 * * * Request timed out.
5 * * * Request timed out.
6 * * * Request timed out.
7 21 ms 21 ms 20 ms 151.164.103.236
8 31 ms 29 ms 29 ms cgcil02jt.ip.att.net [12.122.81.17]
9 47 ms 47 ms 48 ms ae3.chi10.ip4.tinet.net [173.241.128.29]
10 78 ms 79 ms 80 ms ge-7-0-0.lax22.ip4.tinet.net [89.149.185.222]
11 213 ms 203 ms 202 ms aiix-gw.ip4.tinet.net [173.241.129.186]
12 78 ms 78 ms 78 ms aiix-10ge5-1.cr1.lax3.awknet.com [67.220.95.14]
13 79 ms 78 ms 78 ms 69.42.216.173
You do realize that you just told everyone that your IP is 76.216.52.2 right? First of all your hostname from lightspeed has your IP as part of it. Second you should x out the 3rd and 4th, or 2nd, 3rd, and 4th octets of the IP, not the first.
Now as for the timeouts, that is normal, those hosts just don't reply to ICMP packets. Really you need to do a ping, on windows ping -t, let it run for an hour, see what sort of PL you have.
I highly doubt his IP is what you think it is. His Nat router shows up as 1.1. the 2nd hop should be his default gateway at the ISP.
The 3rd hop is that default gateway's next hop.
But what I do agree with is what you are saying about how routers act very odd when you direct traffic right at them instead
of through them. Sometimes using traceroute as a tool to look for loss does not work very well. Best to fire up a ping for 15 minutes
and then look for the percentage loss. If you see a huge amount of loss that is when traceroute works very well to see where things
are stopping. With all that said I would be pissed with a traceroute like that if that was my ISP. My customers would never tolerate
that crap.
And as for what software I use.. i am on win 7 64 bit, guiminer using pocblm and pheonix 1.5 with phatk 2.2. 5830s, a 5850, and 6950s.
August 18, 2011, 10:27:23 PM
Just did a tracert to uswest and this came back... Doesn't look good to me, but not really sure if the timeouts are causing all of my stales or not (I did xxxx a couple things out). I am on ATT Uverse at 30+ mbps. Could ATT have some network issues that are causing my stales?
Code:
1 <1 ms <1 ms <1 ms 192.168.1.1
2 * * * Request timed out.
3 21 ms 21 ms 22 ms 76-216-52-2.lightspeed.xxxxx.sbcglobal.net [xx.16.52.2]
4 * * * Request timed out.
5 * * * Request timed out.
6 * * * Request timed out.
7 21 ms 21 ms 20 ms 151.164.103.236
8 31 ms 29 ms 29 ms cgcil02jt.ip.att.net [12.122.81.17]
9 47 ms 47 ms 48 ms ae3.chi10.ip4.tinet.net [173.241.128.29]
10 78 ms 79 ms 80 ms ge-7-0-0.lax22.ip4.tinet.net [89.149.185.222]
11 213 ms 203 ms 202 ms aiix-gw.ip4.tinet.net [173.241.129.186]
12 78 ms 78 ms 78 ms aiix-10ge5-1.cr1.lax3.awknet.com [67.220.95.14]
13 79 ms 78 ms 78 ms 69.42.216.173
You do realize that you just told everyone that your IP is 76.216.52.2 right? First of all your hostname from lightspeed has your IP as part of it. Second you should x out the 3rd and 4th, or 2nd, 3rd, and 4th octets of the IP, not the first.
Now as for the timeouts, that is normal, those hosts just don't reply to ICMP packets. Really you need to do a ping, on windows ping -t, let it run for an hour, see what sort of PL you have.
August 18, 2011, 10:21:33 PM
Just did a tracert to uswest and this came back... Doesn't look good to me, but not really sure if the timeouts are causing all of my stales or not (I did xxxx a couple things out). I am on ATT Uverse at 30+ mbps. Could ATT have some network issues that are causing my stales?
Code:
1 <1 ms <1 ms <1 ms 192.168.1.1
2 * * * Request timed out.
3 21 ms 21 ms 22 ms 76-216-52-2.lightspeed.xxxxx.sbcglobal.net [xx.16.52.2]
4 * * * Request timed out.
5 * * * Request timed out.
6 * * * Request timed out.
7 21 ms 21 ms 20 ms 151.164.103.236
8 31 ms 29 ms 29 ms cgcil02jt.ip.att.net [12.122.81.17]
9 47 ms 47 ms 48 ms ae3.chi10.ip4.tinet.net [173.241.128.29]
10 78 ms 79 ms 80 ms ge-7-0-0.lax22.ip4.tinet.net [89.149.185.222]
11 213 ms 203 ms 202 ms aiix-gw.ip4.tinet.net [173.241.129.186]
12 78 ms 78 ms 78 ms aiix-10ge5-1.cr1.lax3.awknet.com [67.220.95.14]
13 79 ms 78 ms 78 ms 69.42.216.173
August 18, 2011, 09:42:15 PM
What mining software are you using? I split my miners btw central and west and getting the same results. Any tips el?
August 18, 2011, 09:07:20 PM
@ d.james & fcmatt, which server are you using? I am using central. Thinking of switching a few to west to see if it is a server issue...
uscentral.btcguild.com is what i am using.
August 18, 2011, 08:57:32 PM
@ d.james & fcmatt, which server are you using? I am using central. Thinking of switching a few to west to see if it is a server issue...
August 18, 2011, 08:54:55 PM
What is the deal with all of the stales?
since this morning, I am at 4.5% on 110,000 shares. What kind of stales is everyone else seeing? I used to be at around .6% - .7%. Why does nobody talk on this thread anymore?
since this morning, I am at 4.5% on 110,000 shares. What kind of stales is everyone else seeing? I used to be at around .6% - .7%. Why does nobody talk on this thread anymore?i am at .67% stales when discussing 7 video cards and about 47000 shares. Very reasonable to me.
Jump to: