[CLOSED] BTC Guild - Pays TxFees+NMC, Stratum, VarDiff, Private Servers - page 215.

os2sam

legendary

Activity: 3583

Merit: 1094

Think for yourself

Quote from: Icon on October 13, 2013, 08:17:04 PM

Quote from: os2sam on October 13, 2013, 07:38:43 PM

Quote from: mdopro1 on October 13, 2013, 07:28:23 PM

20% is quite high but I'll still look into it and see how it works for a few days.

That is way higher than my experience. My problem with load balancing was that the split was very uneven. But now with the new quota method it may be better.

Ah not sure of the % i mine at 16 gh/s and lost 2 gh/s doing load-balance, and the fact the usb miner leds come on and off and stay on for a sec then go back mining. Like they where waiting on something...

I haven't load balanced since my GPU days. It's just too easy to setup multiple instances now with the usb devices now I just have no need to load balance. But when I did it with my GPU's my stales increased, but no where near 20%. Maybe 1 or 2%.

Icon

hero member

Activity: 821

Merit: 503

Quote from: os2sam on October 13, 2013, 07:38:43 PM

Quote from: mdopro1 on October 13, 2013, 07:28:23 PM

20% is quite high but I'll still look into it and see how it works for a few days.

That is way higher than my experience. My problem with load balancing was that the split was very uneven. But now with the new quota method it may be better.

Ah not sure of the % i mine at 16 gh/s and lost 2 gh/s doing load-balance, and the fact the usb miner leds come on and off and stay on for a sec then go back mining. Like they where waiting on something...

Also something you can try if you want in bfgminer with no over head is manually set the com ports for each instance of bfgminer to use. Like run 2 instance of bfgminer and set one to mine for btc guild other for eclipse and use the format: -S erupter:\\.\com42 -S erupter:\\.\com44. basically you tell each com port which pool to mine for.. Works really well and 0 overhead just have to run 2 instance of bfgminer and you can even get back your failover pools

and its hard set the hash rate 50/50 or ever how many com ports you assign to each pool.

os2sam

legendary

Activity: 3583

Merit: 1094

Think for yourself

Quote from: mdopro1 on October 13, 2013, 07:43:42 PM

Quote from: os2sam on October 13, 2013, 07:38:43 PM

Quote from: mdopro1 on October 13, 2013, 07:28:23 PM

20% is quite high but I'll still look into it and see how it works for a few days.

That is way higher than my experience. My problem with load balancing was that the split was very uneven. But now with the new quota method it may be better.

What quota method?

You can set a hash rate quota per pool. Such as set one pool to 30% another to 20% and another to 50%. Something like that. Haven't really looked into it myself yet. It's the the CGMiner readme I'm sure.

mdopro1

sr. member

Activity: 308

Merit: 250

Double your Personal Bitcoin Funds.

Quote from: os2sam on October 13, 2013, 07:38:43 PM

Quote from: mdopro1 on October 13, 2013, 07:28:23 PM

20% is quite high but I'll still look into it and see how it works for a few days.

That is way higher than my experience. My problem with load balancing was that the split was very uneven. But now with the new quota method it may be better.

What quota method?

os2sam

legendary

Activity: 3583

Merit: 1094

Think for yourself

Quote from: mdopro1 on October 13, 2013, 07:28:23 PM

20% is quite high but I'll still look into it and see how it works for a few days.

That is way higher than my experience. My problem with load balancing was that the split was very uneven. But now with the new quota method it may be better.

mdopro1

sr. member

Activity: 308

Merit: 250

Double your Personal Bitcoin Funds.

Thanks guys, I'll look into that. I've always wanted to mine in two pools but splitting hashing power manually is too much work. 20% is quite high but I'll still look into it and see how it works for a few days.

Icon

hero member

Activity: 821

Merit: 503

Yep --balance --load-balance options, thing is it basically connecting and reconnecting to the servers you add, now you loose failover option when doing load balance or balance options and there is some hashing power you'll loose due to the over head of constant connects and reconnects. about 20% from what i have tested.

os2sam

legendary

Activity: 3583

Merit: 1094

Think for yourself

Quote from: mdopro1 on October 13, 2013, 06:44:26 PM

Can BFGMiner connect to two pools at the same time? I didn't know that.

CGMiner can so I suppose BFGMiner maintained that capability when it was forked. That is one of their main strengths is to have multiple pools for failover, balancing or one of several other mining strategies.

mdopro1

sr. member

Activity: 308

Merit: 250

Double your Personal Bitcoin Funds.

Quote from: eleuthria on October 13, 2013, 12:00:50 PM

There were some reports from users with bfgminer using startup scripts that immediately connect to multiple pools NOT using failover-only getting messages related to the pool giving them old/stale blocks on stratum.btcguild.com . Last night I pushed out an update to the code on that server which hopefully fixes the problem. This error was related to how the server was changed to filter/validate miners to separate botnets/DDoS traffic from legitimate users.

If you were having this problem previously, please let me know if it is still happening. I have not patched the eu-stratum.btcguild.com server with this update since I want to make sure it works first, but I have so far been unable to duplicate the issue myself.

Can BFGMiner connect to two pools at the same time? I didn't know that.

bobitza

sr. member

Activity: 560

Merit: 256

Quote from: eleuthria on October 13, 2013, 12:00:50 PM

If you were having this problem previously, please let me know if it is still happening.

It connected right away now. So whatever you did, it worked. Thanks.

eleuthria

legendary

Activity: 1750

Merit: 1007

There were some reports from users with bfgminer using startup scripts that immediately connect to multiple pools NOT using failover-only getting messages related to the pool giving them old/stale blocks on stratum.btcguild.com . Last night I pushed out an update to the code on that server which hopefully fixes the problem. This error was related to how the server was changed to filter/validate miners to separate botnets/DDoS traffic from legitimate users.

If you were having this problem previously, please let me know if it is still happening. I have not patched the eu-stratum.btcguild.com server with this update since I want to make sure it works first, but I have so far been unable to duplicate the issue myself.

demonmaestro

hero member

Activity: 574

Merit: 500

Mining for the hell of it.

just sounds like a startup procedure to me

Side note: DONT POST THE URL.

bobitza

sr. member

Activity: 560

Merit: 256

I keep getting "Stratum from pool 0 is requesting work update" messages when trying to connect to the pool, instead of shares. I can see I get a first share (probably the bot checking) and then the pool gives me a new URL and then some message that the pool is issuing work for an old block and then the work updates. Any idea about what I'm doing wrong?

eleuthria

legendary

Activity: 1750

Merit: 1007

Quote from: PatMan on October 12, 2013, 08:01:07 PM

Quote from: eleuthria on October 12, 2013, 07:23:55 PM

Quote from: PatMan on October 12, 2013, 06:47:58 PM

Website doing funky things atm, not loading correctly - part of the new transition?

Should be fixed now. If the page has virtually no styling to it, clear your cache (or hold CTRL and hit F5 to do a force reload).

You got it my man - working Grin

Glad to hear it

. The reverse proxy was setup to cache all the static files locally, but it wasn't redirecting the CSS imports properly so outside of the main CSS file nothing was styled. I changed the links slightly so your browser *should* automatically get the corrected version now even if you had the bad version cached.

PatMan

hero member

Activity: 924

Merit: 1000

Watch out for the "Neg-Rep-Dogie-Police".....

Quote from: eleuthria on October 12, 2013, 07:23:55 PM

Quote from: PatMan on October 12, 2013, 06:47:58 PM

Website doing funky things atm, not loading correctly - part of the new transition?

Should be fixed now. If the page has virtually no styling to it, clear your cache (or hold CTRL and hit F5 to do a force reload).

You got it my man - working Grin

eleuthria

legendary

Activity: 1750

Merit: 1007

Quote from: PatMan on October 12, 2013, 06:47:58 PM

Website doing funky things atm, not loading correctly - part of the new transition?

Should be fixed now. If the page has virtually no styling to it, clear your cache (or hold CTRL and hit F5 to do a force reload).

PatMan

hero member

Activity: 924

Merit: 1000

Watch out for the "Neg-Rep-Dogie-Police".....

Website doing funky things atm, not loading correctly - part of the new transition?

eleuthria

legendary

Activity: 1750

Merit: 1007

The website speed should see a decent speed boost once your DNS updates. Last night at the start of the server move, an Amazon EC2 small instance was being used with iptables to simply proxy requests to the real server (keep it protected from attack) to get off of Cloudflare ASAP. Now migrating DNS to a dedicated box using a more sophisticated reverse proxy setup to speed up the general navigation. It was never obscenely slow, but there was definitely a minimum ~150ms delay added to each page load due to the EC2 proxy. The new server should reduce that delay by a reasonable margin.

This should not trigger SSL complaints like the previous switch. The new reverse proxy responds with the same hostname and SSL certificate as before. Hopefully by now most/all users are no longer seeing a Cloudflare issued certificate due to the slowness of nameserver updating.

eleuthria

legendary

Activity: 1750

Merit: 1007

Quote from: -Redacted- on October 12, 2013, 05:18:53 PM

Quote from: eleuthria on October 12, 2013, 03:32:21 PM

New attack detection systems have been added now which have stopped the attacker from successfully determining if their attempts are valid usernames or not.

At the rate you're making security fixes, your website probably has better security now than NASA....

Hah. Luckily to date no security fix has been vital, it's mostly about limiting the effectiveness of brute forcing by either eliminating methods to reduce entropy (finding usernames via password reset attempts) or eliminating the ability to use a single machine for more than a dozen or so attempts. Same thing with the recent changes to the Stratum frontend where it's validating legitimate miners over botnets/attackers. It wasn't vital, but it severely impacted the efficiency of the future attacks as well as minimized the actual damage they did.

-Redacted-

hero member

Activity: 574

Merit: 501

Quote from: eleuthria on October 12, 2013, 03:32:21 PM

New attack detection systems have been added now which have stopped the attacker from successfully determining if their attempts are valid usernames or not.

At the rate you're making security fixes, your website probably has better security now than NASA....

Topic: [CLOSED] BTC Guild - Pays TxFees+NMC, Stratum, VarDiff, Private Servers - page 215. (Read 903150 times)