[CLOSED] BTC Guild - Pays TxFees+NMC, Stratum, VarDiff, Private Servers - page 88.

eleuthria

legendary

Activity: 1750

Merit: 1007

Quote from: BrandonMcPherson on April 28, 2014, 07:11:48 PM

I noticed and considered posting, but my superstitious side said "Don't! You'll jinx it."

It seems, BTW, that the new graphing format makes the trends look starker by reducing the number of spikes in either direction. That's an observation, not a complaint. Improved responsiveness for mobile users is a Good Thing™. I look forward to seeing a positive luck streak not dip below the neutral line, too. If we're lucky (badaboom!), it will be a multi-day 125% luck streak to balance out this stomach puncher we're enduring right now.

Yes, the new system is 3 months of luck, but each data point is the combination of 3 shifts. It does smooth out the graph, in the same way our high 'N' value on PPLNS smooths out per block luck you would see on Prop/Score/DGM (you'll never see a shift at 300-2000% luck, and you'll likely never see one at 1-10% luck either).

BrandonMcPherson

full member

Activity: 205

Merit: 100

I noticed and considered posting, but my superstitious side said "Don't! You'll jinx it."

It seems, BTW, that the new graphing format makes the trends look starker by reducing the number of spikes in either direction. That's an observation, not a complaint. Improved responsiveness for mobile users is a Good Thing™. I look forward to seeing a positive luck streak not dip below the neutral line, too. If we're lucky (badaboom!), it will be a multi-day 125% luck streak to balance out this stomach puncher we're enduring right now.

Edit: And yay for crossing the 9 PH/s mark. it's good to gain a little share back.

eleuthria

legendary

Activity: 1750

Merit: 1007

Quote from: guytechie on April 28, 2014, 06:51:37 PM

This luck is a killer.

Nobody felt like pointing it out when 1D/3D/1W were over 100% last week Sad

. Think it was on the 24th that all 3 were > 100%, which we haven't seen in a while. Still not enough to make up for the bad.

guytechie

hero member

Activity: 677

Merit: 500

This luck is a killer.

Taugeran

hero member

Activity: 658

Merit: 500

CCNA: There i fixed the internet.

Quote from: eleuthria on April 26, 2014, 07:52:31 PM

Quote from: MoreBloodWine on April 26, 2014, 07:47:48 PM

Quote from: eleuthria on April 26, 2014, 07:44:29 PM

Quote from: MoreBloodWine on April 26, 2014, 07:21:38 PM

Hey Eleu, any idea when you might be able to do something about renaming workers to where it doesn't really "apply" unless the pools reset.

I renamed a couple of workers and the one running work I have right now is running on a "name" it's not supposed to be.

I guess ATM all I can hope for is an unexpected pool reset like when you removed the PPS stuff.

This will be resolved the next time the stratum servers are restarted. Unfortunately, right now it's not a high priority since this problem only happens when people rename workers to a previously used name...which really doesn't happen often. If it really bugs you...name them something you haven't used before, and the problem will go away.

If there is a reason to restart the stratum servers, the issue will be gone. Otherwise they won't be restarted until something else comes along that warrants restarting them.

Not really a OMG I'm gonna throw a fit bother but minor annoyance.

The renaming was more about having them in a specific order, I guess I could say alphabetically even though there's like 1 out of place ein that scenario.

I do recall you at one point talking about trying to "patch" that when you have a bit of free time or something heh.

Well, the patch is *done*, and compiled. The problem is that patch can only be deployed by restarting the server process, which I do not do often due to some miners not handling disconnects gracefully. Since this really is a cosmetic issue more than anything, caused by a non-standard renaming case, it's not something that warrants restarting the server process on its own.

It looks like bfgminer is going to support SSL/TLS stratum soon, so I'm going to start working on an SSL/TLS implementation for the pool server. That change will be large enough to warrant restarting the servers.

HTTPS stratum?

eleuthria

legendary

Activity: 1750

Merit: 1007

Quote from: MoreBloodWine on April 26, 2014, 07:47:48 PM

Quote from: eleuthria on April 26, 2014, 07:44:29 PM

Quote from: MoreBloodWine on April 26, 2014, 07:21:38 PM

Hey Eleu, any idea when you might be able to do something about renaming workers to where it doesn't really "apply" unless the pools reset.

I renamed a couple of workers and the one running work I have right now is running on a "name" it's not supposed to be.

I guess ATM all I can hope for is an unexpected pool reset like when you removed the PPS stuff.

This will be resolved the next time the stratum servers are restarted. Unfortunately, right now it's not a high priority since this problem only happens when people rename workers to a previously used name...which really doesn't happen often. If it really bugs you...name them something you haven't used before, and the problem will go away.

If there is a reason to restart the stratum servers, the issue will be gone. Otherwise they won't be restarted until something else comes along that warrants restarting them.

Not really a OMG I'm gonna throw a fit bother but minor annoyance.

The renaming was more about having them in a specific order, I guess I could say alphabetically even though there's like 1 out of place ein that scenario.

I do recall you at one point talking about trying to "patch" that when you have a bit of free time or something heh.

Well, the patch is *done*, and compiled. The problem is that patch can only be deployed by restarting the server process, which I do not do often due to some miners not handling disconnects gracefully. Since this really is a cosmetic issue more than anything, caused by a non-standard renaming case, it's not something that warrants restarting the server process on its own.

It looks like bfgminer is going to support SSL/TLS stratum soon, so I'm going to start working on an SSL/TLS implementation for the pool server. That change will be large enough to warrant restarting the servers.

MoreBloodWine

legendary

Activity: 1064

Merit: 1001

Quote from: eleuthria on April 26, 2014, 07:44:29 PM

Quote from: MoreBloodWine on April 26, 2014, 07:21:38 PM

Hey Eleu, any idea when you might be able to do something about renaming workers to where it doesn't really "apply" unless the pools reset.

I renamed a couple of workers and the one running work I have right now is running on a "name" it's not supposed to be.

I guess ATM all I can hope for is an unexpected pool reset like when you removed the PPS stuff.

This will be resolved the next time the stratum servers are restarted. Unfortunately, right now it's not a high priority since this problem only happens when people rename workers to a previously used name...which really doesn't happen often. If it really bugs you...name them something you haven't used before, and the problem will go away.

If there is a reason to restart the stratum servers, the issue will be gone. Otherwise they won't be restarted until something else comes along that warrants restarting them.

Not really a OMG I'm gonna throw a fit bother but minor annoyance.

I do recall you at one point talking about trying to "patch" that when you have a bit of free time or something heh.

eleuthria

legendary

Activity: 1750

Merit: 1007

Quote from: MoreBloodWine on April 26, 2014, 07:21:38 PM

Hey Eleu, any idea when you might be able to do something about renaming workers to where it doesn't really "apply" unless the pools reset.

I renamed a couple of workers and the one running work I have right now is running on a "name" it's not supposed to be.

I guess ATM all I can hope for is an unexpected pool reset like when you removed the PPS stuff.

This will be resolved the next time the stratum servers are restarted. Unfortunately, right now it's not a high priority since this problem only happens when people rename workers to a previously used name...which really doesn't happen often. If it really bugs you...name them something you haven't used before, and the problem will go away.

If there is a reason to restart the stratum servers, the issue will be gone. Otherwise they won't be restarted until something else comes along that warrants restarting them.

MoreBloodWine

legendary

Activity: 1064

Merit: 1001

Hey Eleu, any idea when you might be able to do something about renaming workers to where it doesn't really "apply" unless the pools reset.

I renamed a couple of workers and the one running work I have right now is running on a "name" it's not supposed to be.

I guess ATM all I can hope for is an unexpected pool reset like when you removed the PPS stuff.

Sir Alan

full member

Activity: 221

Merit: 100

Quote from: ?? on ??

What irritates me is, that I have problems only to BTC-Guild.

It's not uncommon for only one particular site to be inaccessible.

If you have a helpful neighbour, ask if you can take your pc to their house and hook up on their internet connection. If you can get through to BTCguild, then you will at least know it's not your pc.

If you are using a wireless connection, try connecting with a cable instead. You could also try (temporarily) changing the security level on your router: if it's on WPA2 try changing to WEP - this has even been known to fix a problem when the pc was connected by cable, even though it really should have no effect!

hurricandave

legendary

Activity: 966

Merit: 1003

May be an issue with recent updates to your anti-virus software. AVG does not play nice with mining at times as well as Windows Defender.

os2sam

legendary

Activity: 3583

Merit: 1094

Think for yourself

Quote from: Sir Alan on April 26, 2014, 03:53:26 AM

If you then still get the timeout messages on TRACERT, you may need to contact your ISP.

More routers and/or servers are starting to not reply to ICMP packets. Ping and Traceroute use ICMP packets. That's becoming normal behavior these days for security reasons.

Taugeran

hero member

Activity: 658

Merit: 500

CCNA: There i fixed the internet.

Quote from: Sir Alan on April 26, 2014, 03:53:26 AM

Quote from: guytechie on April 26, 2014, 01:32:13 AM

Maybe there is a bad hop between you and the pool server. That's just how the internet works. It may heal on its own (finds a better route), or whoever administers that particular hop notices and fixes the problem.

Have you tried cold re-booting your router? Disconnect the power for 30 seconds, then re-connect. I used to be in tech support for online banking, and this was our usual fix for people who were unable to connect to their banking service - worked almost every time. If you then still get the timeout messages on TRACERT, you may need to contact your ISP.

+1

Sir Alan

full member

Activity: 221

Merit: 100

Quote from: guytechie on April 26, 2014, 01:32:13 AM

Maybe there is a bad hop between you and the pool server. That's just how the internet works. It may heal on its own (finds a better route), or whoever administers that particular hop notices and fixes the problem.

Have you tried cold re-booting your router? Disconnect the power for 30 seconds, then re-connect. I used to be in tech support for online banking, and this was our usual fix for people who were unable to connect to their banking service - worked almost every time. If you then still get the timeout messages on TRACERT, you may need to contact your ISP.

guytechie

hero member

Activity: 677

Merit: 500

Quote from: ?? on ??

Quote from: eleuthria on April 25, 2014, 10:16:20 PM

EU-Stratum rejects pings, so that isn't unexpected. But as I stated before: This isn't a pool problem. There is nothing I can do on my end because there is nothing wrong on my end.

I do not blame you.
I'm just trying to figure out why it only affects your pool.

Maybe there is a bad hop between you and the pool server. That's just how the internet works. It may heal on its own (finds a better route), or whoever administers that particular hop notices and fixes the problem.

As a network admin, we see this happen before. As Eleuthera says, it's out of our hands.

eleuthria

legendary

Activity: 1750

Merit: 1007

EU-Stratum rejects pings, so that isn't unexpected. But as I stated before: This isn't a pool problem. There is nothing I can do on my end because there is nothing wrong on my end.

Minor Miner

legendary

Activity: 2478

Merit: 1020

Be A Digital Miner

Quote from: ?? on ??

Quote from: eleuthria on April 25, 2014, 12:26:19 PM

The problem isn't the pool. It's likely a bad route between you and the server, likely at your ISP since there have not been massive drops in pool speed indicating a wide spread regional issue.

Funny thing is, I mine for months / years with no problems at BTC Guild. And since 3 days only disconnections to both server.
Is there a way to test the ISP?
At other pools (eg multipool) I have no problem
EU-Server and US-Server are both exactly at the same time dead.
But eg multipool is still alive Huh

I'm stumped.

Do you have a firewall software that may have recently updated? We use Kerio and when it updated a few months ago it kept dropping miners off of BTC Guild until we fixed it.

skleven

sr. member

Activity: 281

Merit: 250

The only thing wrong is bad luck... 5hours since last block now...

eleuthria

legendary

Activity: 1750

Merit: 1007

The problem isn't the pool. It's likely a bad route between you and the server, likely at your ISP since there have not been massive drops in pool speed indicating a wide spread regional issue. A 400 TH/s swing is virtually nothing against a 8 PH/s pool.

3 captures from users that are in different regions.

Testing123

hero member

Activity: 561

Merit: 500

Quote from: BrandonMcPherson on April 25, 2014, 03:38:55 AM

Lost a good chunk of a petahash today?

The pool hashrate is pretty volatile this week. Looks like some big miners came and gone, and came and gone again.

Topic: [CLOSED] BTC Guild - Pays TxFees+NMC, Stratum, VarDiff, Private Servers - page 88. (Read 903163 times)