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

[eleuthria]

The Stratum pool finally crashed again, and I've found the piece of code that caused it.  I'll be working on a fix tonight and hopefully deploy it tomorrow.  Downtime wasn't long, and the pool is back up and running while I work on the fix.


EDIT/UPDATE:  I've restarted the primary stratum server again, with a fix deployed for the crash.  It's going to run in debug mode for the next month, but that shouldn't affect user performance in any way.

[1714018366]

1714018366

[1714018366]

1714018366

[1714018366]

1714018366

[1714018366]

1714018366

[eleuthria]

There were 2 unplanned restarts on Stratum today, to do final cleanup on the previous crash bug.  Sorry for the inconvenience, but I wanted to get these fixes out ASAP since I've been pretty sick the last few days and wasn't sure how fast I'd be able to restart the servers if they crashed in the middle of the night.

[eleuthria]

The Stratum information page (Stratum Protocol under 'Help and Support) has been updated to be less verbose.  It now talks less about internals of Stratum and more about the general concepts.  Information on how to utilize Stratum is now split into 3 sections: Native Support, Windows Proxy, Python Proxy.

Users who have not recently utilized the Stratum pools will see a link at the top of the page about switching to the new protocol.

[eleuthria]

The Stratum server will be having a planned reboot on Sunday (November 4th) at noon (Pacific time).  The restart is expected to last approximately 5 seconds.  This should be the last restart prior to the reward being halved at the end of November.

[eleuthria]

It's that time of year again in the US.  Daylight Savings Time appears to be providing very funky numbers for Hall of Fame speed estimations/pool speed estimation.  It should return to normal after 2 hours.

[organofcorti]

It's that time of year again in the US.  Daylight Savings Time appears to be providing very funky numbers for Hall of Fame speed estimations/pool speed estimation.  It should return to normal after 2 hours.

I was wondering about that 9Thps pool speed

[eleuthria]

It's that time of year again in the US.  Daylight Savings Time appears to be providing very funky numbers for Hall of Fame speed estimations/pool speed estimation.  It should return to normal after 2 hours.

I was wondering about that 9Thps pool speed

You weren't the only one wondering!  Saw a couple blocks hit on blockchain.info, thought maybe luck was finally going to turn positive after the 2 month long slump, and saw 9 TH/s!  Then I knew something was wrong because it was saying 8.5 TH/s of it was on the getwork pool ;p.

[dmcurser]

ok i guess i should just ask this here. I am currently using bamt latest version .5 i think i am woundering if i have to do any updating to it to be able to mine on the pool i pointed my quad miner twords the pool earlyer and it seemed i only mined lightcoins so any help would be much apperciated.

[eleuthria]

ok i guess i should just ask this here. I am currently using bamt latest version .5 i think i am woundering if i have to do any updating to it to be able to mine on the pool i pointed my quad miner twords the pool earlyer and it seemed i only mined lightcoins so any help would be much apperciated.

You wouldn't be able to mine litecoins with BTC Guild...sounds like you weren't pointing it here?  You shouldn't have to do anything special to point an FPGA at the pool (I run my Ztex miners with cgminer).

[dmcurser]

i didnt mean lite i ment name coins
if i look at my user stats it shows accepted share on the name coin side not on the btc side
im a point just my quad miner there in a min i was just asking if the software in bamt needs to be updated or if it was ok to use with your pool
i belive for the miner im using bfgminer progrtam



hmm nvm edit seems to be showing now
but still wanting to know if bamt software is up to date enough lol

[eleuthria]

Payouts will be disabled for about 10 minutes after the 1 PM (PST) automatic payout is processed.  I need to replace the wallet.dat which has become so heavily fragmented that roughly 5 payouts per day have been timing out waiting for bitcoind to respond with the txid.  The payments always go through, but it's requiring me to make daily checks to catch any payments missing a txid (and unfreeze the account it affected).

After payouts are restarted, the pool servers will have rolling restarts.  Stratum based servers should only be down for about 2 seconds, getwork pools may take 1-2 minutes.

UPDATE (1:23 PM PST): Servers were all restarted.  The new payout wallet is being refilled with coins from the old wallet.  Automatic payouts will process at 2 PM.  Manual payouts will be re-enabled once the old wallet has been emptied into the new wallet (approx 30 minutes).

UPDATE (1:50 PM PST): Manual payouts are enabled again.  Will be doing some work to make the next switch a bit easier.

[iCEBREAKER]

I need to replace the wallet.dat which has become so heavily fragmented that roughly 5 payouts per day have been timing out waiting for bitcoind to respond with the txid.  

What about putting wallet.dat on its own partition, for faster/less defrag?

[eleuthria]

I need to replace the wallet.dat which has become so heavily fragmented that roughly 5 payouts per day have been timing out waiting for bitcoind to respond with the txid.  

What about putting wallet.dat on its own partition, for faster/less defrag?

It's not hard drive fragmentation.  The wallet continually gets cluttered up by new addresses for change, as well as all the transactions.  It takes a few months, but eventually it becomes unusable.  Not a problem for everyday users, big problem for pools where there are constant payouts.  I could clean it up a little by patching it to always send change back to the original address, but that is only a bandaid in my experience.  The problem will still eventually happen.

[eleuthria]

I've started a poll regarding Merged Mining (Namecoins).  As I stated recently in slush's thread, the namecoin project feels like it's abandonware.  Very little work has been done since merged mining was released last year, usage is not picking up, and the value of NMC continues to become a smaller and smaller percentage of income.  I'm likely going to drop NMC merged mining and stick purely to BTC unless there is a demand for it to remain available.  In that case, I may setup a special getwork-only enabled server for people who want to continue mining NMC.

[dmcurser]

just wondering would there be any interest or ability to merge mine light coin?

[eleuthria]

just wondering would there be any interest or ability to merge mine light coin?

Litecoin can't be merge mined.  Different algorithm for hashing.

[iCEBREAKER]

Where's the 'I hate Namecoin and want it dead' option?   

Just kidding, I fully support the dot.bit project and think it will eventually find its killer application.

[dmcurser]

ok just figured i might ask

[beekeeper]

Man, I think you have a very nasty bug in workers management.
If I use de server (176.9.42.247), and I have one worker working (my client connected to it and mining), and I change its name, then hell breaks loose. The old worker will still be available and new worker name wont be usable anymore (anything I send trough it isnt recorded), even if I change it later, create a new worker with this new name after I freed the name. More, any worker which was somehow in contact with this scheme wont be usable anymore. Check user 41599, worker with extension _h3m5.

[eleuthria]

Man, I think you have a very nasty bug in workers management.
If I use de server (176.9.42.247), and I have one worker working (my client connected to it and mining), and I change its name, then hell breaks loose. The old worker will still be available and new worker name wont be usable anymore (anything I send trough it isnt recorded), even if I change it later, create a new worker with this new name after I freed the name. More, any worker which was somehow in contact with this scheme wont be usable anymore. Check user 41599, worker with extension _h3m5.

Old workers will still be available and still work because the server caches them.  Similarly, BAD worker logins are cached.  You obviously renamed your worker but tried to connect before renaming it, causing it to get added to the bad worker list.  The server caches bad logins to prevent somebody from flooding the server with bad logins and forcing constant MySQL lookups to bog the server down.  Old mining software used to reconnect with no delay when a connection failed, which resulted in flooding the pool server -and- database server due to the worker lookups.

If you were using the DE Stratum server (which can be connected to by ANY stratum supporting client by just pointing them at de.btcguild.com), this problem would only persist for about a minute before the bad worker cache was cleared.  I don't recall how often the getwork pool clears its worker cache.  At this point the getwork pool is barely supported, and simply left online while people upgrade to stratum mining clients.

Either way, this only occurs because you were trying to connect to the pool using invalid credentials initially, THEN renamed the worker later, leaving the old name in the bad worker cache.