|
Pool's going down for DB upgrade..
|
|
|
|
It looks like we're running into a mysql bug.. http://bugs.mysql.com/bug.php?id=16708Basically what's happening is a bunch of updates are hanging and everything is grinding to a halt. I just restarted the db.. Hopefully it helps. Otherwise we may be looking at an upgrade to 5.6.6. |
|
|
|
Multiport is dropping out once every minute. This is on the load-balanced server. 2013-07-10 11:01:06: Listener for "multi": [2013-07-10 11:01:06] Stratum connection to pool 0 interrupted 2013-07-10 11:01:16: Listener for "multi": [2013-07-10 11:01:16] Stratum from pool 0 requested work restart 2013-07-10 11:01:41: Listener for "multi": [2013-07-10 11:01:41] Stratum from pool 0 detected new block 2013-07-10 11:02:10: Listener for "multi": [2013-07-10 11:02:08] Stratum connection to pool 0 interrupted 2013-07-10 11:02:15: Listener for "multi": [2013-07-10 11:02:14] Stratum from pool 0 requested work restart 2013-07-10 11:02:24: Listener for "multi": [2013-07-10 11:02:23] Stratum from pool 0 detected new block 2013-07-10 11:03:08: Listener for "multi": [2013-07-10 11:03:08] Stratum connection to pool 0 interrupted 2013-07-10 11:03:14: Listener for "multi": [2013-07-10 11:03:13] Stratum from pool 0 requested work restart 2013-07-10 11:04:10: Listener for "multi": [2013-07-10 11:04:08] Stratum connection to pool 0 interrupted 2013-07-10 11:04:16: Listener for "multi": [2013-07-10 11:04:15] Stratum from pool 0 requested work restart 2013-07-10 11:04:19: Listener for "multi": [2013-07-10 11:04:18] Stratum from pool 0 detected new block 2013-07-10 11:04:54: Listener for "multi": [2013-07-10 11:04:53] Stratum from pool 0 detected new block 2013-07-10 11:05:11: Listener for "multi": [2013-07-10 11:05:11] Stratum connection to pool 0 interrupted 2013-07-10 11:05:18: Listener for "multi": [2013-07-10 11:05:18] Stratum from pool 0 requested work restart 2013-07-10 11:05:32: Listener for "multi": [2013-07-10 11:05:32] Stratum from pool 0 detected ne EDIT: And now it has died completely.  I've had to restart the pools several times in the past hour due to code changes.. |
|
|
|
|
There were some db issues last night causing the stratum disconnects.. I have patched stratum and I think I have the problem sorted out..
|
|
|
|
I mined on the Multiport Stratum server for about 50 minutes total at 740 kh/s yesterday. None of my stats have updated and it currently states that I haven't submitted any shares. I had zero HW errors during this time and maybe two rejects. I haven't had any problems with my other pool and I switched back to it after I noticed that nothing was updating. I'm intrigued by the idea of mining the most profitable coin at the time, but I'd like to be assured that I'm getting credit for mining. It has been over 12 hours and my stats haven't changed from all zeros.
What's your username? |
|
|
|
There was a small hiccup in the server, website was down for me as well as multiport, but seems to be fine again now.
Yeah, right after I said smooth sailing we had an issue with the DB (ofc).. Good news is that I found out that a query option I added to some of the queries a while back when I was a complete SQL noob was the reason for a lot of the stratum disconnects.. So I removed that and it seems to have helped things. |
|
|
|
There is 1 implausible scenario (Mt. Gox official explanation) : Mt. Gox sends so many wires it overwhelmed the 2nd largest bank in Japan's ability to send wires.
There are 2 plausible scenarios, one bad, one worse.
Scenario 1) There is major regulatory pressure that has come down and we are only seeing a bit of it from our vantage point. Dwolla seizure is just the tip. Many banks are refusing to accept wires from Mt.Gox account. I heard from OneFixit on IRC that his wire from Gox to Citibank never showed up, he called Citibank and they said they refused the wire and will refuse all future wires from Gox.
Gox is frantically trying to open accounts in other banks or in other names to disguise the origin of the transfers so it can resume sending wires out. How long before they get caught? This is the bad scenario.
Scenario 2) Gox does not properly segment customer funds, and without regulatory oversight the recent secular shift in market sentiment to more bearish tone has caused a major imbalance in wires in vs. wires out. They don't send wires out because they don't have a segregated pile of customer money. They spent all the money on hiring new people, or expanding, or speculating, who knows. All that matters is they don't wire out till people wire in. Gox is not a bank that has many loans out, where fractional reserve banking works, it is an exchange or brokerage where customer accounts should be segmented. Gox has been in seg fault for weeks and would be instantly shut down if there was any sort of regulatory oversight on their operations. This is the worse scenario.
Take your pick, they both suck. Or take plausible scenario 3, which is 1 + 2 simultaneously.
I suspected scenario 1 several weeks ago when the dwolla seizure happened. The halting of US withdrawals fits well into that scenario. It's possible they are scrambling to comply with KYC requirements, but this shouldn't be an issue for people who have already been validated. |
|
|
|
|
Everything is running on the new pools now.. I moved DNS over earlier. Seems to be smooth sailing since the earlier issues were resolved.
|
|
|
|
When I run a cpu miner (just to get a few extra hashes), I often get something like this: coin-miner 0.64 Copyright (c) 2011-2013 Ufasoft http://ufasoft.com/coin
Mining for stratum+tcp://xxxxx.1:x@pool1.us.multipool.in:3339
Using CPU (3 threads)
New Stratum data with Clean
7/9/2013 3:40:35 PM Result: 4bdeb57c accepted
New Stratum data with Clean
Error 80072746: An existing connection was forcibly closed by the remote host.
New Stratum data with Clean
10.32 kHash/s 3 CPU threads
Any idea why the connection keeps getting closed? The miner keeps running and it's not a significant hasher, but it is a little curious... Cheers! How long ago was this? 7/9/2013 3:40:35 PM (Mountain), but I get it about every time I use either coin-miner or minerd. Having said that, It's at work, so I wouldn't be surprised if there was a firewall issue involved, so it could very well be on my end. Again, not a major complaint, since it's not a big hasher. BTW, thanks for helping out on MNC - it was getting lonely there for a while  I was just testing things  |
|
|
|
I will if can. BTW.. in cgminer, how do you output to a file? I tried the 2>logfile.txt as in the README but it doesn't work. I've looked/read all over and it all says the same thing yet I cannot get it to work ever. Has anyone done this or are doing?
I'll try to Flound, will post asap.
T.
or just email it to me. |
|
|
|
|
Can you post a screenshot?
|
|
|
|
When I run a cpu miner (just to get a few extra hashes), I often get something like this: coin-miner 0.64 Copyright (c) 2011-2013 Ufasoft http://ufasoft.com/coin
Mining for stratum+tcp://xxxxx.1:x@pool1.us.multipool.in:3339
Using CPU (3 threads)
New Stratum data with Clean
7/9/2013 3:40:35 PM Result: 4bdeb57c accepted
New Stratum data with Clean
Error 80072746: An existing connection was forcibly closed by the remote host.
New Stratum data with Clean
10.32 kHash/s 3 CPU threads
Any idea why the connection keeps getting closed? The miner keeps running and it's not a significant hasher, but it is a little curious... Cheers! How long ago was this? |
|
|
|
|
OK, I found the issue.. It's not really related to the load balancer itself, it's more of a problem with the secondary pool server, basically the issue is that the wdc and dgc daemons were killing the cpu on the server, causing the pool servers to disconnect. So when a pool server disconnected, stratum stopped running, resulting in all of the connections migrating to a different server.. Eventually only one server was running and couldn't handle the load.
I've upgraded the secondary pool server to a faster instance which should take care of the issue. I just rebooted all of the pool servers, so your connections should be returning soon.
I think this is also the reason people were getting kicked off when the multiport switched... All of the connections were going to one server and it started dropping people..
|
|
|
|
Dont count money to account for WDC blocks
Say what now? |
|
|
|
7:13 AM (CST) stats hosed again. Miners working but LTC "My Stats" and pool stats down.
BTW Flound, just curious as to why Block payouts do not match to account payout? Block payouts total do not match my actual balance payout. For the last 16 blocks, I added up the individual (all confirmed) blocks pay together and my total confirmed balance is less than my blocks total.
It's not a big amount so really inconsequential but math is supposed to be accurate so..
Be well.
T.
Can you post the actual along with what they were supposed to be? Well.. it must've just been a time 'til actual posting or something. Before when I posted it was off by .08xxx but it is now correct to the digits. No worries, it's all good. Be well. T. It might have been that a block was showing as confirmed in the block list (because that just goes by the # of confirmations) but the payout code hadn't run for that block yet. |
|
|
|
Hi flound1129 I just noticed this quirky behaviour. I'm mining nvc direct on pool1.us and when the multiport switches in and out of nvc, my mining client loses connection and unable to get work from the pool. Why is that? Shouldn't mining direct be uninterrupted?
It does same on LTC. Don't know why but it does and it usually causes several rejects due to when it reconnects the job is now different so "job xxx not current" etc. I just accepted this as part of the pools "quirks" even though it shouldn't be happening. I'm confident Flound will figure it out one day that the multi should somehow?? be on its own and all the coins should be on solo ports themselves. Just my 2 Satoshis Be well. T. P.S. Wakey wakey Flound, time to get up! We need you man. Good Morning West Coast! Yeah, I'm looking at this now.. I agree that shouldn't happen. There are some quirks with both the new LB setup and the backend.. This may be due to me switching the healthcheck. Hope to have them resolved soon. You may see some further disconnects over the course of the next hour or so while I troubleshoot the issue. |
|
|
|
7:13 AM (CST) stats hosed again. Miners working but LTC "My Stats" and pool stats down.
BTW Flound, just curious as to why Block payouts do not match to account payout? Block payouts total do not match my actual balance payout. For the last 16 blocks, I added up the individual (all confirmed) blocks pay together and my total confirmed balance is less than my blocks total.
It's not a big amount so really inconsequential but math is supposed to be accurate so..
Be well.
T.
Can you post the actual along with what they were supposed to be? |
|
|
|
DCG direct appears to be down. Haven't tried any of the other directs yet...
MNC is working ok
DGC up now |
|
|
|
2013-07-08 22:05:38: Listener for "multi": [2013-07-08 22:05:37] Pool 0 stratum+tcp://pool1.us.multipool.in:7777 not responding!
switched back to original multiport and it is working.
Yeah there was just a hiccup.. I was changing some stuff around on the load balancer. Should be ok now. |
|
|
|
|
Show Posts
