bigb159
|
|
April 18, 2013, 10:19:23 AM |
|
Thanks for getting this back up, Slush!
Now I have a minuscule amount of coin tied up in bitminter. I may have to mine there until I hit their threshold.
I've been looking around: how does one set up poclbm to fall back on a backup pool when default pool connections fail?
|
|
|
|
ninjaboon
Legendary
Offline
Activity: 2128
Merit: 1002
|
|
April 18, 2013, 10:34:13 AM |
|
Thanks for the info, I'll never choose OVH as my server hosting partner.
|
|
|
|
PetrovMichael
Newbie
Offline
Activity: 34
Merit: 0
|
|
April 18, 2013, 10:55:37 AM |
|
before APR 16 ... time [73ms] [78ms]... today [213ms]...[234ms]... Good point... I'm preparing smart routing algorithm, which will automatically move your miner to the closest pool endpoint. I'm keeping all this very seriously... Slush, thank you ! i'm still with you, even i see when timing (the max i saw) [2108ms] (or [1748ms] or [1320ms]) share still accepted...
|
|
|
|
OlgaA524
Newbie
Offline
Activity: 54
Merit: 0
|
|
April 18, 2013, 12:04:14 PM |
|
Thank you for your tremendous efforts Slush! get some rest
|
|
|
|
tnkflx
|
|
April 18, 2013, 01:08:50 PM |
|
just the question... using mining_proxy-1.5.2 without any optional arguments... before APR 16 i saw access time [73ms] [78ms]... today [213ms]...[234ms]... is it important ?
Good point. Servers were located in OVH/France until yesterday. I move machines to OVH from Linode after previous DDoS attacks year ago. These days, sales representative at OVH promised me that they'll actively help me with DDoS attacks, because they have great infrastructure and skilled admins. The reality of recent days was that they were even worse than Linode guys, because at Linode I was able to create multiple machines dynamically and mitigate DDoS by moving pool around many IPs. In OVH I had multicore machines with dozen of IPs, and I was prepared to mitigate DDoS by filtering the traffic by strong hardware and by routing traffic all around many of private IPs. But. These idiots at OVH blocked ALL these IPs at once since attacks started, claiming that the attack is affecting other customers. So I paid overpriced infrastructure for many months just for *nothing*. The only "solution" they offered me were Cisco firewalls, able to filter DDoS up to throughtput of 400 Mbit/s (excuse me??). When I rejected this offer, they simply rebooted my machines to "rescue mode", telling me that my servers are damaging their infrastructure and asking me to move my machines away. So - never ever OVH. Yesterday I migrated the pool to Amazon EC2, that's why the pool was completely offline for some time. I selected datacentres in USA, because they have the best infrastructure so far, and that's why you see higher latencies than "usual". But this is just a temporary problem, I'm configuring more machines around all Amazon datacentres (USA, Europe, Asia, ...) and I'm preparing smart routing algorithm, which will automatically move your miner to the closest pool endpoint. I'm keeping all this very seriously, but it is hard to mitigate DDoS when you depend on group of impotent idiots at OVH which are able to send you only prefabricated template reports "your server is having issues, we restarted it / null routed all IPs". Is cloudfare a viable solution?
|
| Operating electrum.be & us.electrum.be |
|
|
|
tarrant_01
|
|
April 18, 2013, 01:20:51 PM |
|
I noticed that I am getting the message that some of my workers are using the deprecated "getwork" protocol but all my workers are using cgminer 2.11.3 (one down from the current 2.11.4). The message suggests that I use cgminer 2.8.x or higher. What else could be wrong?
|
1P95gCUCw3Tjb7yyoYtW3ARZZQyTpFgk6H
|
|
|
DoomDumas
Legendary
Offline
Activity: 1002
Merit: 1000
Bitcoin
|
|
April 18, 2013, 01:24:10 PM |
|
I nominate Slush for pool operator of the year!
I vote for Slush ! Keep up the excellent work, and updating user like you do is really really appreciated.. I'll always mine on Slush's pool !
|
|
|
|
Hippie Tech
aka Amenstop
Legendary
Offline
Activity: 1624
Merit: 1001
All cryptos are FIAT digital currency. Do not use.
|
|
April 18, 2013, 01:31:06 PM |
|
Very suspicious behaviour by OVH. I smell a false flag.
Demand a refund in BTC !!
|
|
|
|
mcdougrs
Newbie
Offline
Activity: 15
Merit: 0
|
|
April 18, 2013, 02:32:23 PM |
|
That is potentially embezzlement; a felony in many countries. Have fun with that....
Not if I have permission.
|
|
|
|
Synapse_On_Slush
Newbie
Offline
Activity: 22
Merit: 0
|
|
April 18, 2013, 02:50:52 PM |
|
Since the DDoS I've had problems getting my 7970 ATI mining backup to it's ~614 MH/s I had using GUIMiner with cgminer.exe on stratum.
I can't remember how I got cgminer.exe working with GUIMiner, when I try to add it now it says that cgminer.exe is not supported and refuses. Very confused as I've not updated since January 2013 and it obviously allowed me to add it back then...
Anyway, my problem is - I'v now tried just going with cgminer.exe straight from command line but I'm only getting 575 MH/s... Any idea why it's slower going direct with cgminer than using it through GUIMiner?
I've had to use cgminer directly as GUIMiner has (apparently) started using getwork (according to my slush pool account page). I tried telling guiminer the stratum.bitcoin.cz with port 3333 but it just kept saying "starting..." and nothing more. So I figure GUIMiner is temporarily broken/forced to using getwork so I'll use the (confusigly) slower cgminer.exe direct from command line.
Anyone else noticed this? or am I just odd having been using GUIMiner to run cgminer.exe all this time? :/ It was definitely using stratum before the DDoS.
|
|
|
|
slush (OP)
Legendary
Offline
Activity: 1386
Merit: 1097
|
|
April 18, 2013, 03:07:23 PM |
|
I noticed that I am getting the message that some of my workers are using the deprecated "getwork" protocol but all my workers are using cgminer 2.11.3 (one down from the current 2.11.4). The message suggests that I use cgminer 2.8.x or higher. What else could be wrong?
It is possible that cgminer hanged with getwork when stratum was inaccessible. I don't know how cgminer handle this internally. However stratum.bitcoin.cz should work for now without issues.
|
|
|
|
slush (OP)
Legendary
Offline
Activity: 1386
Merit: 1097
|
|
April 18, 2013, 03:08:49 PM |
|
Since the DDoS I've had problems getting my 7970 ATI mining backup to it's ~614 MH/s I had using GUIMiner with cgminer.exe on stratum.
I don't think this issue is related to Stratum or DDoS itself. Stratum has been designed especially to provide local work generation, so even high latency and/or packetloss should not affect your mining speed (in contrast to Getwork). Do you use same flags and other cgminer settings?
|
|
|
|
joolzg
Member
Offline
Activity: 76
Merit: 10
|
|
April 18, 2013, 03:10:07 PM |
|
my miners are not connecting, with debug im seeing
couldnt rsolve api-stratum.bitcoin.cz
joolz
|
|
|
|
slush (OP)
Legendary
Offline
Activity: 1386
Merit: 1097
|
|
April 18, 2013, 03:12:31 PM |
|
couldnt rsolve api-stratum.bitcoin.cz
I realized this second ago. I already fixed DNS, it will propagate in 15 minutes... I'm sorry
|
|
|
|
joolzg
Member
Offline
Activity: 76
Merit: 10
|
|
April 18, 2013, 03:15:20 PM |
|
thanks
|
|
|
|
cosurgi
|
|
April 18, 2013, 03:17:39 PM |
|
AFAIK Amazon EC2 is practically unDDoSable, I recall a slashdot story from a year back, where some poor idiots tried to ddos amazon, and they failed. There was some fuss about it and their threats were "serious". But I don't remember the details. Perhaps you could google that.
|
|
|
|
slush (OP)
Legendary
Offline
Activity: 1386
Merit: 1097
|
|
April 18, 2013, 03:18:12 PM |
|
Btw I received many offers for DDoS mitigation services or proxying the traffic thru some well connected nodes. Please don't take it personally, but it is hard to trust unknown strangers from Internet in this area.
|
|
|
|
slush (OP)
Legendary
Offline
Activity: 1386
Merit: 1097
|
|
April 18, 2013, 03:21:14 PM |
|
cosurgi, thanks, I'll google it. Right now I'm designing "final" EC2 architecture so any experience in this area is valuable.
|
|
|
|
Synapse_On_Slush
Newbie
Offline
Activity: 22
Merit: 0
|
|
April 18, 2013, 03:28:14 PM |
|
Just checked things and using CGMiner through GUIMiner as I have done before is also now doing 575 MH/s so it not what I though (GUIMiner using some fancy options with CGMiner to make it run faster). In answer to you question, nope, I don't have any options or anything. The only tweaking I've done is OC'd my GPU with Catalyst. I'll just wait and see what happens, it might speed up again in a day or two or it might be something's changed with my desktop, I could try restarting it. Thanks for all your hard work btw Slush You're a legend
|
|
|
|
PuertoLibre
Legendary
Offline
Activity: 1890
Merit: 1003
|
|
April 18, 2013, 03:45:05 PM |
|
AFAIK Amazon EC2 is practically unDDoSable, I recall a slashdot story from a year back, where some poor idiots tried to ddos amazon, and they failed. There was some fuss about it and their threats were "serious". But I don't remember the details. Perhaps you could google that.
Has anyone ever tried making Amazon EC2 instances that mine BitCoin/LiteCoin/NameCoin (Via virtual GPU or raw CPU mining)? I am just curious if it has ever been attempted. If so, was it profitable? If it was profitable, does Slush accept that kind of setup at his pool?
|
|
|
|
|