Bitcoin Forum

I've had no response from it for a few days now....

Works fine for me.

+1

Have you tried things such as traceroute to attempt to isolate the issue?

Slush is very much up and working.

after adding a / to the url I run it now and get

[2011-02-25 19:47:17] 2 miner threads started, using SHA256 'c' algorithm.
[2011-02-25 19:47:31] JSON decode failed(1): '[' or '{' expected near '<'
[2011-02-25 19:47:31] json_rpc_call failed, retry after 30 seconds
[2011-02-25 19:48:02] JSON decode failed(1): '[' or '{' expected near '<'
[2011-02-25 19:48:02] json_rpc_call failed, retry after 30 seconds


any ideas?

Permissions problems, most likely.

I'm going to take a guess that your options parameter is not correct for the url it should be

--url http://mining.bitcoin.cz:8332

I'm getting the same message myself

[2011-02-26 20:54:18] HTTP request failed: couldn't connect to host
[2011-02-26 20:54:18] json_rpc_call failed, retry after 30 seconds
[2011-02-26 20:54:19] 1 miner threads started, using SHA256 'c' algorithm.
[2011-02-26 20:54:48] HTTP request failed: couldn't connect to host
[2011-02-26 20:54:48] json_rpc_call failed, retry after 30 seconds

i am using

minerd --url=http://mining.bitcoin.cz:8332 --user=xxxx.xxx --pass=xxxxxxxxxxx

maybe I am missing something as well. Any help would be greatly appreciated

Is there any possibility to log what the miner receives? Looks like server is sending non-json response, but I don't see why exactly. Log should help...

I think this is another issue than response parsing. The pool is up and running, so I don't see the reason why you cannot connect, except that there is something wrong on your side (firewall etc).

Please edit the forum topic, pool wasn't dead for longer than 10 minutes since it started on December 2010 ;).

It could be your DNS server.  ATT DNS is having issues at the moment.

From a command prompt try


and see if that resolves to an IP address.  If so, you may want to change your DNS until ATT gets theirs fixed.

";; connection timed out; no servers could be reached"

I am beginning to think I am having problems outside the miner or pool...

I should note that this is a new machine on a new network. Fresh install of ubuntu 10.10

Check for firewall or dns problems

it was firewall/iptables. FIXED. Thanks for the sounding board!!

As of now, it is responding as "502 Bad Gateway" (nginx/0.7.65).  The pool client is also showing bad response from server.  I suspect that I will not be able to get new work from them or at least won't be able to submit finished work.  So, maybe time to setup with an alternative pool as backup (I am tempted to put another (older) video card in my machine for the sole purpose of GPU computing (460GTX overclocked) and use it for another worker anyway, but I am not sure I have enough power and don't want fans screaming louder than they are.  Anyway, since this site is based on scoring which weights your uptime in a round for payout, this type of outage is rather unacceptable (site should be pooled with multiple IP addresses and redundant backbone connections (using a good ISP perhaps) and reliable software that doesn't all go down at once.  I like slush, but this is not good. ???

I'm having the same error (502 bad gateway), but I'm still able to mine and my shares are getting accepted.

I think that only the site is broken, the pool itself may be still working fine.

There is no need to create a new "alternative" pool because we already have enough different pools.
My pool has servers in different DCs with automatic failover for high availability :)

bitcoind freeze on webserver, it's restarted and running again. Mining was not affected.

Yes, it appears that the miner was working, but I had no way to tell if it was working on real data or just eating GPU cycles [especially stale data].  Looks like the slush contribution is very small according to bitcoin watch as well.  Slush screwed up it appears.  In any event, I am working another pool until I am convinced that Slush has things under control. 

BTW, the site was back up shortly after I wrote my original messages.  I noticed that 502 errors have been a common theme for that pool over the last couple of months, so I really wonder about the ability of the system :(  I hope it is just my apprehension and that it will turn out to be fine and stable with significantly better uptime in the future.

Actually not working now.

Current server load (60 sec average):   0 getwork/s
Connected workers:    2749
Total score of current round:   20818426.1946
Shares contributed in current round:   233693
Approx. cluster performance (30 min average):    12.060 Ghash/s

And that takes down the entire site?  Seems you should have a monitor for the daemon.  Further, the site should be more robust to at least say something like "we are having technical difficulties, but rest assured, the pool is still up and your miners are processing valid shares" or something like that.  Also, I would recommend updating your web server to the later version (I haven't looked at the change log for that web server, but I hope you have the security holes covered).  Apache does work well, and security analysis is done on it quite diligently.  Just my 0.02 BTC :)

Can't connect... "Problems communicating with bitcoin RPC"

It's really unfortunate---my other pool has had some of its own technical issues lately so I was coming back to this one, but I can't even connect. If I had already gone to bed by now like I should have, I probably would have disconnected and not found out for at least 6-8 hours. And my solo mining has been luckless since day 1 (as in, 0 blocks generated in a month), so even falling back on that seems like a bad idea. This kind of sucks.

I feel your pain. I am not mining at bitcoinpool until he resolves this issue. I asked in his official pool thread that I'm wondering if the pool is/was under attack.

Because data on site is hardly coupled with info from bitcoind - yes. But it took down _only_ the site, not mining system.


OK. What's with monitoring? When I found problem, I should stop my car, find nearest pigeon and send packets to repaird lockups?

I wonder the same thing and is why I pointed out his software version [as shown on the 502 error page].  nginx isn't exactly the world's most popular web server and I suspect running an old version without doing conscientious security audits leaves you wide open.  In this case, he claimed it was bitcoind [which I doubt based on the slush share of zero on Bitcoin Watch].  Anyway, I hope he addresses it soon and once and for all.  Being new to the forums here, I looked back a little and see that at least the 502 error is not a new problem.

I didn't catch relevance between nginx and current pool troubles discussed here. Am I missing something?


bitcoin watch is loading pool stats from site. When site is down, bitcoin watch displays 0% in the graph.

OK, good.  That makes me feel better.  Any idea why bitcoind keeps dying and why it takes your site down?  I didn't receive my payout last night as I should have.

Yes, payout problems are related, its calculated on the webserver. Everything will be fixed today.

bitcoind hangups are documented and last bitcoind version already fix that. Unfortunately I have many custom patches in bitcoind, so update is not so easy - I need to test it well.

I'm developing a bitcoin website too, and these bitcoind hangups happen with an annoying frequency. Do you know where can I get the latest bitcoind version?

https://github.com/bitcoin/bitcoin

But you must build it from sources.