1921
|
Economy / Computer hardware / Re: [WTS] BFL 60 GH
|
on: November 16, 2013, 05:28:49 PM
|
If I do not hear back in a couple of hours... they are yours. 2 shipped BTC6
im in the UK is that ok we can sort out shipping costs
|
|
|
1931
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: November 07, 2013, 11:23:41 AM
|
one of my mini's has died!
i woke up to find it had 0gh/s on the pool so i logged in to try to reboot from the interface. rebooting didnt help so i just turned off the device, waited 10 mins and tried to turn on again.
when i turned on the fans went on for a bit as usual when you turn on, but then i smelled strong scent of something being burnt. i then switched off and waited about an hr to try again, this time the fan doesnt even turn and then i see some gray smoke coming out from the air exit area, i have shut it off and pulled the plug right now, anyone have this happen to them before? what are the possibilities?
do you guys think i can fix this and have it run again?
please help me troubleshoot!
i was running 0821 firmware at 400mhz.
I had the same thing happen with two of my batch 3 units. Unfortunately there is nothing you can do to fix this unless you can isolate specifically what part on the PCB died and replace it. This would require some skills that not everyone posses. If it's a blown capacitor it might be a lot easier to fix. I am hoping that the Avalon team would introduce Avalon modules with their new 55nm chips so I can just swap the old cards with new ones, but who knows if and when this will happen. so should we refrain from running on 400mhz then? my mini is only 2 weeks old, if even that. is 400mhz unsafe for mini to be running? I would not recommend running them over 375 i have 6 running at 375 and they all work fine no problems at all and I'm still getting 56 Ghz on each
|
|
|
1932
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 31, 2013, 03:46:13 PM
|
I would recommend you install CK's latest firmware, I think that's still 0821. I have seen a couple of people complain they have had nothing but problems with the latest Avalon firmware installed on their machines.
Flash the machine with the CK firmware, do not restore the settings. And add your mining servers back manually and confirm if it's working.
I am not sure if you have are doing this, but my Avalon's are set to failover and I have two pools added to them, I found when I only added one Pool the hashing sometimes would not start at all, it's only when I added the second pool (as a failover) I didn't have any issues at all and at restart the hash kicks on pretty quickly.
Keep in mind when you update the firmware (pushing it back to CK's 0821) the default settings will be applied to the Avalon, so you will need to connect to it on 192.168.0.100.
When you connect a computer do your network does your computer get an IP address? If so there is a DHCP server (application if you want to call it) running likely on your internet modem or router. Avalon's by default enable DHCP server so you need to make sure this is disabled as two DHCP servers on the same subnet (192.168.0.xxx) will fight with one another as you are technically only supposed to have ONE DHCP server in your network environment.
So disable DHCP on your Avalon, assign the Avalon a static IP within the range that your DHCP doesn't operate in; example again DHCP servers may say the first IP is 192.168.0.100 with a pool of 50 so .150 is the last IP it will hand out. You have IP's 192.168.0.1 to .99 and .151 to .254 which you can statically assign out to your equipment. If you assign a DHCP server address statically when a second device grabs that IP from the DHCP server which doesn't know what machines have that IP assigned statically you will have an IP conflict problem between the two devices.
The last thing I forgot to ask or mention is plug your computer into your network and confirm you can get out on the internet. Then go to command prompt (if you are running windows) and type "ipconfig /all" without the quote, then look for the network adapter that has 192.168.0.x for the IP, copy down the numbers, IP, Subnet, Gateway, DNS.
On your Avalon you it should have it's own IP, subnet, the gateway must match what your computer says, and DNS I set mine to 8.8.8.8 which is Google's openDNS server.
Re-flashed 0821 without keeping settings, connect wifi, dhcp (no hash), switch to manual IP, HASHING! Thanks all. Hi if you have the DHCP option ticked in Network settings-> DHCP and DNS then this will override you routers DHCP ip management.
|
|
|
1933
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 29, 2013, 08:35:01 PM
|
No such luck. But as if on cue, Avalon got back to me and gave me some things to try opening up the box. They want me to direct connect the FPGA hub to my laptop and see if I can get it to hash. They are thinking the router is flawed.
Hmmm not sure how a reset can do that Do you have wire shark? Or I use another program called Who's on my wifi It's very good at picking up computers
|
|
|
1934
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 29, 2013, 08:18:30 PM
|
I could use some help and am getting no response from Avalon support - I bought a number of miners in the auction but one has been problematic. Had a harder time getting it to communicate with wireless but did so eventually. Yellow light was flashing when hashing and speed was up down substantially. I wanted to go back and start from scratch with it. So under SYSTEM - BACKUP/FIRMWARE - RESET TO DEFAULTS I selected "Perform Reset" The screen showed the machine processing for quite a while so I let it go and left it for a while. Came back 15-20 minutes later and it was still processing. I clicked on one of the other tabs in the menu and it went blank like it was not connected anymore. So I shut it down and restarted it. Tried accessing it with 192.168.0.100 and its Wifi address. I get the black screen saying "LuCI - Lua Configuration Interface" and it tries to connect, but times out. I have let it sit on and off for quite a while over a few days. But still the same result. Unable to connect. Both the green light and the amber (yellow) light are on full time. No flashing. I tried using the tiny reset button by the ethernet connection. The blue light flashes rapidly as has been described, but still won't come up on my laptop. Any suggestions? Did it brick?
Hi it might of changed its IP address to 192.168.1.100 Try changeling your laptop to 192.168.1.2 and see if you can ping the other address
|
|
|
1935
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 29, 2013, 03:45:17 PM
|
Long time Avalon owner. Always had an issue on my Batch 2/3 4 module where if I oc'd too much I would get a decrease in GH eventually (sometimes 20-40%). Run those @ 335 (auto 325-335) 24/7 around 107GH and they maybe restart cgminer once or twice a week but never slow down. Really loved the 20130821 firmware. So I now have 7 minis running (stock 20130723), and let me tell you now 2-3 of them have so far had this issue where they slow down to under 30GH as low as 4GH or so? GAH! Changed the worst two to 20130821 and will see. However I am disappointed, it appears the --avalon-auto option doesn't work? (currently set 350 clock on page with these options: --quiet --avalon-auto --avalon-temp 50 --avalon-cutoff 70 --avalon-freq 350-375 ) < Those settings never show above 350 on the status page. Setting 375 manually works. put a sheduled task to restart cgminer, hash is same, hw error increases. i had same issue, after restarting cgminer every 4 hrs untouched for weeks, and constant max.. im having problems with my miner dropping low hash rate after hashing for a while, how do i set a scheduled task to restart cgminer? sorry. very new here! thanks for any help Scroll back a few pages. it's here! hi if you start reading from here https://bitcointalk.org/index.php?topic=140539.msg3419505#msg3419505you should find the info you need any problems let me know
|
|
|
1936
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 27, 2013, 10:46:41 PM
|
Is the fact that the task is invoked logged anywhere? I'm thinking it would be handy to know which task and how often its happening. Not necessary but would be interesting. I guess just watching how long its been running will tell me that its been run.
Thanks for the explanations. I'm going to add it to mine now too.
Whats the temperature where your miners are? I'm wondering if its caused by them getting to warm or ?
Hi ok you can see if it's been called in your system logs As for temps 34 46 50 Mine are in a small garage 3 side by side I also have a knc 275ghz miner 2 * bfl sc 60 Not sure on the temp in there as I have fans pulling in the cold air at the bottom of the doors and a big fan pushing the hot air out at the top
|
|
|
1937
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 27, 2013, 10:19:26 PM
|
pjcltd,
I woke up to mine in that non hashing mode again.
Hows the task working for you? What does the ...\.. mean?
Thanks.
OK what we need to add is this */5 * * * * cgminer-api devs | grep "MHS 5s] => ....\.." && /sbin/reboot ok the ....\.. bit means that we are looking for a hashing speed less than 10000 mhz and if it finds it it will reboot the miner and this will clear the cache issue. i have also added */5 * * * * cat /proc/loadavg | grep "2.\." && /sbin/reboot this will check the CPU load every 5 mins and if it finds that the cpu load is high then it will reboot the miner I have 4 miners with this setup and they have all been working all day (i'm in the UK) i would have normally had to reboot them twice today. I'm not saying that this has fixed the issue but its a dam good start at a workaround. Let me know how you get on with this. (don't forget this only adds 2 checks in your scheduled tasks and nothing else if you find this has not worked for you let me know. I'm trying to understand how ...\.. = 10000 mhz I tried a google search and nothing. What do you mean by this? (don't forget this only adds 2 checks in your scheduled tasks and nothing else Thanks, jddebug (linux newbie) [/quote] ok when you use 'grep' to find something in a file you can use wildcards i am using ....\.. to look for any number thats less than 10000 that has a decimal point i.e. 9959.01 = match (reboot the miner) 10110.11 = no match (do nothing) hope that makes a bit more sense 'What do you mean by this? (don't forget this only adds 2 checks in your scheduled tasks and nothing else' ok i am just telling everyone that we are adding 2 check's into the system. some times people will get you to add things to your miner that could cause you issues.
|
|
|
1938
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 27, 2013, 09:59:27 PM
|
pjcltd,
I woke up to mine in that non hashing mode again.
Hows the task working for you? What does the ...\.. mean?
Thanks. [/quote]
OK what we need to add is this */5 * * * * cgminer-api devs | grep "MHS 5s] => ....\.." && /sbin/reboot
ok the ....\.. bit means that we are looking for a hashing speed less than 10000 mhz and if it finds it it will reboot the miner and this will clear the cache issue.
i have also added
*/5 * * * * cat /proc/loadavg | grep "2.\." && /sbin/reboot this will check the CPU load every 5 mins and if it finds that the cpu load is high then it will reboot the miner
I have 4 miners with this setup and they have all been working all day (i'm in the UK) i would have normally had to reboot them twice today.
I'm not saying that this has fixed the issue but its a dam good start at a workaround.
Let me know how you get on with this. (don't forget this only adds 2 checks in your scheduled tasks and nothing else if you find this has not worked for you let me know.
|
|
|
1939
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 27, 2013, 08:09:28 PM
|
I'm sorry to post this a second time but what settings do I change to be able to remotely view the Avalons cgminer stats?
Edit: Never mind I see the answer several pages back.
Hi witch page is it on ?
|
|
|
1940
|
Bitcoin / Hardware / Re: Avalon ASIC users thread
|
on: October 27, 2013, 10:38:32 AM
|
Hi i have 4 Avalon Mini's and they are all having the same problem
The is this when i start up the unit
Load Average 0.12, 0.14, 0.14
but after the miner has been on for 5 - 6 hours the load average goes high to
Load Average 24.10, 23.4, 22.10
then the miner has gin rate will drop to 3.05 GH/s i then have to power off the miner witch is a pain if i am at work any advice ?
thanks in advance Paul
Paul, A number of mine do that from time to time too. I found it stopped when I put them in a cooler environment. A few degrees was all it took for all but one of mine. What I'm now dealing with is the mini will start doing duplicates instead of new work for some reason. My hashrate is the same at the pool but its sending thousands of duplicates. Hi thanks for the reply There is so much info on this thread i have just added this to one of my miners */5 * * * * cgminer-api devs | grep "MHS 5s] => 0.00" && /sbin/reboot in Scheduled Tasks so lets see if that helps also have upgraded the firmware to 20130923 ( yes i did do a backup first) What does that task do exactly? i found it here https://bitcointalk.org/index.php?topic=303376.msg3362789#msg3362789ok slight change to this if you set it to list */5 * * * * cgminer-api devs | grep "MHS 5s] => ....\.." && /sbin/reboot once the miner gets below 10000 Ghz the server will reboot i am testing this on 2 miners that have had this problem i will let you know how i get on.
|
|
|
|