| Membergroup | Average accuracy | | Staff | 88% |
Question: In what cases would staff report rather than self action? I always have a look at the modlog/deletion log whenever I want to see if it was dealt with or not. Unless it's simply a moved topic or any other action which isn't broadcast on the modlog. I was surprised to see myself with 99% accuracy yesterday on my 1142 reports. Then, I screw it up within a few hours and go back down to 98%.  40% report to post ratio, confirmed grumpy old man |
|
|
|
Migration testing ended up running an import script on the server ended up taking up way too much resources on the server. We are figuring out a better way to not starve the production box. This wasn't because of a DDOS. Hi,
was the forum under a ddos attack from someone ? I wasen't able to access the forum for about 30mn-1h . I was getting a Database error at the begginig saying "please try again. if you come back to this error screen, report the error to an administrator." for a while then that Gateway error which is known when ddos attack happens as far as I know
any Staff member can confirm that ? or members can confirm that It was really down and not just for me
Why are you running an import script on a production box? Kinda not profesional instead you should have had another server spinned up. Made that server a mysql read only slave and then run the import script off of that. I'm sure we could all pitch in for some new hardwa.... ohwait  Although /s because I'm not a donator. |
|
|
|
Last time there were avatars, someone hacked the forum by uploading javascript to the avatar. This happened:
Also, there was a meter saying that your bitcoins were going to be transferred to "CosbyCoins" (I wasn't here when the forum got hacked, but I have read about it.)
There should be avatar uploads in the new forum software, going to be released around February.
Video: https://www.youtube.com/watch?v=LKrOHAfMdxI&hd=1 |
|
|
|
|
I can't work out what the hell this thread is / what it has become. I can't see any connection between the discussion and mprep unless I'm missing something?
|
|
|
|
I've got a few S4 that recently started having every single share discarded. All of them started this behavior at the exact same time.
Temps are showing as 0, and all ASICs appear to be online. Any ideas? I tried reflashing the firmware, but that hasn't appeared to help.
Running through some steps to double check: 1) I assume you've done hard restarts 2) I assume your mains power is okay and unchanged. Green lights on the back of the PSUs? 3) I assume your pool is fine (what is it)? As its at least receiving shares to discard If multiple units all had exactly the same thing at once, it can only be the 2 things external to the S4s: Internet or power. |
|
|
|
I've got a few S4 that recently started having every single share discarded. All of them started this behavior at the exact same time.
Temps are showing as 0, and all ASICs appear to be online. Any ideas? I tried reflashing the firmware, but that hasn't appeared to help.
Replied to your other post, we will get confused if we discuss it in two places. |
|
|
|
Thank you for the great photos.
Keep up the good work.
Do you know if the power supply can be 9v dc like the S5?
It has full DC2DC so it wouldn't give the intended effect. |
|
|
|
Temperatures should not be let to rise above 70C Where did you get this information?? Is it from Avalon chip's datasheet? No, its just a sensible number. Chips are no where near any boundary under stock cooling as it works so well. and you will need a typical computer PSU with 4xPCI-E 6 pin power connector I'm wondering why does they put 8-pin connectors on miner's boards?.. Thanks for the nice guide and photos. I liked the '360 deg' view photo.
As Rabinovitch said, the power connectors clearly look like 8 pin and not 6 pin PCI-E.
Cheers
Correct, they are 8 pin however to keep from confusing people I refer to them as 6 pin as that's what they will have seen on every other miner and are familiar with. If they already know the difference between 6 and 8 pin, then they don't need the information in the first place. Its a bit of a catch 22. You could probably also get away with only 2 PCI-E 6 pin, and even on one cable strand if its 16AWG, but again if you're in a position to use a more extreme arrangement then you likely already know that. |
|
|
|
Did anyone noticed bitmain have increased shipping costs for S5 (was like 50 and now it is 80)
Hi, the shipping cost will be automatically calculated by the system when you place the order in our official website. I did not change delivery address. On the day of S5 release it was ~50USD and now it is 80USD for the same address. Shipping is sometimes subsidised by Bitmain, sometimes not. So shipping price may vary as that changes. |
|
|
|
Out of curiosity, what weighting did you use when you measured the SPL of the fans?
what is SPL? I am not familiar with this term. Sound pressure level. It's the dBs you got. Why was this question asked? [The thread never got continued]. Short or long weighting should come out near identical, one just takes longer to get to the reading. There is a lot more than just averaging going on between A weighting and C weighting. C is much flatter and extends significantly farther into the low part of the spectrum. Many people have noted the higher pitch of the S5's stock fan running full out. Nevermind, I thought you were referring to something else entirely. Hence my confusion. |
|
|
|
So is a corsair cx750m not beefy enough for these? I have two S5 running on one corsair cx750m each (which only have two PCIe cables each that split to provide two plugs each for a total of four). I did have one unit shut off randomly after a couple hours of mining yesterday but I didn't pay attention to whether the PSU had shut off or just the S5. The S5 was power off though, no lights, no fan, nothing. Turning the PSU off and back on brought it back up and it has been running since. The other unit has been running the whole time.
So as I posted above, one CM750M is absolutely fine for 1x S5. Its unlikely you could get away with it on just one cable strand, but two cable strands is more than enough [and I've been running for weeks]. |
|
|
|
Ohh great my S5 Shuts down after half an hour minig.
I'm 100% newbie and really dont know what should i check.It's beeping and have to turn off and on to start again.Overheat mightbe the problem but its in a garage which is cold.
I see there has been plenty of suggestions and nothing really helping. Can you post a screenshot of the status page while its beeping please? That will tell us if its pool, power or other. |
|
|
|
Just going back and confirming / correcting some assumed fixes. For the S5 what is the recommended worker difficulty. I have been using 2048. Is that high enough?
1024 should be good enough. 2048 is probably too high. The only difference you should see between 2048 and 1024 is more varied hashrate near the start before it averages out. Network load is negligible either way and both are still low enough that discards shouldn't be significant. EVGA 1300 for 2 S5 or EVGA 1000 for 1 S5 What's the recommended power supply for an S5?
EVGA PSUs are quite hard to come by in the EU, and when they are their prices are really out of wack. A Corsair CX750M would be recommended as an alternative for 1 unit, and a CM1000 potentially for 2. Someone will have to test it first (I only have 1x S5) but I believe it should be okay due to the crazy headroom naturally on a CM1000. Are you using four separate cables?
Separate?You mean each plug straight from the psu?No.I have 2 cables and with two heads each and one cable for one board. Should i buy 2more modular cable? Yes, you should have four cables. Pushing close to 300 watts in one cable is simply too much for most standard PCI-E cables. My guess is they are getting too hot and the PSU's thermal protection is shutting it off. If you feel the cables they are probably hot to the touch after the miner has been working a while. Each input to the S5 should be a separate cable. If the PSU has four separate PCI-E outputs you should be running a cable on each one. Nothing wrong with 300W on a single cable [READ PCI-E CABLE not PCI-E connector. Its assumed that there are 2 PCI-E connectors]. Even 18AWG PSUs will do 400W almost inherently. 500W things will start heating up and is very boarderline, 600W is no go on 18 AWG. Think of examples of this - ASICMiner Cube was about 330W on a single strand, Antminer S1s were ~300W on a single strand with COLD cables, ASICMiner Tubes were 440W on a single strand (with unusual PSU selection). PSUs with 16AWG can do significantly more. |
|
|
|
Out of curiosity, what weighting did you use when you measured the SPL of the fans?
what is SPL? I am not familiar with this term. Sound pressure level. It's the dBs you got. Why was this question asked? [The thread never got continued]. Short or long weighting should come out near identical, one just takes longer to get to the reading. |
|
|
|
I'm able to connect to the miner again but it's not hashing. Here are the screenshots:
What PSU are you using? Thank you very much for responding. An Aerocool Strike-X 1100W 90A at +12V; but the PSU cannot be causing any problems, it has to be a network thing because as I said the miner worked perfectly at my home (previous location) during more than 20h and everything was ok. Now I'm trying to make it work from an office at university, I do not need any login to access the internet, I just got a router and connected it to a RJ45 from the wall, internet works nicely, the miner does not. Any ideas? Is it maybe a ports thing? It could be internet related as part of the problem but it wouldn't usually display like that - you'd just get pools with no connection. It very much looks like a hardware issue with the main culprit being power [as you have no modules shown]. PSUs can die at the most random times as well. The last thing you can try is updating the C1's firmware to the latest version, but I don't think that's your problem, and you'll need a second PSU to test. Hello again, a couple of days and nothing seems to work. I took it back home (from university office) today and it is working great, no problem at all. As I told you the PSU is not the problem, it's less than 24hours old (the PSU) and same the miner. Curiosity, at uni it shows X.X.X.0 in hardware software version and at home 3.3.3.0 This shows that it is definitely a network's problem, but how could I fix it??! Thank you very much in advance for any possible solution. I wish Bitmain had a half decent customer support after shipping. It's an "only selling" company and it seems like I'd waste my time waiting for them to give me a solution so I'm not gonna even try. Unfortunately I'm not sure why it wouldn't work if you're using DHCP, sorry. |
|
|
|
i get sometimes 4 x's from my s2, it's bbb related, or something with the mainboard!
This is the C1 thread. Post a screenshot in the S2. |
|
|
|
Thanks your a life saver - I was just about to submit it as a TAR - Flash went fine Thx. the .gz file. Do not extract or it will not work. It wouldn't accept the file if you tried to give it the extracted part. |
|
|
|
I just cant seem to program a static IP when I try to log in to the new STATIC ip it says PAGE NOT FOUND like its not finding the web UI Also what is that little black button do by the SD card slot and can you put a SD card in it I don't know just got it that's why I am wondering  ?? Man the fan is real load to load its sounds like the miner is ready for takeoff like a boing 747 starting up!!!! Also cant put S3 fans on it either fan cable to short if I get an extender will it work they look way different ? You can get a quasi static IP by setting that device to always use the same IP via your router. Its like fixed DHCP. SD card slot on an S3? If you mean S5, I don't see what button you're referring to. The SD card slot is not used. |
|
|
|
I'm reading back and forth about using 2 PCI-E's for stock, but 4 would be better. Looking for some guidance on how to power two S3's. Please let me know which PSUs and how arranged.
All Corsair PSU's-
3x CX500M, 2x CX600, 3x CX750W, 1x HX750W, 1x HX850W
Thanks a million dogie (and anyone else!)
For 2 S3s, a single CX750M will do. |
|
|
|
Well unfortunately something went wrong with my test and both miners stopped mining. When I went to inspect the issue the PSU was off.. Nothing on the PSU was running. Not sure what happened, maybe the PSU is faulty. I will try another power supply if I can but I can confirm it was running at least 36 hours. @irobb how is your looking? Still going? Still running with no zombies. If I can find one cheaper than I got off Amazon I will get another since I'll be returning this unit this week. Why are you returning if you got it to work without going ZOMBIE? Because only 3 of the 4 asic's work in mine. If most of the units are going zombie because of thermal load is it possible yours isn't because it's only running 3/4? I would think yours would only draw 3/4 the power and generate 3/4 the heat. Its a myth / coincidence that zombie and thermal paste have a connection. |
|
|
|
|
Show Posts
