 Show Posts
|
|
Pages: [1]
|
Hi.
i got this error when runing daemon: 21:50:04.421288 ERROR Exception: Failed to read from IInputStream
I've seen this pop up every now and then. When it happens one can no longer run the daemon. Removing the blockchain database/cache (the .sld folder) and restarting the the daemon (which resyncs the complete blockchain) is what fixes it for me every time. Seems to me the code is vulnerable to database corruption, and is unable to recover on its own. This is likely a problem with the original code, not just the Soldo code-base. |
|
|
|
Cryptonight resistance against ASIC was broken
proofs please. They are talking about it here: https://bitcointalk.org/index.php?topic=3103990.0. I don't see much skepticisim in the thread, and in fact it looks like several CryptoNite-based coins are planning hardforks to introduce stronger ASIC resistance. I in no way, shape, or form pretend to be knowledgeable on this topic beyond what is in the thread, but based on what I see there (and over on Reddit) I don't think dismissing this without any thought at all is not the best face to present... |
|
|
|
|
For those who have been asking 'how many coins per day?' type questions:
The answer is, of course, highly variable, but here is a datapoint from today.
In a 6 hour period, with a hashpower of roughly 4 KH/s (spread over 5 systems), I mined 17 blocks and got 18 coins.
Though I have no idea if that'll ever be worth anything, it feels like progress, so I'll be here until the Whales come with their GPU-Farms and ruin all the fun...
|
|
|
|
|
Good News(TM): I've updated my Docker container to build #41. YAY!!
Bad News(TM): A change I wasn't ready to publish yet snuck in. BOOO!
It is still usable, but you have to be sure to pass in the user if you are mounting volumes because the container no longer runs as root (YAY!!). I'll work on polishing this up and documenting it this weekend, but I really gotta get back to studying for some exams this week.
Side-note: Now is the time to offer any advice/questions/criticizims regarding the container as I'm hopeing to get a lot of work done on it this weekend, including some usability enhancements.
|
|
|
|
Ryzen 1700 3.6ghz (8 threads) - 2.2 kHs
Testing has shown that for the Ryzen 1700, 6 threads is the 'sweet spot'. I (and others) have seen the same, or even better, hash rates with 6 threads then 8. Even better if you can pin each tread to its own core. I get 2.5KH/s with this setup. However, if you are getting better rates with 8 threads than 6, I'd like to know the details (i.e., platform, commands run, etc...) |
|
|
|
|
FWIW, I have updated the Docker container awmyhr/soldo-miner to bulid 39, so if you are using that please re-pull...
|
|
|
|
Very rough benchmarks (for 2 threads on each system): CPU GHz KH/s Host OS
i7-5600U 2.6 0.48 Fedora 27
i7-5600U 2.6 0.48 Fedora Atomic Host
i5-3340M 2.7 0.33 Fedora 27
i5-2500S 2.7 0.64 macOS 10.11
A10-9700 2.4 0.21 Fedora 27 I run the daemon (sldd) in its own container on the AMD system, then the miner (sldm) in a container on each system pointed to the daemon host. Small update: Today i upgraded my AMD system from the A10 to a Ryzen 1700. Aside from letting the BIOS 'auto-overclock', I haven't done much tuning on it. With 2 threads, I'm getting about 0.82 KH/s. With 7 threads I'm getting about 2.30 KH/s! (I didn't run 8 threads cuz my boy is using one core for Minecraft, lol...) |
|
|
|
incredibly, I wonder who this lucky guy is?  Is there any actual lottery statistics?  Just got home and checked my miners, no Lambo for me... |
|
|
|
Hello all. I posted this on the official forums, but not sure how much overlap there is... I have updated my soldo-miner Docker container to the latest build 38d release: https://hub.docker.com/r/awmyhr/soldo-miner/Also, I believe I have nailed down the proper build flags so that this *should* run on 'mondern' (as GCC's 'generic' flag defines it) x86_64 hosts. I've tested on the 5 systems I have at my disposal, but if you're having issues please let me know... (For those interested in how the sausage is made: https://github.com/awmyhr/docker-soldo/tree/develop ) Very rough benchmarks (for 2 threads on each system): CPU GHz KH/s Host OS
i7-5600U 2.6 0.48 Fedora 27
i7-5600U 2.6 0.48 Fedora Atomic Host
i5-3340M 2.7 0.33 Fedora 27
i5-2500S 2.7 0.64 macOS 10.11
A10-9700 2.4 0.21 Fedora 27 I run the daemon (sldd) in its own container on the AMD system, then the miner (sldm) in a container on each system pointed to the daemon host. |
|
|
|
Guess it's pointless for me to mine with my CPU, with 5 threads im only getting about 0.11 kh/s, kinda impossible to find a block with that speed?
FWIW, I've been able to find blocks running at 0.2-0.3 kh/s overnight. |
|
|
|
Due to recent DDOS attack on many server pools, we decided to purchase anti-ddos protection in the datacenter. It is expensive, so we are forced to set a 1% commission for the portal.
We have no other choice. We want to support YENTEN and protect our customers. We do not want our customers to lose their profits. It is better to prevent a threat than to fight it.
This is fine and understandable, but you need to stop advertising '0% fee' when it is acutally a '1% fee'... |
|
|
|
|
Finally got some time to mess around with this coin again, but ran into the problem were I was completley unable to connect to the network. Is the whole thing down, or am I doing somethig wrong again? I saw a huge list of IPs eariler, and fed that to my daemon, but none came back as active...
|
|
|
|
Hello
I am using a Ryzen 1600, getting around 450 h/sec. Using 12 threads. The CPU temp stays at around 50 deg.
So is it ok ? or can I get more performance, I am using minerd miner.
You may not be getting the best possible performancse. I'm running the AMD A10 (4 cores) and getting around 400 h/sec, or about 100 h/s per core. Given that this is the super-cheap/low-end version of the Ryzen, I'd expect better over-all performance for the 1600. At this time I do not have temp readings. For what its worth I'm useing the JayDDee optimized cpuminer ( https://github.com/JayDDee/cpuminer-opt), which I have compiled myself and run in Docker containers https://hub.docker.com/r/awmyhr/cpuminer/ (NOTE: I have yet to document usage, still in very early stages of getting that all setup). |
|
|
|
That did it, thank you very much! I knew it was something simple... |
|
|
|
|
Hello, I've been trying to compile the unix headless binaries per the doc in an Ubuntu container (same base container I've used to build numerous other coin code) with all listed pre-reqs, but the make fails immdiately:
root@6e14f56a7f01:~/gtoken/src# make -f makefile.unix
g++ -c -O2 -pthread -Wall -Wextra -Wno-ignored-qualifiers -Wformat -Wformat-security -Wno-unused-parameter -g -DBOOST_SPIRIT_THREADSAFE -I/root/gtoken/src -I/root/gtoken/src/obj -DUSE_UPNP=1 -I/root/gtoken/src/leveldb/include -I/root/gtoken/src/leveldb/helpers -DHAVE_BUILD_INFO -fno-stack-protector -fstack-protector-all -Wstack-protector -D_FORTIFY_SOURCE=2 -MMD -MF obj/alert.d -o obj/alert.o alert.cpp
alert.cpp:273:1: fatal error: opening dependency file obj/alert.d: No such file or directory
}
^
compilation terminated.
makefile.unix:167: recipe for target 'obj/alert.o' failed
make: *** [obj/alert.o] Error 1
I didn't jump straight to opening an issue as perhaps I've missed something simple, anyone here have experience compiling this?
|
|
|
|
|
