 Show Posts
|
|
Pages: [1]
|
i understood that previously, and raised a question for dealing with that kind of security issues to get clean (useful) currency. if that kind of bug occurs -> wallets can be stolen, market crash, and etc, if you can transfer money from dirty wallet, a.k.a money laundering.. I suggest to insert some kind of system for upgrading wallet/client/pool to new version, to prevent direct transfer from bad/dirty client and/or wallet, what i did easily. Also determine in currency policy what/when how coin idea deal with those kind of things. It is ok now when testcoin, but for real coin there should be some strategic rules to maintain whole process in tact edit: what concerns me, that dirty client and/or wallet was accepted by network... If the pool has already paid the money, and after that they have left, then they also leaves from those to whom it paid.
that was not clear to me (probably the meaning was lost for me in translation at some point). In the post above it is valid (same) process in currency exchange sites as you described in program? Cause I bet that all participants have hope to push this coin in to the 'real' world too. So wondering if you have developing this project in that manner, which can be implemented within a time without refreshing whole mining progress... |
|
|
|
|
nice thing,
but if I missed please repeat how do you deal with actual cheaters in those two scenarios:
I. max pool connections can be filled up with 'empty' miners -> i'm not talking about botnet, just i.e. thousand (or more) miners -m 0 and and just wait for some coins xD,
II. or do it with purpose to fill up pool and eliminate possibility for other miners(people) connecting to pool -> getting sort of solo mining carrier with no or very little concurrents -> which effects many things.
|
|
|
|
|
hello again,
0. if you did not already -> change coin amount storageing type from float (single) to decimal, as should be used to store currency values) - or then it will take enormous amount of storage space in pool ?
1. please add -w option for initial run (alongside -m -t options, w for passWord xD) for those who are confident, that they have secure systems.
2. a. what is the mechanism of miner/pool versions to allow to be connected
b. and have a save wallet?
context: had updated from 'yesterday's' to 'today's' version -> "f*ed" up the wallet (lost address and all testcoins in it).
but that should not be, unless that was intentional with new release to wipe everything clean (should be mentioned in release's description maybe?.. or hint for some way to fix this may be in next sentence)
for the first run it requires to set password and generates host keys. but after update it just asked for password, so user may assume that wallet was saved, but after checking -> oh you have been robbed. (I think that requires some fixing too)
c. also there is no way to check wallet offline [and no last known date when that wallet was validated by network -> maybe that can be extracted from log data, but it is not convenient for the average user. so implementation would be nice].
d. that means validation only by network, but what if one pool has been disconnected (like ISP problem occurs for couple of hours, when can not connect outside your country) and some people still mining on it, how validation process is going for that ?
how it is decided which pools data is valid and which ones has not valid data?
e. I have checked this thing: changed with yesterdays from today version, and all older testcoins are here. shouldn't be the same wallet for any version of miner? this coin/program version validation process seems weird (or non existent)
P.S. sorry if I have messed up this part, cause I did testing tasks when wrote, so there will be redundant questions, but leave it anyway to give more (raw) info for dev...
3. there are another connection check delay issue (or not) -> connected to the testnet. set mining 0, or connected with default -m 0. disconnected LAN cable (wifi was off too. ie. discord gets its status changed within less than 5secs, so i was really offline).
if I give commands as 'state' 'account' 'balance' -> there was ok, said that for different time intervals check two times one was about 2m 55s, other 2m 35s -> margin of error is +- 2sec (set some delay in script how often write a command to check state). I did not test it more, because of that around 20 sec gap.
i suggest make that more fixed than this. and give some option like default time to check if connected and manual option for time to check connection status (min value maybe 30sec or more)
above my concerns about essential functionality. especially number 2. It smells like #2 for me. so I gave that number on purpose...
these are just bonuses, but still useful
4. a.>> some status automatisation: some hashrate or smth like that status in miner? (helps to check better mining performance within cpu that are like 2c/4t and 4c/4t)
b. >> maybe sdtout can be with some time stamp {in windows there none, idk about linux version}?
it may be useful in end user (which usually do not know about .log -> and to find something in it requires some patience too...) perspective when checking account or sending
c. >> in 'state'. please add mining threads count instead only "mining on" status. like "mining on "+N+" threads"... cause if there are 2c/4t cpu, can be pressed by accident 7 instead of 4, and that is not good for mining (already tried).
until the time where it will be some implementation in GUI with logged data...
please do not quote my large posts and save thread space. I insert simple numbering to maintain basic clarity for diff parts.
edit: please do not requote posts (and save forum thread space for more readability)
edit#2:
what the ?
cheatcoin> account
qft1tyDsTPOt9fFmukRjeChH7BcAq+dt 24448.194650061 key 0
cheatcoin> account
qft1tyDsTPOt9fFmukRjeChH7BcAq+dt 24448.194716423 key 0
haven't been mining at all just connected to the testnet
and I did not made any transfer
|
|
|
|
|
2. a. for this as I understand that pool does not require even a dedicated regular HDD because of low IOPS? I actually never tried too simulate that much of open files,but i will try a bit later to see how windows reacts XD, and join a next day for testnet (after when responsive deamon comes out).
2.b. what is main reason for maximum 4096 for pool: this for whole pool or just a single port? there is posibility to run multiple same pool ports? and how they react to each other? i can try to run, but i'm not sure how properly evaluate results (that is why i'm asking)
2.c. btw what is you policy on public/private knowledge of existing pools (maybe your one known and other pools like in dark net-need to import address manually before miner connect), cause i don't get it how connecting to the testnet/mainnet works when multi pool network starts.
or program is still under heavy development part for this thing?
3. maybe yes , maybe no, maybe idk, but the fact is that amd users are/will be basically screwed if in this currency avx512 gives it's full benefits, cause even new AM4 platform does have only avx.
avx2 is not very common after this year launch and large amd sales if recall correctly, avx512 even less at least in user level market, so it maybe won't be for now.
those 8-10x speed is between avx2 and 512. avx is even more slow. after my comments now we may have more specific audience for mining this currency xD
btw i know that few asic was litterally about 1000x faster, but some are in similar range in hash/w (gpu mining in that case is even better, because gpu has multipurpose)
well that is my very limited experience. and no need to repond for this part anymore. unless someone can provide actual data for hash difference (if any is for now in kinda early dev. stage)
|
|
|
|
|
dear, cheatoshin0 ,
I have several questions.
1. can you add to daemon some response to user? like status report what program is doing or done what it supposed to do or does nothing, because can not connect and etc.? maybe already added? if yes -> then thank you
2. please add some guideliness for the pool parameters (since you are running one) vs pool load. cause maybe there can be reasonable amount of small pools not running on server grade rack (or even so then just few spare nodes)?
i have static IP. but have no idea about suggested loads even on the test net...
3. about miner, algo and performance: since it is sha256 and should be ASIC resistant. Yes? how miner performs within processor which are limited to avx, avx2 and avx512?
because in some sha256 benchmarks avx512 capable processors runs like ASIC against avx or even avx2, but in other common task performance difference is within the reason.
If avx512 capable processors has that amount of mining performance gain, then mining is rigged from the begging...
|
|
|
|
maybe limit pool to have active connections to 16-20 another pool for reasonably small time period. after that period some part of active connection canceled and reconnect to other pools. some kind of slow/delayed syncro with all pools in reasonably big time period. but one pool does not have too many active connection if overall pool count is big ?
I think that's a good idea. Maybe we should once every few minutes to randomly tear one of the connections giving others the opportunity to connect. It is difficult to imagine how much we can have pools on restart. On the first launch there was only 4 open ip addresses. I recently revised the log file dnet.log. It turned out that my node received more than 1000 connections at the time the network was launched. Maybe it was an attack, but their addresses were different and it seems they using our protocol. Of course, it was impossible to work. I think if we had about 20 pools and about 50 miners in each, it would be the best option to start. if we have small connection limits. it should be able to detect cpu threads (it was already done) also with network adapters. so if somebody have high-eng rig or server it can be easier to config the load for them. also some options to LAN connection priorities against public network should be added (it has similarity to p2p networks from my point of view) |
|
|
|
Thanks dev, will we run into the same issue again? Will your nodes will be able to handle more than 20 connections each?
I think that now it is not good time to substantially rewrite the full node. It is necessary to make the functionality of the pool in it and write a miner that could be compiled and run on any hardware. In the pool, work with the miners will be arranged another way. There will be one thread that will consistently bypass the miners, their number may be more than 20. If any of the miners are forgotten, then this is not terrible for the network, just this miner will go to another pool. But if there are more pools than 20, this is another case. It should be noted that each pool has an open ip, so they do not have to connect directly to each other, but may be through intermediaries. Therefore, I think the number of connections to other pools can be limited to 20. maybe limit pool to have active connections to 16-20 another pool for reasonably small time period. after that period some part of active connection canceled and reconnect to other pools. some kind of slow/delayed syncro with all pools in reasonably big time period. but one pool does not have too many active connection if overall pool count is big ? |
|
|
|
|
OK. i'm a bit confused. because there couple of runs, with -d m 1; only m 1 , only -d what purpose of that?
i have my public IP, router, what ports need to be OPEN ? what procedure to run in sync? i was not being able to start with test net and Its looks confusing for first time to use. what and why we need to change in netdb.txt ?
how to determine if I'm in sync with society and how not? (because it looked different few times with those hashes was same and different but other stuff was weirdly inconsistent so..) my questions is messed up , I know that, cause I dont know what too ask even XD
|
|
|
|
I like your coin dev and I like your code, would be nice if you added more comments in your code though  but I understand it enough. I will run a full node when main net goes live. Though I agree with most of the other people on here. CHEAT should be renamed. It has a very negative connotation in English speaking countries. F - those english speakin.. they are minority against all xD Personally my opinion is that this is the best name for crypto coin. because in general all currency, and crypto currency as well or even more is sort of cheat mode in order to exchange some material goods and/or services (because by itself it has no real value at all). Why is this coin called CheatCoin? Care devs explain further why is this project created?
Please read our white paper http://cheatcoin.atwebpages.com/Cheatcoin.doc. The goal is to create a distributed payment system with a large capacity, which is based on the DAG instead of blockchain. The name is already recognized as bad. However it attracts attention well.) The project is very interesting and it looks like you (dev) is really serious about it. I suggest to make a rebranding before launch, cheating sounds really bad to me... I agree with you. Cheat might attract people's attention, but doesn't sound like a serious project. I suggest no rebranding or renaming. It is humans right for intellectual freedom to attach any kind of meaning for words. i.e. geek today it has more positive meaning than about hundred years ago.. there is no bad words, but now cheatcoin miner and wallet is little bit behind from starting. From comments above it can not be launched due flaws in security (maybe it was improved a bit) and sync algorithms. Good money trading system should trustworthy , but since is lacking in QA. it is a bit rush to start in less than 3 weeks of simple level testing. My personal opinion first 'real' coins probably can be asked to be eliminated before it evaluates to real cash-money market, or after that we can wait for huge hacking attack after this coin gets significant dollar value (everything has extreme security level until it represent no value). For darknet transactions - it is whatever if someone is buyin' and someone is sellin' then nobody cares about security nonsence until somebody gets spotted stealing coins in large amounts. I would prefer quality product. Now it looks fragile to be real one. It needs to gain few layers of armour. Sorry for criticism, but I don't have good solutions, cause this is not my area of expertise and I do not see exact weak spots (but there was some major and I believe there is few left). |
|
|
|
thanks a lot, it is now understood.
and yes, my testnet wallet.dat is exactly 96 bytes.
another question:
is it enough to backup wallet.dat + dnet_ket.dat in order to make a backup for my account ? or are there more files i need to backup ?
In case of data loss, it is enough to back up only the file wallet.dat. The file dnet_key.dat identifies your host on the network, but has nothing to do with coins. But since private keys generates from the random data saved in the dnet_key.dat file, someone, having captured the file dnet_key.dat, can restore your keys. Therefore, both files wallet.dat and dnet_key.dat should be kept secret. Probably, it makes sense also to save the file netdb.txt. It contains the ip addresses and ports of servers running on the network. Other files do not need to be backed up or in secret storage. 1. I have transferred 100000 coins to the address YjZxlonGl+tSjibh1GJr5//Cydlo9Uvn. But you probably can see them after synchronizing with the network. I think that is not secure. maybe add password or smth to encrypt personal wallet and/or dnet_key and can start only when manually password is entered with some option to be not secure and be more secure? like repeat password within some user specified time period after last time pass was entered - time frame should be defined by network sync (or smth like you probably know better). |
|
|
|
Posted by: itbrnnet Does anyone know if bitzeny/cpuminer is supported on Windows Server 2003 R2? I've recently tried to start it but got a message that normaliz.dll is not found. Tried to download it but there was no change in the miner's behavior.
Is there minimal requirements to the release of the OS?
P.s. I have a server with 4 x AMD Opteron 880 and win server 2003 r2 installed. Also it's possible to install Red Hat Enterprise 5. Go for the linux. always choose linux... unless realy need MS or Apple BS... you forgot to mention if 2k3 R2 (btw amazing os) is 64bit or 32bit (on this one you can not mine with precompiled minerd on older AMDs - cuz no sse4 on them...). I do not now if you can compile it from source on or for 32bit Win OS win no SSE4, but if you can than you could run it. Also I want know how minerd works on multi cpu systems with old OS (if manual CPU affinity assignment is required or not). -> just curiosity Posted by: codeany lapool is so sloooow and you are mining at 1% pool fee. try soup.misosi.ru, this pool does not apply fees and much more stable lapool was more stable, steady and accurate in terms of hash detection and given works. dashboard info too. in lapool I had 99% efficiency and 1% fee (in all 5 mined coins from beginning). in misosi I have now mined only ~2 coins, with 80% eff. and no fee. (~1 coin with 4threads, another one with 3 threads, cuz 4threads seems unstable) also mining itself on misosi:16002 looks weird. on 16001 does not even mine , cuz my cpus (all have 4 threads each) are not strong. mining on all 4 threads is unstable. on 3 threads is more stable and reports same hashrate like on 4threads in both misosi and lapool. which one is better (if both are runing xD)? so if I will notice any significant and worth to mention changes. You will be posted. |
|
|
|
|
hi,
anyone is mining on lapool.me ?
i was started to mine about two days ago. mined about 5 coins on lapool.me:3014 .
then server side had problems and it was mentioned on website after couple of hours, when i noticed that minerd was sending stratum_subscribe send failed.
so i did not pay much attention for that.
i'm wondering what did gone wrong with pool and/or minerd that when notification was removed I still getting same message.
thanks in advance
|
|
|
|
|
