Did you even read the OP ? I guess not..

---

He wants to create one address (implies 1 private key) for each user, as mentioned in his OP.
Not a separate wallet for each user.

---

Why does the user need the key to decrypt the private key if he doesn't have access to the private key anyway ?
Also, you most likely don't want each user to be able to have access to the private keys. This creates room for exploitation.

The funds of the user are managed in a database and the coins should get consolidated anyway (when the fees are low).

---

^This.

No system is completely secure.
You need to hire experts which can be made liable in case of an obvious data breach. Not some random online guy.

Vulnerabilities will exist, it is important to have a plan to reduce possible damage and to make sure that the total amount of damage is limited.
For example, it has to be completely bulletproof that an attacker can't add withdrawal requests AND top up the hot wallet from your cold wallet. This would be a disaster.

---

That's currently not the topic.
Such a module won't help you if an attacker can manipulate the database which handles the withdrawals.

It is necessary to have a good concept, then focus on the hardware you are using.

I fixed that for you, LetsEncrypt offers free TLS certificates.


Scammer don't even need to pay for it. HTTPS only ensures that the traffic between the client and the server is encrypted.
This does not imply that you are communicating with the server you want to communicate with.

Unfortunately a lot of people think that the green lock besides the address bar means it is the legit site they wanted to visit 

Not necessarily.
Blocks contain ~2500 transactions currently.

So if it is less than ~2500 / X, you could theoretically assume it will get included into the next block.

But note that this is also just an estimation. Your transaction is not guaranteed to be confirmed within the next block if the priority shows < 2500 / X.
That is not even guaranteed if it shows 1 / X (which would mean your TX has the highest priority of all).

Same for me.

It takes very long to load (not 60 seconds, but about 30).

Additionally the watchlist exactly shows what i can see under the 'Show new replies to your posts'-tab, not what actually is in my watchlist.
When i click on 'Edit watchlist', i see my (real) watchlist entries (which differs heavily from the 'Show new replies to your posts').



Edit:
It is fixed for me now. Everything back to normal (loads fast + correct watchlist).
I am still curious what exactly happened.

The address given is the address the nano s has the private key of.

So it can always be used. The '1st verify step' is just to ensure that the address showing on your computer really is the one generated on the nano s (to be safe against a compromised computer).
The '2nd verify' (after copying it etc.) is for the user to make sure that his computer is not compromised with a clipping board malware or something similar (which changes any BTC address copied to an address from the attacker).


You can use all addresses given by the nano s without verifying them.
This is just done to protect against some possible attack vectors (e.g. compromised computer / human error).

Don't store them on your online server.

If you can allow to delay payouts / withdrawals by a few hours:
- Create a 2-server-setup with one being your webserver and one being 'offline'.
- Make use of 2 wallets (hot- and cold wallet)
- Every X hours batch the transactions, verify them, do sanity checks, and create a payment request to your 'offline server'.
- Your 'offline-server' should then handle all payments (make sure to secure it with a firewall and only allow specific incoming requests from your webserver + outgoing requests to broadcast the transactions) from the hot wallet.
- Top up the hot wallet with the cold wallet funds when necessary (manually) or withdraw from the hot wallet to the cold wallet (automatically).


If you want 'instant' withdrawals:
- Make use of 2 wallets (hot- and cold wallet)
- Only keep a limited amount on your hot wallet (on the web server). It has to be an amount which you can afford to lose in case of an attack.
- Top it up with your cold wallet funds when necessary.


But generally:
If you are going to build a trading platform yourself (without developer who build the whole software with security in mind (from the very beginning)), you are guaranteed to have vulnerabilities.
And these vulnerabilities will be exploited if it is worth it for the attacker.

If you really want to build a professional platform and need to handle funds of users, hire some competent security-minded developer.
It is not going to be cheap, but it will save you a lot of money in the long run (given that you really want to build a professional platform).

Generally, these tips are good, but there are still some things which need some improvement / are not completely correct.




You don't need a complex password, if it is long enough.




This is true, but only for windows and android / iOS.
I wouldn't call it crucial for MacOS / Linux.




HTTPS alone doesn't secure you too much.
You need to make sure that you are on the correct site (and theoretically that the certificate has been signed by the correct CA if the website owner doesn't use certificate pinning (to be on the very safe side)).

A small typo can lead you to a website which looks like the original one and is using HTTPS. But i'd call that the fault of the user.

But yes.. HTTPS over HTTP. Always.




To be more precise:
The server 'sends the cookie' to the client (Set-Cookie header in the HTTP response).
And the cookie doesn't contain any information about what you have done. It is simply an identifier (which can be used to track you across several site (by the site admins) / replace login information).

The only way a hacker can read the data you are receiving / sending to / from your browser is when you either don't use HTTPS or your browser / computer is compromised.
In the second case, deleting cookies / history doesn't help you. It only helps you if your computer gets compromised afterwards. In this case the attacker won't be able to reconstruct what you did in the past, but will still be able to read your future traffic.




Computers of which you don't control the hardware are always risky.
But using an open Wifi is absolutely fine if you encrypt and route your whole traffic via a trustworthy server (e.g. a small VPN server at your home).

If HCP's advises did not work, make sure that your software and firmware is up-to-date:

• Nano S Firmware: 1.5.5
• Electrum: 3.3.4
• Check whether your windows installation is up-to-date

You should make sure that everything is updated - even if that's not the root of the problem you are facing.

How did you create your wallet in the first place ?

Normally, if you create a new wallet file, your wallet software will generate a long random number (seed) and encode it into 12 / 24 words (mnemonic code; easier for you to write down).
Then, based on this seed the private keys are generated. The private keys are needed to spend BTCs.
Those private keys are used to derive public keys, which are then used to generate addresses.

If you have the seed, you can create the private keys -> public keys -> addresses.
But if you only have an address / public key, you won't be able to get its private key (which is necessary to spend the funds associated with this address).


So, did you create the wallet yourself ? Were you given a mnemonic code to backup ?
Or did someone else 'send you the wallet' (Note that the only proper way of transacting bitcoins is to receive them to an address you own (with its private key / seed).


If you were just given the wallet file by a stranger, you unfortunately have been scammed.

You can do CPFP by spending the freshly generated (unconfirmed) outputs in a new transaction with a fee which is high enough to pay for both transactions.

Choose the unconfirmed 'coins' as input and set a relatively high fee to perform a CPFP.


Or - if the transaction is marked as 'replacable' - right click on the transaction and choose 'increase the fee'.
But this requires you to have more funds 'in this wallet' (which i guess is not the case here).

You just need to 'reverify' the address if you copied it, to make sure you are not infected by any clipping board malware.

If you KNOW the address is correct and it is yours, you don't need to verify anything.

You receive bitcoins regardless of whether you click 'verify' or not on your nano s.

Setting 'block confirmation within X blocks' is just an estimation based on the current mempool / fees.

If there are going to be more transactions the next hours, it means that it will take longer until it gets confirmed.

It seems you are the sender, so in this case you have 2 options to 'speed your TX up':

• CPFP: If you didn't spend the whole inputs, and still had change coming back to you, you can spend this change with a very high fee (enough to pay for both of your transactions)
• RBF: If you have the RBF-flag set (default in electrum, i think; btw, which wallet are you using?), you can 'replace the fee' of the transaction and increase it.

If it is not urgent, you could simply wait until it confirms.
Based on your statement you have chosen a fee below 10 sat/B, right ? If you take a look at the amount of transactions waiting to get confirmed in the mempool, it still could take some time.

You are wrong.

Just because YOU can't manage a windows system / network properly, it doesn't mean that it is not possible.
While i agree that it takes way more effort to create a somewhat secure system running windows, it is - by far - not impossible.

We would have all of your governments and any other critical infrastructure compromised already.. The majority is using windows only..

You should keep away from the thought that not a single person can do things which YOU can't do..

Yes, i were talking about out-of-the-box in one sentence:


All i have said was that any system can get compromised and that no system is secure directly after installing.
It is the configuration which makes it more or less secure. The possibilities and simplicity of unix-based OS's are the reason linux can be made more secure in an easier way.
This has nothing to do with linux = secure; windows = horrible. It all depends on the configuration and the management.. always.




Just because you didn't see X, it means X doesn't exist ? What kind of an argument is that ?

Also.. why are you hating so much ? There is not a single reason to be aggressive at all..


Edit:
Just checked your post history and it seems you have a bad day today..
What about we stop the discussion here now and talk some other day about this topic (given that you want to properly discuss this topic) ?

Just tested it with Debian stable (testing has newer versions), and it seems it really only has python 3.5.3 in its repositories while electrum needs python 3.6.

However, it can be installed this way (python 3.6.3; just tested on debian stable):

You will have python3 (3.5.3) and python3.6 installed.

Then you'll need to install pyqt5 via:

Afterwards electrum can be run using:

inside of the electrum directory.


Of course this also works with newer versions of python.

And dynamite will destroy both.

With proper network- / privileges-management and some common sense a windows network can be as secure as a linux network.

It all depends on the security mechanism / -management.
A miserably managed linux network is way more prone to being compromised than a moderately good managed windows network.




So.. you mean there never were linux kernel exploits, privilege escalations or any other exploits which only affected linux and were severe (*cough*  shellshock  *cough*) ?


Just because the majority of malware doesn't work on linux, it doesn't mean that linux is more secure.

If you consider a non-techy guy who barely can open his browser and type into google.
Without any security practices, it is not harder to compromise his computer running linux than if he would use windows. Same applies to a MAC, iOS, android etc..

Most people using linux do have more clue regarding IT / security / etc.. And that's the reason why it is 'easier' to compromise a windows system. Most windows user just don't know what they are doing at all..

Which compatibility issues ?
Python3 is available on every distro. The only dependency it has is python3-pyqt5, which is also available on all distros.

Or are you talking about electrum from PIP ? In this case, yes. It is outdated.

But you don't have to install electrum from source then.. so i guess you are talking about outdated python? 
Which distro does not have python3 in their repository ?

 

The 'electrum support' for linux is the same as for windows.

What do you exactly mean?

Electrum runs even better on linux than on windows, or are you talking about the 'support from electrum'. In this case.. there isn't a support at all, because it is not a company.




Yes, and it is over.
All the attacker could do is to show a popup with a custom message.




IMO (if you don't want to use a hardware wallet), electrum.

You don't know how much inputs you are consuming / outputs you are creating  

Look at the preview tab (before sending the transaction).

We need to know how large it is (either in Byte or in amount of inputs/outputs).

While i agree that it is safer to use Linux, i would not agree that linux is a more secure OS per se.

It is the configuration of your computer, software, network which makes your system secure or not secure.

More than 90% of the malware is written for windows.. but this still doesn't mean that you are more secure (especially not in a targeted attack).
In a targeted attack it doesn't matter at all which OS you are using. Bugs and exploits exist (and can be found) for every OS / system setup.


Generally, yes. Linux is safer (out-of-the-box). And you are more secured against the day-to-day threats, yes.
But saying linux is more secure per se, is kind of wrong.




Password managers are a good idea, yes.
But in some cases you need to memorize your password (e.g. for an account you need to log into from different devices from different networks).
In this case you need some password you can memorize.

Then, you can easily go without special chars by increasing the length.

An explanation on length beats complexity regarding password security: https://bitcointalk.org/index.php?topic=5132378.msg50625648#msg50625648




Definitely, but maybe not manjaro.

For a linux newbie, some easier-to-learn distro might be more helpful (e.g. ubuntu / mint).