Me and CfB already use Jira to track issues in our private repository on Bitbucket. It will be most convenient for me to have all bugs related to the nxt-client package there, where the code is, not to have to go to other websites. I enabled Jira for our public repository too: https://bitbucket.org/JeanLucPicard/nxt-public/issues and suggest that we start entering bugs and feature requests (such as API requests) there. I allowed anonymous issue submission for now, if spam becomes a problem I will change it to require Bitbucket login. Is it possible to add to the list of "Kind"? That way bugs for all the different NXT issues can be tracked in one system and you won't get bogged down by having to go through bugs in unrelated areas. Didn't see a way to modify the "Kind" menu. But I don't want any issues there unless they are related to the NRS Java server, and possibly the JavaScript client (critical bugs only). Bugs and feature requests for the other Nxt projects really don't belong there, I will not have time even to look at those. |
|
|
|
I know there are a lot of people following this thread who would like to contribute, but with it being so technical, not sure of how they can.
I have an idea that will let pretty much anybody contribute to NXT, but it will require someone to manage a bug tracking system. Not sure what the best bug tracking system is nowadays, but if anybody is familiar enough with it to set one up, that would be fantastic.
This way, people can report API bugs, usability bugs and everything in between.
Me and CfB already use Jira to track issues in our private repository on Bitbucket. It will be most convenient for me to have all bugs related to the nxt-client package there, where the code is, not to have to go to other websites. I enabled Jira for our public repository too: https://bitbucket.org/JeanLucPicard/nxt-public/issues and suggest that we start entering bugs and feature requests (such as API requests) there. I allowed anonymous issue submission for now, if spam becomes a problem I will change it to require Bitbucket login. |
|
|
|
SEO is irrelevant...
...because Google-killer will be based on Nxt.  Google will be assimilated. Resistance is futile. |
|
|
|
I agree with this and I believe it's an excellent idea and naming strategy for marketing and promotion purposes. The only thing I would change is the spelling and use the conventional correct spelling so it's easy for new users to find, spell, and search online such as:
Nextcoin
Nexttalk
Nexttrade
and wrap it all under one umbrella such as "Next Economy"?
I'm not sure about the "Next Economy" but I'm gonna definitely start referencing your terms in my upcoming Nextcoin videos unless I see a better naming convention.
Tai Zen
We have already discussed that several times. It is Nxt, not Next, and there is no coin in it. SEO is irrelevant. |
|
|
|
Isildur23 got 5k bounty from me for gathering info about current projects that need support. This is what we already have. Please correct, expand it. I'm going to post polished version in new topic To clarify this, I am working on the core of the Nxt server as a lead developer. There is no such thing as Nxt wallet. Currently the server package also includes the javascript client, but when we get a native or alternative client this part will be removed and I will continue to work on the Java based sever core only. I am not asking for bounties, this is why I don't want to put my Nxt account number there. If somebody really wants to send me a personal thank you for my work, there is a bitcoin address in my profile. But don't include me when figuring out how to distribute bounty funds. |
|
|
|
The SHA1 in the beginning of that file is because of the GPG signature it is signed with, nothing to do with the other sha256 value. You can verify the signature of that file using "gpg --verify nxt-client-0.5.5.zip.sha256.txt.asc", and check sha256 of the zip file by running "sha256sum -c nxt-client-0.5.5.zip.sha256.txt.asc", with the zip file in the same directory (ignore errors about invalid lines, look for the "OK" message). |
|
|
|
I think this is a great idea and I will support including it in the official client. |
|
|
|
Just wanted to say that 0.5.3 is running incredibly well for me. Is anyone still having problems with stuck blocks?
mine keeps crashing or something cuz Firefox keeps asking me if I want to stop or continue the script... dont know whats the cause... Yes that is at startup had the same problem with chrome, this needs to be fixed. At startup with blocks.nxt and transactions.nxt missing or up to date? If you are downloading the blockchain from scratch, or a significant number of missing blocks, it is to be expected that the browser UI cannot handle that many blocks. This will not be fixed as there is nobody working on the browser UI. You need to wait until the blockchain is up to date, and then open the browser. Or restart the browser only when it hangs, not the java process too. |
|
|
|
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Stable release 0.5.5: http://download.nxtcrypto.org/nxt-client-0.5.5.zipsha256: aa20fb32fe7e6cda7debee59928d5aea8ff7c581c58a56ff68e8252eaf9368c7 Change log: Added getGuaranteedBalance API: http://localhost:7874/nxt?requestType=getGuaranteedBalance&account=account&numberOfConfirmations=numberOfConfirmationsreturns the balance of an account only considering transactions with confimration count >= numberOfConfirmations. The API requests getAccountBlockIds and getAccountTransactionIds now return sorted results. Added getAccountPublicKey and getGuaranteedBalance to admin.html page. More performance optimizations. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIcBAEBAgAGBQJS0YsaAAoJEFOhyXc7+e2AtwoP/349GjR+VuIXo39X4l2N+Ute 1TNEW11wzwnwoFeaejNu3/HKTTooS37G6I0zxPFaFooCSoJZ7HAtqG9qYhF1byqF BYJCmM1tL65hD5Q40Jbijx+x7RBlqyHxSr07Uq/eWn2ILYkWNvHui3v3VE+GVBTm 0DBaacGAODh0SwXm6oZY1bToFMp9+kRCszvYeUJCvNXZRAj5gecgeVlc8ituFQRK azLFYsVj7hqRSkjwl/rRuuZGu02hJBrfNqunYBtCe/EjdFkwFdpuZpnpSeEXoxJH TT7OzIx1+0M1SjH98Hcbrx+4DbEtfsRedF861AChSG4Jb4/O1n0iQ6c0rsj/dZYE FO+Jcgcr0Rn76FKp8afukr1N+XjdigkJp/22LkWWk1q4/ByEaCv7FuKqCacSQ6N3 +H60NQ1fuOlkwcTeQtN+YSJyAVMLNeoIXRusOPPCeDHhK6RFadaWk9PoUgxUJtOj TRzsHQ0lZV6jiiUExogQGjsgVi2p0KcUQgaYlYpzLHzS2JH9IYWpON03RpJN2rlQ GNS87EYLfFWqWIQqXazS8tSufxBIF9OaTsqnpedXoxvevbLr5KyuT2VlmF/OAm4a z0/vZfDvUnHj3sX0lKxDlrvpSaQWV0CZeuQGMY76vVzeidhU3MAKgkCp+tAenZzH 2x3Wh5NbuNPKg6plCCiK =U0LQ -----END PGP SIGNATURE----- |
|
|
|
Hey guys,
I'm currently generating me a 777****777 address. Is my calculation correct that it takes about 900,000 generated addresses to obtain one in my wanted format with 6 wanted digits characters (the 1st digit can't be a zero).
The script is running on my PC and on my university cluster which isn't much faster. It's unfortunately running with the NXT API and in python. So it's not the fastest. I would like to make a native C version.
For this I would have to work me into the algorithms. Is this something the community wants?
You can start from my vanity generator: https://bitbucket.org/JeanLucPicard/vanity/src and add a Filter class that matches addresses in the format you want, should be just a few lines of code. Slower than C, but it can go through 8000 accounts per second on my laptop, and I just generated two accounts that start with 777777 in 400 seconds. |
|
|
|
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Experimental version 0.5.4e is available at: http://download.nxtcrypto.org/nxt-client-0.5.4e.zipsha256: 56b84fbcaecedb276ca2ca478c2af0fe3177da184c53863a5fb37f2807d4e451 Change log: Added Arbitrary Messages code, to be enabled at block 40000. New API requests added: getAliasId http://localhost:7874/nxt?requestType=getAliasId&alias=aliasreturns the id of the alias given alias name getAccountBlockIds http://localhost:/7874/nxt?requestType=getAccountBlockIds&account=account×tamp=timestampreturns the blockIds generated by the account after timestamp getState will now also return cumulativeDifficulty and totalEffectiveBalance. This allows the chance of an account to forge a block to be calculated by the user as myEffectiveBalance/totalEffectiveBalance. Significant memory consumption and performance optimizations to reduce the load on the GC and the total memory requirements. It should be possible to run with -Xmx512M if needed, more is better but should not be necessary. I hope Raspberry users will be happy with this release. This is experimental because of the multiple memory related optimizations, but should perform better than 0.5.3. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIcBAEBAgAGBQJSz8FUAAoJEFOhyXc7+e2AON8QAKXOHRKzk+XixZ2wg2AHVAJ1 QTYjccUl2w7wOIM7zYp4bdRSgoI4mi/bZT52I/Ktc9AD9rD46LGoaya00wqTGFD3 Ra/pKSGRXODWUNUpBRDQoc3urLiyNkjs8s9DpNrNZrnlGohvvde/k2h3W3Y0m2lz S68jUwHCGCMLz6H5jS7cqefSzkhE85S3MmbZlmfo+ov7SNEAn9bcCJbbGcWCbPXN 0S6dDYUMS9ERPq47iED90d0dtEs6MnfhMmx7pQUI9G2YwKRsRHnTwg5tQttA3Lh7 H7pYl4kPOPBrRZyxhNdKeHEOknoEZmDM0FiQIOc3idC1Vjh03NeQpNj8rCcAWvYo wPldZahvXtJr7iOwlSylkjeeT4rPf0oKC5NvuIKB8aT0x+38smNKicDN1bZS5HnO LyY/6yIMSSRGm9zwGFJoF50tfSVdwyj07bJJL0hzMEQrc65rWRs/739yn7n028PV 455glB1w93fH/i98YweCaxWuD1wI+zy18xb2xxESFF/WXarBTeTyxUTQyeRVnqQ7 02qiOjedcySlv+D44x0/GAiTeHvOft6vsTWfRH8IzCh2u/6FbNH33qLl1DyFVhSp RKpeyvAnQIUmGRu7gvT5UZY6j23jNPfgSfGF1B1jd/+OBMbdiKoZDh8UTmGwl5HD +Hbn8JOHqhcNcQc0PlHw =yFER -----END PGP SIGNATURE----- |
|
|
|
I suggest @info.nxtcrypto link @Luc's BTT post for each client update, so we can do a fast simple comparison with @Luc's post and confirm the sha256sum. If hacker replaced the download file and also replace sha256sum at info.nxtcrypto, it's not so easy to find it, but I think hack those 2 and Luc's account at the same time is more difficult.
See those *.asc files in the http://download.nxtcrypto.org/ directory? Those are GPG signatures of the corresponding zip files. If you download both nxt-client-0.5.3.zip and nxt-client-0.5.3.zip.asc in the same directory, you can run "gpg --verify nxt-client-0.5.3.zip.asc" to verify my signature of the zip package. This gives you one independent way of checking, and a hacker cannot provide a signature for a modified zip package without somehow stealing my private GPG key. The nxt-client-0.5.3.zip.sha256.txt.asc is again a GPG signed file containing the sha256 sum. You can run "gpg --verify nxt-client-0.5.3.zip.sha256.txt.asc" to verify its content, then run "sha256sum -c nxt-client-0.5.3.zip.sha256.txt.asc" which will say "nxt-client-0.5.3.zip: OK" if the sha256 sum matches (ignore the warning about the extra lines, those are the gpg signature). Finally, the value of the NRSversion alias on the blockchian contains the sha256 sum of the last stable release. That gives you quite a few independent ways of verifying the package. Add to this getting your public key based on "GPG key fingerprint" and it will be nice guide. And yes, we definitely need nxt-client-latest to automate all this steps $ gpg --recv-keys FF2A19FA $ gpg --fingerprint FF2A19FA Then compare with fingerprint in my profile info. |
|
|
|
I suggest @info.nxtcrypto link @Luc's BTT post for each client update, so we can do a fast simple comparison with @Luc's post and confirm the sha256sum. If hacker replaced the download file and also replace sha256sum at info.nxtcrypto, it's not so easy to find it, but I think hack those 2 and Luc's account at the same time is more difficult.
See those *.asc files in the http://download.nxtcrypto.org/ directory? Those are GPG signatures of the corresponding zip files. If you download both nxt-client-0.5.3.zip and nxt-client-0.5.3.zip.asc in the same directory, you can run "gpg --verify nxt-client-0.5.3.zip.asc" to verify my signature of the zip package. This gives you one independent way of checking, and a hacker cannot provide a signature for a modified zip package without somehow stealing my private GPG key. The nxt-client-0.5.3.zip.sha256.txt.asc is again a GPG signed file containing the sha256 sum. You can run "gpg --verify nxt-client-0.5.3.zip.sha256.txt.asc" to verify its content, then run "sha256sum -c nxt-client-0.5.3.zip.sha256.txt.asc" which will say "nxt-client-0.5.3.zip: OK" if the sha256 sum matches (ignore the warning about the extra lines, those are the gpg signature). Finally, the value of the NRSversion alias on the blockchian contains the sha256 sum of the last stable release. That gives you quite a few independent ways of verifying the package. |
|
|
|
Can CfB/JLP comment on VPS requirements for the new 0.5.3 code? If a hallmarked VPS with say, 30M on the account was running, how much mem would it need? (not my hallmark, BTW, Im hosting someones hallmark for them)
NCC-1701-D is running on 8 GB VPS, with -Xmx6144M, and java is currently using 3065M. On another server I have java is currently using 2124M. I am going to run some memory profiling again soon to see what is taking how much space and if there is any room for improvement before switching to storing blocks and transactions in a database, which would be a significant change. |
|
|
|
Now you confused me. Auto-rescan is supposed to happen or not? If yes, and you won't add GUI auto-recovery, it would be useful
to have at least a pop-up saying "Please reload this page to update transaction status" or something similar.
The re-scan happens as a recovery step when your client is intentionally being fed with wrong blocks and transactions by a bad peer. Then the blockchain is reloaded from the .nxt.bak files, the unconfirmed transactions discarded, and the peer blacklisted. But it is too complicated to also force the GUI to recover. Usually it is pretty obvious that something is wrong as the Recent blocks count becomes negative, there are lots of orphaned blocks and as you noticed thousands of bogus unconfirmed transactions. If a reload of the browser page fixes it, I think this is good enough for now. |
|
|
|
What happened in the old version when the account number was pasted twice, so it read 1295619013897570058912956190138975700589 ?
In the old version 1295619013897570058912956190138975700589 would be interpreted as 6667516829191161453. Now it gives an error. |
|
|
|
What does it mean [2014-01-08 19:28:58.571] ...Done
on log?  That it is done rescanning the blockchain after an error. |
|
|
|
Released 0.5.3:
http://download.nxtcrypto.org/nxt-client-0.5.3.zip
sha256: 23fc36fba166e00299003407169a26515e6d67c8094b5a06f9c795cc62ca83a7
Change log:
Fixed blockchain rescanning. Clear unconfirmed transactions on rescan. Not fixed. NXT just did auto-rescan and afterwards 17,653 unconfirmed transactions are still on the list. Machine runs Win7 64-bit. Did you reload the page in the browser? I have no plans to fix the GUI to recover by itself on this error. |
|
|
|
We don't integrate some sort of CRC because of saving a tiny bit of memory in the core code and because sort-of-CRC could be done in a/the client, right?
No, because in the core and when sending across the network the transactions are already signed with the sender public key. Any corruption in the transaction content would automatically invalidate the signature. |
|
|
|
Just wondering if you guys are using unit tests now?
I don't use them, looks like a waste of time to me. I've seen a lot of projects that had a lot of specifications, testing and other voodoo magic things and still failed. It is too early to add unit tests now. I am not a big fan of them, but they do have a value. But there is so much refactoring and re-writing needed now, that adding unit tests to it will only make it more difficult, more stuff to refactor and re-write. Unit tests are not that useful in algorithm design and in cryptography. You cannot prove the validity of an algorithm using unit test, you can only prove that the test cases that you could think of worked. Just like you cannot prove a mathematical theorem by inspection. I can write an encryption algorithm based on the assumption that all odd numbers are prime. If I test it for values of 1, 3, 5, 7, 11, 13, it will pass the unit tests with flying colors. Would you trust it though? |
|
|
|
|
Show Posts
