Nofish
nofish@zeroid.bit

ZeroNet dev.

Help distribute this user's images
Mute nofish@zeroid.bit

Activity feed

Nofish commented on Styromaniac's post: @Styromaniac: That's weird, because it's a pure-html page without iframe/wrapper/js The supported network should be in the first line: `rev3866 | ['32.121.25.310'] | Port: 13392 | IP Network: ['ipv4']...`
Nofish commented on Styromaniac's post: Do you have ipv6 connection on mobile? Do you have ipv6 in "IP Network:" list on the /Stats page?
Nofish commented on Styromaniac's post: So you downloaded a big file, then started to download another file with the same hash and it overwritten the downloaded piece status, right?
Nofish commented on Moor's post: Yeah searching in files tab is only supported in py3 version.
Nofish commented on Blurhy's post: No, it's not possible, the iframe and the wrapper required for multiple reasons. (displaying permission requests, sidebar, loading screen, etc.)
Nofish commented on Styromaniac's post: @Styromaniac: Yes, if you wants to be checked, then you should set it to automatic.
Nofish commented on Styromaniac's post: @Styromaniac: It could be possible by checking if there is any incoming connection happened that points to that ip. Right you can do it manually by checking it on the /Stats page
Nofish commented on Styromaniac's post: If you enter an external ip = forcing zeronet client in port opened mode without testing it.
Nofish commented on Styromaniac's post: It should be fixed now, but you have to untick/tick "Follow username mentions" in ZeroMe 3dot menu After that it will display the post creator username at the beginning of the post
Nofish commented on Styromaniac's post: Correct. Btw right now for me it's looks like china is detecting/filtering ZeroNet based on p2p behavior, so I think dht would be also detected.
Nofish commented on Styromaniac's post: what do you mean by instability? the client crashes or other type of problems?
Nofish commented on Styromaniac's post: The posts are timestamped based on the machine time. There is a feature that calculates the possible correct time based on the connected peers time, but right now it's only give notification your time is more than 10 minutes off. It could be possible to add a feature to ZeroTalk (and other sites) that uses this to correct the smaller time differences.
Nofish commented on Styromaniac's post: @Styromaniac: You can delete your sent messages, but not the incoming ones (since they are stored in other user's data file)
Nofish commented on Styromaniac's post: I have fixed both, but on ZeroMail I recommend to delete earlier sent messages if you run out of the space.
Nofish commented on Styromaniac's post: Hm, I have ` "styromaniac@zeroid.bit": {"max_size": 800000},` in data/users/content.json, I just signed and publised again
Nofish commented on Styromaniac's post: When did is started happening?
Nofish commented on ssdifnskdjfnsdjk@zeroid.bit's post: In tor always mode every connection is made using the tor proxy, so local connections will not work.
Nofish commented on leftside's post: Should be fixed in rev3864, please confirm if it's working after the update
Nofish commented on leftside's post: Thanks, I was able to reproduce it, fix is on the way. :)
Nofish commented on leftside's post: I have modified the ZeroHello and added the db error to the stats section. What message do you see bellow that line that ha n/a for exectution time?
Nofish commented on 60's post: As far I know using a GPL licensed module does not requires the whole project to switch to GPL. Like iOS also using many libs that is under open-source license. But if its necessary I can add an MIT license header to it. (probably we should also include GPL license header to every source code file to make it valid)
Nofish commented on leftside's post: @leftside: Do you see the checked "Follow in newsfeed" in the posts's 3dot menu? (should be visible after hovering cursor to the topright corner of the post) At ZeroHello if you press the "from X site in Xms" text, does it have a " ZeroMe Post follow" section?
Nofish commented on leftside's post: There was no change. Do you see the checked "Follow in newsfeed" in the posts's 3dot menu? Also: are you using py3 or py2 version?
Nofish commented on Quantumworld's post: Thanks for the suggestion, `--offline` mode is added to Rev4066 Py3 version. (Also possible to change it on-the-fly on the /Config page)
Nofish commented on Styromaniac's post: You mean after click on a link on a raw page, then pressing the browser back button breaks something?
Nofish commented on 60's post: @60: The idea of the archive file is to be able to remove the separate user files to speed up the initial sync time and the reduce the required number of files to store
Nofish commented on 60's post: It should be changed in 1Udb.../data/usersdb/1Ekf2.../content.json, but for some reasons it does not contains user_name entry, I will check it
Nofish commented on ssdifnskdjfnsdjk@zeroid.bit's post: Do you mean this one: https://github.com/HelloZeroNet/ZeroHello/issues/115 ? I still can't reproduce it.
Nofish commented on Styromaniac's post: Yes, it excludes pinned and big files
Nofish commented on Styromaniac's post: I suspect there is a bug if you have more than 100 pieces in the download queue (updating: x number) I will try to reproduce/find a solution for it after py3 release
Nofish commented on leftside's post: @leftside: The problem is in the markdown syntax parser. (you have to use `[title](link)` format, so the last `)` is problematic) Maybe it could include a feature and detect if there is a `(` in the link, then the closing `))` should consider as part of an url.
Nofish commented on Sirlotaplot's post: Thanks for reporting, fixed in 3846
Nofish commented on Kaffie's post: @Geekless: Probably returning an error is better as you can find out the list of clients who ignores you by try to get your data file anyway. It's a planned feature (I will prioritize it), but right now I want to freeze the source code until py3 transition is done.
Nofish commented on Kaffie's post: Right now muted users only removed from the database, but remain on the file system. There is a proposal to also remove these files: https://github.com/HelloZeroNet/ZeroNet/issues/1855
Nofish commented on Aoghepha90020801's post: I just did!
Nofish commented on LoliGod's post: It should be fixed in Rev3821
Nofish commented on LoliGod's post: @1pgallery: I will try to fix it tomorrow.
Nofish commented on LoliGod's post: @1pgallery: According to the error you have lots of content.json files in your site. The sidebar calls `site.content_manager.listContents()` which should return the list of non-user content.json files. (normally its just a few for every site) It's possible to fix the query, but do you really need this large ammount of content.json files?
Nofish commented on ssdifnskdjfnsdjk@zeroid.bit's post: I have added a "RECENTLY DOWNLOADED" section to the ZeroHello (lists pages that finished downloading in last 3 days)
Nofish commented on ssdifnskdjfnsdjk@zeroid.bit's post: We could have a "Recently added" section for sites that were added and successfully downloaded in last X days
Nofish commented on P2P's post: Thanks for reporting, some more information would be nice: - How many trackers have low success rate (<20%) on ZeroHello with and without meek proxies? - Does it only affect new users or older ones also unable to connect? - How many connections do you see in /Stats page? Unfortunately the previously used remote desktop not working anymore, it would be great help if someone would able to provide a ssh shell or an rdp access.
Nofish commented on Tusiko's post: Probably this is a limitation in Chrome and multi-threaded db will not improve on this: Chrome allows only 6 pending requests per host. So if you have a site that includes 6 unreachable/slow optional files, then the Chrome will block all requests until that 6 one times out. Firefox handles these situations better. (cancels when you navigate away from the page)
Nofish commented on Znnex's post: Please try to update in latest Rev, there a bugfix in Rev3737 that affects the muting feature
Nofish commented on Tusiko's post: Thanks for reporting I was able to reproduce and fix it in Rev3739
Nofish commented on Tusiko's post: @Styromaniac: Yeah, any idea how to reproduce it?
Nofish commented on Moor's post: Thanks for reporting fixed in Rev3729. The problem was with the win-style data_dir configuration value. (I think it also affected earlier versions of ZeroNet)
Nofish commented on Pexo's post: @Pexo: you can use gitcenter for that: http://127.0.0.1:43110/1GitLiXB6t5r8vuU2zC6a8GYj9ME6HMQ4t/
Nofish commented on Manwe's post: I just tested and it's says new version available for me. Do you see this message in the 3dot menu on ZeroHello? (Under "Show data directory"?)
Nofish commented on Pexo's post: @Pexo: Thanks for reporting I will check it
Nofish commented on Styromaniac's post: @Styromaniac: Fixed!
Nofish commented on Styromaniac's post: Thanks for reporting, it should be fixed now
Nofish liked Pexo's post
Nofish commented on Pexo's post: You are right, it should be the same. I just added a separate send s button. So you can use tab, enter or ctrl+enter to send the message.
Nofish commented on Styromaniac's post: Thanks, fixed!
Nofish commented on Kaffie's post: It should work, currently there is around 10 trackers can be found using the shared trackers feature.
Nofish commented on Yuzu's post: Thanks for reporting, fixed in Rev3568
Nofish commented on Musickillerwork's post: @Musickillerwork: Fixed! (it was limited to 25 character)
Nofish commented on Musickiller's post: Thanks for reporting fixed! (changed to `text.replace(/(\s|>|^)(@[^\s]{1,25}):/g, '$1<b class="reply-name">$2</b>:')`)
Nofish commented on Musickiller's post: @styromaniac@zeroid.bit: Fixed! :) (it did not matched if the comment began with the username)
Nofish commented on Musickiller's post: Thanks for reporting, changed the username-bolding regexp, it should work now. @styromaniac@zeroid.bit: Test 1, @Musickiller: Test 2
Nofish commented on Styromaniac's post: Thanks to lola's fix it should work again in Rev3537
Nofish commented on Binchan's post: @Binchan: It's the same problem: there is no peers connected to your client. (You can verify it using the http://127.0.0.1:43110/Stats) The client tries to keep 6 connection per site by default. Do you have lines "Need connections:" in your log/debug.log file? (there should be one for each site in every 5 minute)
Nofish commented on Binchan's post: That error means there is no connected peers. You can ignore it
Nofish commented on version@zeroid.bit's post: @P2P: I just tested it using a Mainland China remote access I got some week ago and I was able to ping my home ZeroNet node without any proxy/vpn/tor. And looks like they also lifted the ban from port 15441 as it's also working again. I think the issue is with the trackers.
Nofish commented on Sgushennayavoda's post: When/where do you see that error?
Nofish commented on Styromaniac's post: @styromaniac@zeroid.bit: yes: http://127.0.0.1:43110/Stats
Nofish commented on Styromaniac's post: Do you have normal amount (100+) of peers for the site? Do you see tls-rsa encrypted connections on /Stats page?
Nofish commented on P2P's post: @P2P: Yeah that wasn't pretty, fixed in Rev3406
Nofish commented on 猫叉酱's post: Thanks for the the info, looks like it's works again. I will post more info on ZeroBlog later today.
Nofish commented on P2P's post: @P2P: That's the next step, I will check how does shadowsocks works. New Rev3402 out with port randomization and fixed implicit tls. I just tested using the China VPS I got (thanks again) and looks like it's works again, but unfortunately you can only connect to peers that using non-standard port (15441). I will post about it on ZeroBlog later today to increase the updated (and by that non-standard port) clients.
Nofish commented on P2P's post: @P2P: The cert itself is random, but some of the details fixed right now (eg. it's issued to "Example Company") Later we can also randomize these details, but creating a DPI rule based on it is more complicated and probably it's not necessary yet.
Nofish commented on P2P's post: Thanks, I already got one and using that I was able to confirm both DPI and Port blacklisting. The implicit SSL (make data transfer looks like https connection) works against DPI. I'm working on adding port randomization, it should be out today.
Nofish liked P2P's post
Nofish commented on Binchan's post: I have just increased the optional limit to 100MB/user
Nofish commented on ZAlex's post: Thanks for reporting, it should be fine now
Nofish commented on Hermes32's post: You can fix hugo urls by adding `relativeurls = true` and `baseURL = ""` to your config. @Binchan: Namecoin only allows 512byte value, so it won't fit. Maybe it could be solved by adding "*" entry support and then the javascript would redirect to the given site.
Nofish commented on Binchan's post: @Version: Yes, it's legal here. They tried to ban the encrypted communication 2 years ago (https://www.deepdotweb.com/2016/05/16/hungary-passes-new-law-to-ban-full-encryption/), but it did not passed. There will be elections next month, probably the current (not-so liberal) ruling party will win for the 3rd time in row. I expect it to be worse after that.
Nofish commented on Git Center's post: @Git Center: Oh I thought it's just for cache or other meaningless data. Well then, think the best solution is create a random id, store using UserSetSettings and write it to that directory/file. This way it's also multi-user compatible.
Nofish commented on Binchan's post: It was one of them :) Others: - https://www.engadget.com/2014/12/16/pirate-bay-shutdown-explainer/ - http://blog.bittorrent.com/2014/12/10/project-maelstrom-the-internet-we-build-next/
Nofish commented on Git Center's post: @Git Center: Yeah, it will, so you can use it to store some private data, but it's not recommended because it won't work well in multi-user environments.
Nofish commented on Git Center's post: @Git Center: site.storage.getPath does not allows ".." in the file path, but if you find any security issue please report it. (preferably a private way)
Nofish commented on Deadjoker's post: @Binchan: I think an rss would be more suitable for third-party application notifications.
Nofish commented on Deadjoker's post: Currently the site's notification feed is defined by a sql request. To make zeromail notifications possible we need to allow javascript logic to return the latest message list. It could be possible by extending the newsfeed plugin to support javascript function call signups. After that when the user visits ZeroHello, the plugin would create a sandboxed iframe to load and run the newsfeed returning javascript function and then inserts the items returned to the newsfeed.
Nofish commented on Jro's post: Please create a github issue with the details if you can reproduce it on multiple machines.
Nofish commented on Binchan's post: I also got lots of reach from ICO projects, but I decided to keep the project indepentent from third-party money. If there is a reason I'm happy to use the donated bitcoins to support zeronet-related projects, but I'm not sure if money has positible effect on open source projects. Also I prefer if somone developing something on moral motivates instead of financial.
Nofish commented on Nofish's post: @styromaniac@zeroid.bit: Can you please give more detail on that problem?
Nofish commented on Nofish's post: @Binchan: Yeah, will do it tonight :)
Nofish commented on Styromaniac's post: Thanks, fixed! :)
Nofish commented on Styromaniac's post: Thanks, fixed!
Nofish commented on Styromaniac's post: There was no change that affect cpu usage, but you can try to rollback to previous version.
Nofish commented on Binchan's post: Actually firefox always worked: http://127.0.0.1:43110/kaffie.bit/test.html
Nofish commented on Styromaniac's post: There is no bw prioritization / limitation built-in yet, I can't think any easy way to do that, so I need some more research on that.
Nofish commented on Nofish's post: @ZeroLSTN: "What's the problem with having many different domains available?" It's not a problem, the PoW required because it prevents potential bad actors creating new ID provider in no time, so making the black list ineffective. @Kaffie: I not sure if I want to make the ID system to depend on the NMC, because if anything goes wrong, then it would affect many sites.
Nofish commented on Nofish's post: @Kaffie: Accepting all by default is makes the blacklist useless as a potential bad actor can create 1000s of new cert provider per second. As a new user you don't have to do the proof-of-work, but you can use any other existing cert provider.
Nofish commented on Nofish's post: @ZeroLSTN: The current, white-list based system will here to stay, but the new system will accept IDs from unknown provider if the provider's address meets the requirements (matches the required regex pattern) specified in the content.json
Nofish commented on Nofish's post: My proposal for Proof of work based ID providers: https://github.com/HelloZeroNet/ZeroNet/issues/1258 I just generated a 1ZeroiDJnkHkugPNd8UzSwceH8HfsnYtC address in 3 hours, so maybe we should make it harder like "1ZeroiD[0-9]" should take ~5days on my machine
Nofish commented on Musickiller's post: @styromaniac: I don't think that any other plugin has significant impact (not sure about the Charts plugin either)
Nofish commented on Nofish's post: @Jro: Sure, the current system will here to stay, but there will be a possibility to also accept unknown zeroid providers with the specified prefix. (I propose 1ZeroiD, but it will be configurable by the site owner)
Nofish commented on Git Center's post: Have you signed your main site before?
Nofish liked Lxpz's post
Nofish commented on Musickiller's post: You can try disabling the stats module by renaming plugins/Chart to plugnis/disabled-Chart. If it causes the problems, then we could add an option to reduce the sampling rate (currently 5 minute) or disable to most cpu hungry ones (peer statistics) In the log/debug.log you can search for "Global collectors done in" to see how much cpu time it takes to collect the data. For me it's 6-12ms + inserting 1ms
Nofish commented on Nofish's post: @Kaffie: The problem with the id provider black list instead of white is it's also easy to create an id provider. As a solution we could require some intensive proof-of-work from id providers, for example it has to start with "Zeroid", witch takes around 8 hour to generate on a ~200USD GPU or 30 USD on https://bitcoinvanitygen.com/
Nofish commented on Styromaniac's post: It stores your selected profile in browser's local storage (similar to cookies)
Nofish commented on Jamesholden's post: Not sure, what happens when you press the Sign button on the sidebar?
Nofish commented on Binchan's post: Yes, currently you can't share a file named content.json file, as a standard file. I will think about how to solve it...
Nofish commented on Glightstar's post: Yeah, nice spot, object arguments not supported on that call
Nofish commented on Kaffie's post: @Kaffie: It's only ipv4 right now any you need to maintain the bootstrapper list by hand (you can distribute the config file with zeronet), but peer exchange is supported within one site, so the bootsrapper only needed if you want to visit a site that you have not seen before.
Nofish commented on Kaffie's post: @Kaffie: DHT is not required for meshnet support, you can run ZeroNet right now on meshnet by starting a bootstrap node (enable plugin, add address to client's trackers list)
Nofish commented on Kaffie's post: @Kaffie: All would be nice, but DHT does not works well on Tor and IPV6 is not supported by most of torrent trackers
Nofish commented on Kaffie's post: @Kaffie: I2P support would be nice, but it's not a replacement for Tor connection, as it don't support connections outside of the I2P network, so you won't be able to use it to connect sites that does not have I2P peers. (and the last time I tried it it was much slower, than tor)
Nofish commented on Glightstar's post: It was more tricky, than I originally thought, but added in Rev3176. To take advantage you should have your directory in the ignore pattern followed by `/.*`. So having `data/users/.*` as your ignore pattern will be fast, but `data/users/.*/.*` or `data/users` won't.
Nofish commented on Nofish's post: @Random321: I started it as a hobby project without any expectations, so yes :)
Nofish commented on Glightstar's post: Yeah, great idea, added to TODO :)
Nofish commented on Balancer73's post: Thanks for the suggestion, from Rev3173 it won't ask for permission if it already has it + fixed the NoneType error
Nofish commented on Nofish's post: Got the laptop (Xiaomi Air 12), the build quality is nice, core m3 cpu performance is also perfect, after found a way to install precision touchpad drivers (took 6h) i'm satisfied with it. Have not tried any development on it yet, but I'm pretty optimistic :)
Nofish commented on Herp's post: @Krixano: I got the notification, so it's not required
Nofish commented on Git Center's post: The feedQuery API call tries to parse the query, add some optimization (date_added > now -3 days) and ORDER BY date_added DESC LIMIT 10 and it does not support sub-queries with union atm.
Nofish commented on 霸王寨第一光头's post: @霸王寨第一光头: There is no built-in yet, but here is a proof of concept code to use websocket api from python: https://github.com/HelloZeroNet/ZeroNet/issues/1137#issuecomment-337019878
Nofish commented on 霸王寨第一光头's post: Not currently, you can edit data/mutes.json, but it requires to restart zeronet
Nofish commented on 猫叉酱's post: @Git Center: Every spam list would have one maintainer, so he/she would decide who is going to be added.
Nofish commented on 猫叉酱's post: Thanks for reporting, I think I cleaned up the hubs i have. I think the right solution would be shared/subscribable mute list.
Nofish commented on Krixano's post: @Git Center: Packed + updated the repo, it's smaller and uses less files, but it really needs caching: clicking on v0.6.0 tag loads the `pack-33e03b768ec585e19ceb5401c8d1f84b425c57ff.pack` file (1MB) 70 times with fileGet
Nofish started following ulrichard, U53r, Glightstar
Nofish commented on Glightstar's post: data/users.json settings node within site's part
Nofish commented on Git Center's post: @Glightstar: Just fixed it in rev3157, please update and try again @Git Center: For big files probably ajax would the best as it transfers the files asnyc from websocket, support ranged queries and don't have to pack/unpack base64.
Nofish commented on Git Center's post: @Krixano: using different address for sidebar is a good idea I will restore the redirect. I just tried to pack the repository with `git repack --max-pack-size 1m`. It's a bit slower this way, maybe a read cache could help to avoid re-reading big files using fileGet. The bigger files also triggered an error when reading files from different sites at the same time. (already fixed, but need some more testing, will roll out tomorrow) Btw. why does it needs to read ZeroID site?
Nofish commented on Git Center's post: I changed to link, because I need the sidebar for information and signing. @Sirenyc: Please update to latest revision. There was a problem that affected sites larger than 10MB, but it's fixed now.
Nofish commented on Binchan's post: @Binchan: Not yet. The json files are written by the site's javascript source code, so the client itself can't modify it's content. Maybe it could be solved by adding a placeholder for the timestamp, eg. `{{ publish_timestamp }}`, but not sure if it's necessary.
Nofish commented on Git Center's post: Yes, nice discovery! :)
Nofish commented on mv00_'s post: @mv00_: I'm afraid your comment is not visible here. Try changing your identity to a zeroid one and comment again.
Nofish commented on BTC-E's post: I'm afraid big files won't work well without opened port. Enablind Tor connection helps, but it will be much slower and only allows other tor-connected users to download your file.
Nofish commented on mv00_'s post: Probably it's because the hub i using does not accepts 0ch id provider.
Nofish commented on Nofish's post: Thanks for the suggestions! Eventually I went for Xiaomi Air 12.5" as every review I found was positive about it. The price was also pretty low (560USD) so I don't have to be super careful about it. :)
Nofish commented on Binchan's post: @Binchan: Since 0.6.0 the valid path is `"^[a-z\[\]\(\) A-Z0-9_@=\.\+-/]+$"`, so it's includes `" "`, but not other whitespace characters
Nofish commented on Binchan's post: @Git Center: Why do you want to change it? It allows the space character.
Nofish commented on Binchan's post: @Binchan: It's using pyhon search engine https://docs.python.org/2/library/re.html
Nofish commented on Binchan's post: It should work since 0.6.0 for example: http://127.0.0.1:43110/1uPLoaDwKzP6MCGoVzw48r4pxawRBdmQc/data/users/16YxtnaTrCSbvCXSYxrnvobmYnqtidxG23/Charlie%20Chaplin%20-%20Final%20Speech%20from%20The%20Great%20Dictator-J7GY1Xg6X20.mp4
Nofish commented on mv00_'s post: The mute based on auth_address,so the user name is irrelevant
Nofish commented on Fengyun007's post: @Krixano: Yeah, currently it merges the files signed by the same auth_address on different hubs and yes, probably he/she registered the same address multiple time
Nofish commented on Git Center's post: Looks like there is some problem with size calculation for sites larger than 10MB (you have to change the limit to 30MB+ manually to finish the download), but once it finished downloading it's working fast and fine, grats! (I will try to fix the size calc error tomorrow)
Nofish commented on mv00_'s post: @mv00_: The log looks normal. You had higher peer number before the fix, because the upnp bug blocked the peer cleanup method.
Nofish commented on 西木野真姬's post: Yes, supported trackers: bittorrent http(s), bittorrent udp, custom zeronet
Nofish commented on Musickiller's post: does pressing update on redhub helps?
Nofish commented on Musickiller's post: It should be better now!
Nofish commented on Styromaniac's post: Thanks for reporting, it should be fixed in Rev3146
Nofish commented on mv00_'s post: @mv00_: Thanks, (hopefully) It's fixed in Rev3144, please update & watch for errors
Nofish commented on Musickiller's post: Maybe we could store longer posts in optional files on ZeroMe
Nofish commented on mv00_'s post: @mv00_: Thanks fixed in Rev3141, Is there any other "Unhandled exception"?
Nofish commented on mv00_'s post: @mv00_: That's weird way too many connection and for some reason the dead ones did not get closed automatically. Is there any match if you search for "unhandled" in file log/debug.log ?
Nofish commented on Musickiller's post: The space is more limited here, so if you have any longer thoughts i recommend to start a blog and put a link to that post.
Nofish commented on mv00_'s post: When the updates stops reaching your client how many connection you see in http://127.0.0.1:43110/Stats ?
Nofish commented on mv00_'s post: @mv00_: It wait 2-10 seconds between sites. So it should not take that long. How many site you have in you client?
Nofish commented on mv00_'s post: It takes some time, but it should do it.
Nofish commented on Git Center's post: Yeah, I will check it soon
Nofish commented on Krixano's post: When you are running in --debug mode and you have "This is my site" enabled and your client requests all.js (via http), then your client will merge all *.js and *.coffee in that directory into one file and try to compile .coffee files (using the command specified in `--coffeescript_compiler` or `coffee --no-header -p` if it's undefined) It's pretty simple (just to fit my needs) and probably you want to use [webpack](https://webpack.github.io/) or similar that has more features (jsx, babel, etc.)
Nofish commented on Krixano's post: Yes, the id provider only issues a digital signature that is acts as proof that you registered that username
Nofish commented on Bingyuhuo's post: Thanks, fixed!
Nofish commented on Sksn's post: Changing the address is possible, but the permissions are recorded to site's address not to the domain.
Nofish commented on Krixano's post: The "maps" are relative to the .db file, so in this case anything outside of "data" won't match
Nofish commented on Git Center's post: For me the git repo sites (eg.: 1GTispUAh7VLQjLAkKCUmnpJA7jU6K7Grh) are not loading. (looks like nobody has the content.json file) Can you sign and publish them again?
Nofish commented on Krixano's post: It won't delete it automatically if it runs out of optional files limit + does not counts in optional files limit
Nofish commented on Krixano's post: Is it a big file or a regular file? If you use fileNeed on a bigfile by default it only downloads the piecemap file. To download the whole file you have to add "|all" after the filename when you call the fileNeed request. For regular files accessing it via linking should be the same as fileNeed
Nofish commented on Krixano's post: @Krixano: Probably the table sturcture was out of sync. If you change that value it will delete and re-create that table.
Nofish commented on Musickiller's post: Per-post file would be painful to store/transfer/read. Per-device file is a great idea, but currently only one content.json supported per directory, so not really possible. Currently working on partial solution that check if the file is fully updated before starts writing to it. > Backup is a good idea, but would that not be space/time consuming? We could only keep last ~5 backup.
Nofish commented on Musickiller's post: That's sucks, normally it should give a warning about writing to file that is still in sync. Can you please describe how did that happen? My idea to avoid situation like this in the future: When you modify your file using the API it would automatically create a backup of that file (eg. data.json-bk-1509441647), so if anything goes bad you would able to restore from it.
Nofish commented on 霸王寨第一光头's post: @P2P: Unfortunately right now this is a pretty hdd sensitive function so i'm not sure if it's a good idea, but I will try to find a way of the optimization to make it faster.
Nofish commented on Comeon's post: Unfortunately it's not possible if we want unique usernames.
Nofish commented on 霸王寨第一光头's post: @霸王寨第一光头: You can try to download it again and copy the data directory to the new client to see if it's still happening
Nofish commented on 霸王寨第一光头's post: @霸王寨第一光头: I tried to reproduce it with different Tor versions, but never got that error. You can try to update your tor client to latest version (0.3.1.7) (the you can find the instructions in tools/tor directory)
Nofish commented on 霸王寨第一光头's post: It's not necessary a problem, the not connected peers just dead
Nofish commented on Musickiller's post: Probably it can't detect the free space on your device
Nofish commented on Musickiller's post: sublime text
Nofish commented on Musickiller's post: I made some improvement, on this. It's still now perfect, but at least it's not fall apart
Nofish commented on Musickiller's post: It should be fine now
Nofish commented on 霸王寨第一光头's post: Whats the version of the Tor client? You can check it in log/debug.log, search for "VERSION Tor"
Nofish commented on 西木野真姬's post: It's working here, do you see any error in js console? (F12)
Nofish commented on 空澄砂夜's post: @空澄砂夜: Yes it's kind of similar the only real difference is it does not reaches the hard disk. Maybe we could add some POW requirement to it
Nofish commented on 空澄砂夜's post: Yes, here is some ideas: https://github.com/HelloZeroNet/ZeroNet/issues/1145#issuecomment-336490090
Nofish commented on Musickiller's post: Just added a `⋮` menu button that appears when you hover the over file item in ZeroUp and allows you to delete the file without leaving the site.
Nofish commented on Nofish's post: @43110: It could be easy to add random port listening, but I have not found any precedent GFW blocking based on port number. Any opinions on that?
Nofish started following Bluishcoder
Nofish commented on 霸王寨第一光头's post: Thanks for reporting fixed!
Nofish commented on Netwatcher443's post: Looks like we got *.zeronet.io domain + 104.156.231.236 ip ban Any suggestions on solution?
Nofish commented on Netwatcher443's post: @Comeon: ZeroBoard was the first site on ZeroNet when multi-user sites was not possible, so the messages are signed and distributed by contacting the site owner. I was planning to remove it for a while maybe it's the a good time to do it.
Nofish commented on Thiece's post: It's fixed in Rev3102, thanks again for reporting.
Nofish commented on Thiece's post: Looks like the stats counter is broken, working on the fix...
Nofish commented on Musickiller's post: @Musickiller: Got it, it will be fixed in the next release
Nofish commented on Musickiller's post: Unfortunately the site does not works for me. It tries to download user files from data/users/ directory, but the content.json specifies it to data/ directory
Nofish commented on Musickiller's post: It should happen if you try to open a file from a site that you have not visited before.
Nofish commented on 空澄砂夜's post: @空澄砂夜: There is a built-in css merger in zeronet that combintes css files into all.css and also adds vendor prefixes (-webkit-, -moz-, etc.) when necessary. Other than that it's written by hand.
Nofish commented on 空澄砂夜's post: I using sublime text for every kind of file
Nofish commented on 霸王寨第一光头's post: Just increased the limit!
Nofish commented on Elspru's post: @Elspru: Actually you need to update the UiRequest.py file: https://github.com/HelloZeroNet/ZeroNet/commit/bfd3d18a10e3cceddf88946dd4567e51acfd7200
Nofish commented on Elspru's post: It's working for me. There was an error with the Multiuser plugin, but it's fixed in the latest version.
Nofish commented on Musickiller's post: Thanks for the suggestion it, it's a good idea and there is lots of space of improvements on zeromail, but i'm afraid I won't have time to improve it in the next months.
Nofish commented on Musickiller's post: There is no button like that, but if you set the storage limit lower it will automatically delete the most seeded files. (your client's webui last access time also counts)
Nofish commented on Nofish's post: @Freedonn: https://zeronet.readthedocs.io/en/latest/faq/#how-to-create-a-fancy-site-address
Nofish commented on Krixano's post: @Binchan2: It's hard to tell based on this information. Have you opened ZeroHello when you experienced the hang? (loading the newsfeed could block the running on slow hdds) What the DB speeds for you on http://127.0.0.1:43110/Benchmark ? There is some files that does not distributed well on hubs. You can ignore these errors usually these are optional files. (you can check the filename on site's sidebar)
Nofish commented on Krixano's post: @Nofish: My results: http://127.0.0.1:43110/1MgHVPCE1ve6QfKrgsqCURzRj72HrRWioz/?1J3rJ8ecnwH2EPYa6MrgZttBNc61ACFiCj_1 Python 32bit: - openssl verify x 100..........0.123s [x3.01: WOW] - sha512 5M x 10..........0.213s [x2.82: WOW] - Buffered insert x 100 x 100..........0.533s [x2.44: Fast] - Indexed query x 1000..........0.082s [x3.05: WOW] Python 64bit: - openssl verify x 100..........0.070s [x5.29: Insane!!] - sha512 5M x 10..........0.085s [x7.06: Insane!!] - Buffered insert x 100 x 100..........0.476s [x2.73: WOW] - Indexed query x 1000..........0.069s [x3.62: Insane!!] So the cryptography is 80-250% faster, but db only 20% faster on 64bit vs 32bit.
Nofish commented on Krixano's post: @Krixano: The communication is Websocket based that has client for most of the language. All you need is a wrapper_key as http get parameter that can be obtained from site's wrapper http source or from sites.json file
Nofish commented on Krixano's post: @Krixano: Yeah it doesn't looks good, the crypto speed could be blame on 32bit vs 64bit, but not sure why the database is slower. Is it running from same HDD as on Linux? @P2P: Yeah, it's block :)
Nofish commented on Krixano's post: @Krixano: Actually I'm using and developing it on Windows 10. To maximize compatibility currently we using 32bit Python switching to 64bit would significantly (up to 100%) speed up cryptography functions. Is there any value under 1.0x on http://127.0.0.1:43110/Benchmark for you?
Nofish commented on Krixano's post: - Sometimes the image loading can be slow if it has only a few seeder. Usually the images loads within some seconds for me. - Right now ZeroNet is single threaded, it means if some site doing CPU/HDD intensive stuff, then it's also blocks the UI. If you start zeronet using `--debug_gevent` parameter, then it will logs "Gevent blog detected" messages when this happens. To avoid this I'm planning to CPU cost stats to sites and I also try to move SQL to different thread.
Nofish commented on Nofish's post: It went relatively well (still having hard time speaking in front of other people) and there were video recording, but I have no eta on the video yet.
Nofish commented on Musickiller's post: - No idea how it happened. I just checked user json files (25k) on all hubs I serving and only found one another invalid json file. (it looks like edited by hand) - I have seen a comment on your post in the activity list, but after clicking on it displayed "post not found"
Nofish commented on Musickiller's post: @Musickiller: There is a json error in your data.json file on Red hub @ line 587: `"body"?à"So Ubuntu Wiki did`. Any idea how did that happened?
Nofish commented on Nofish's post: @Ivesen: No, not yet.
Nofish commented on Nofish's post: @Ivesen: Yes it will be in english and I think there will be a video recording.
Nofish commented on Jabasukuriputo Wang's post: I'm still using it. It can happen if you try to use it on client that runs using `--ui_ip "*"` and does not added zero to allowed domains using `--ui_host zero`
Nofish commented on Musickiller's post: @Musickiller: Because nobody implemented it
Nofish commented on Musickiller's post: @Musickiller: I'm afraid right now there is no way to attach image to an existing post using the web interface.
Nofish commented on ulrichard's post: @ulrichard: Just move the monkeyPatchAjax part before loading the OpenLayers.js file and it will work!
Nofish commented on Musickiller's post: Nice, does it displays the top-right "x" button? (and does it works?)
Nofish commented on Musickiller's post: Yes, atm it's 16 character max: https://github.com/HelloZeroNet/ZeroMe/blob/master/js/utils/Text.coffee#L23
Nofish commented on Musickiller's post: If you update to Rev2169+, then you will have a top-right "X" close button on the sidebar.
Nofish commented on Krixano's post: Normally it should be unique, but it's the ID provider responsibility to make sure no other user has the same name, so I recommend using auth_address + increasing id (or timestamp) as identifier.
Nofish commented on ulrichard's post: Unfortunately Ajax requests not working anymore: For some reasons the browsers does not allows Ajax call from sandboxed iframe to files without `Access-Control-Allow-Origin: *` http header, which would allow cross-site resource access without any limitation. To fix it normally you only need to add these lines to your site: ``` <script src="ZeroFrame.js"></script> <script>zero_frame = new ZeroFrame(); zero_frame.monkeyPatchAjax();</script> ``` But looks like your site is using deprecated sync Ajax call which is not possible to "emulate" using [ZeroFrame.js](http://127.0.0.1:43110/1HeLLo4uzjaLetFx6NH3PMwFP3qbRbTf3D/template-new/js/ZeroFrame.js), so you need other modifications. Here is the modified index.html using async/await: http://127.0.0.1:43110/1MgHVPCE1ve6QfKrgsqCURzRj72HrRWioz/?1J3rJ8ecnwH2EPYa6MrgZttBNc61ACFiCj_0 I'm still thinking about optional content.json parameter that adds `Access-Control-Allow-Origin: *` http header to files and by that allow Ajax requests without modification of html source code, but it could be privacy issue as any site would able read these files and by that see which other sites are currently seeded by the user. Any opinions?
Nofish commented on Musickiller's post: Thanks, [merged stbinan's patch](https://github.com/HelloZeroNet/ZeroMe/pull/97)
Nofish commented on Binchan's post: @Styromaniac: Most of the work done by @Krixano, I just polised it a bit more :)
Nofish commented on Styromaniac's post: Thanks for the suggestion, moved the tirrger to the bottom on zerohello
Nofish commented on Nofish's post: @Styromaniac: I have no experience in Android development, so I dont know
Nofish commented on Krixano's post: onpopstate won't work because of the iframe, but the [wrapper passes it to the inner frame](https://github.com/HelloZeroNet/ZeroNet/blob/master/src/Ui/media/Wrapper.coffee#L36) and you can catch it using [onRequest](https://github.com/HelloZeroNet/ZeroMe/blob/master/js/ZeroMe.coffee#L261) (btw you can easily convert [coffee to es6](http://decaffeinate-project.org/repl/) if you prefer that syntax)
Nofish commented on Fengyun007's post: @Fengyun007: Thanks, got it, investigating...
Nofish commented on Fengyun007's post: Can you please send your data/content.db file to hello@zeronet.io ?
Nofish commented on Glightstar's post: @Glightstar: Sorry, I was wrong, the url should be the same not the title, so for example change it to `"" as url`
Nofish commented on Glightstar's post: It automatically group it by title, so I recommend using "New comment" as title and username to the body
Nofish commented on P2P's post: Yeah, it's getting problematic. I will write a plan for it this week.
Nofish commented on Musickiller's post: Yeah unfortunately it does not works well with 1000+ contacts. I just restarted the bot and it did replied my test after that.
Nofish commented on Krixano's post: It doesn't sounds good. What operating system are you using? Stopping ZeroNet, deleting data/content.db, then start it again should work. (it takes some minute to re-index the file)
Nofish commented on Nofish's post: @P2P: github shows http://fuliba.net/zeronet.html as source of traffic, but it's unavailable for me.
Nofish commented on Kaffie's post: @jro: For example `re.match("(\w+\d+)+C", "112649159593822679584363.json")` takes 0.7s on my machine, so forcing `.` before `+*{` was the only possible solution I found that still offers backward compatibility for most of the sites.
Nofish commented on Kaffie's post: @Kaffie: I got a security report on CORS, so I fixed on it. @jro: Here is the discussion on redos, unfortunately timeout is not possible: https://github.com/HelloZeroNet/ZeroNet/issues/989, but if anyone can find a better solution i'm happy to change it.
Nofish commented on Peeper's post: @Kaffie: Thanks for the feedback, I'm totally up to this,here is the discussion: https://github.com/HelloZeroNet/ZeroNet/issues/1026
Nofish commented on Peeper's post: @Peeper: Yes if it's already seeded by the client: ``` Page.cmd("mergerSiteList", {query_site_info:true}, function (res) { console.log(res["1BLueGvui1GdbtsjcKqCf4F67uKfritG49"]["peers"]) }) ```
Nofish commented on Nofish's post: @TCP: There is no option for it yet, but created an isse for it: https://github.com/HelloZeroNet/ZeroNet/issues/1021
Nofish commented on Nofish's post: @TCP: The encryption is optional currently and fall backs to unencrypted channel if ssl handshake is failed.
Nofish commented on Krixano's post: @Krixano: Corrent, but the digital signature is generated for content.json file without the "signs" node of the json.
Nofish commented on ZAlex's post: @ZAlex: Thx, fixed in rev2154
Nofish commented on ZAlex's post: Yeah found it. :) It was fixed in rev2151, so please update: https://github.com/HelloZeroNet/ZeroNet/commit/2777c4c5370963ad89221d12eaf53a9d5b28f0ce I think I will release a new version later today.
Nofish commented on Peeper's post: There is a bit more peer than usual, but I don't see any suspicious thing. It's an opensource protocol/app, so you can't do much about making custom modifications, but I don't see any problem with that. If someone does not have spare bandwidth to seed sites, then he/she can use a proxy or simply put it on paused state.
Nofish liked Lola's post
Nofish commented on Peeper's post: Some ideas if we want to spend it: - Get audit from professional security firm. - Create bug-bounty. (Actually I already sent some bounty for bug reporters) - Hire someone for specific task - Create ZeroNet site contest and support content creators
Nofish commented on Nofish's post: @Peeper: I think it's better if it's a separate site, so anyone can clone it and create own version of it.
Nofish commented on 美杜萨's post: It's looks like the proxy is pretty overloaded. The owner can increase it using the `--max_file_opened` argument. Other tips: https://serverfault.com/questions/20387/debian-too-many-open-files
Nofish commented on Styromaniac's post: @P2P: I was able to reproduce the problem (in VMWare + RemixOS 3.0) and I think this commit will fix it: https://github.com/HelloZeroNet/ZeroNet/commit/f30b2b6fc2b799d0bd3e1e81b68634bb1418af61
Nofish commented on Styromaniac's post: Probably compiling and bundling openssl libs with ZeroNet-kivy would be the solution. It would also reduce the cpu usage of the verification of the files by 20 times.
Nofish commented on Styromaniac's post: What does OpenSSL verify line says on http:/127.0.0.1:43110/Benchmark page? Do you have `ssl*`, `crypto*`, `libcrypto*` file in /storage/emulated/0/lib dir?
Nofish commented on 老司机福利's post: @P2P: Thanks for reporting, Rev2127 should fix it.
Nofish commented on Yuzu's post: It's possible if there is a LaTeX rendering lib in js.
Nofish commented on Forza's post: Two idea for this: - Add mandatory language selection when creating the topic and implement a ZeroSites-like filter - Or automatically move foreign language topics to a sub-topic. Not sure yet which one is better...
Nofish commented on Forza's post: You mean here, in ZeroMe? Yeah I was thinking about it and already did some performance tests.
Nofish started following Antyhrabia
Nofish commented on Nofish's post: @Antyhrabia: I just tried and it works for me in FF 54.0 with Multiprocess Windows 1/1 (Enabled by default) And still working after changing the settings, maybe some plugin causing the problem.
Nofish commented on Musickiller's post: @Musickiller: If you are the only seeder of a new site, then it will mark the "content.json" file as possible bad file, because it could not check if you have the latest version or not. Maybe in this case we could hide this message from the UI to avoid confusion.
Nofish commented on Musickiller's post: What you mean by "copy your blog", you cloning it or just visiting it? Which one is the missing file? (you can check the file on the site's sidebar)
Nofish commented on 西木野真姬's post: Can you please explain what you mean by "privatekey sometime will automatic missing problem" ?
Nofish commented on Laithal's post: Yeah, sometimes it decides not to answer, but just did some modification, please try again.
Nofish commented on Nofish's post: Changed: - Arranged categories (other suggestions are welcome) - "Show more" expands in-place (you can reach the category view by clicking on the category title) - Admin view if you have "This is my site" enabled to edit other users submitted links - Changed ordering to use the number of stars - The Other category expands to 3 column - Added sites.zeronetwork.bit domain
Nofish commented on Nofish's post: Changelog: - Added site editing and deleting. Please move the already submitted sites to the new categories! - Notifies if you want to add already submitted site @Kaffie: It need some to move the sites from the Other category to the new ones. (eg. I think you can move the search to "Services" category) I will try to find better way display for other category (eg using 3 column instead of one)
Nofish commented on ks's post: @P2P: Probably depends on shipping method, but free shipping (by simple mail) is around 3 weeks.
Nofish commented on Nofish's post: Added new categories and other features. Editing feature still missing, but coming soon :)
Nofish commented on ks's post: I have used https://wirexapp.com/ around 1 year ago as virtual card to buy something from aliexpress and it worked well.
Nofish commented on Nofish's post: @Mkg20001: Yeah, that's the goal!
Nofish commented on Nofish's post: @Booma: You can filter sites based on language, but clones are also welcome!
Nofish commented on Styromaniac's post: There was an error related to the port checking. It's fixed in Rev2088, so updating to that version probably fixes the error.
Nofish commented on Balancer73's post: I'm currently working on a language separated site directory where anyone can submit new addresses, so probably it will direct the traffic to other sites as well. @Binchan: To keep the site size limit the comments older than 200 days periodically got deleted. In the next few months I'm planning to implement an archive feature that will allow users to decide if they want to download, see and serve old data or not.
Nofish commented on Balancer73's post: @Nofishes: I'm open for suggestions, the "Please use English, you can reach other language forums at the top of this page" warning at topic creation does not seems to be working
Nofish commented on Nofish's post: @Zer0surfer: I usually wait some days before tagging it for bugs/fine-tunes
Nofish commented on Nofish's post: @Skwerlman: Yeah, I have been thinking about it, it would not be hard to add and sounds like a great idea. (same could work for mute list)
Nofish commented on Happyrabbit's post: What OS are you using? On windows it should automatically download the latest one from tor's webpage
Nofish commented on Uv's post: I'm not really following it, but everyone should use the network that is most suited to his/her needs or he/she likes the best. There is already many alternatives: bittorrent/ipfs/maidsafe/i2p-lafs/etc.
Nofish commented on Uv's post: Stop Zeronet, Copy your old users.json to your data dir, Start ZeroNet, visit ZeroID site should work.
Nofish commented on Nofish's post: @Sick: Yes, every concurrency should be possible
Nofish liked Gonzo's post
Nofish commented on Nofish's post: @Sick Kuerana: Yeah, adding support for other address formats would be easy if it based on secp256k1 cryptography
Nofish commented on Nofish's post: @Zeroidea: Not really just "not the right time for it", but were positive, so maybe next year. To be honest I also think my topic would not fit to the session (Our robotic overlords) they arranged me and I also need more time to get used to the audience.
Nofish commented on Nofish's post: @Sick telephone: I took it from [Morton park](https://www.google.hu/maps/@49.2877314,-123.1416204,3a,75y,131.54h,86.12t/data=!3m6!1e1!3m4!1sywPCZC9_QoBbwimxsH2xTQ!2e0!7i13312!8i6656)
Nofish commented on Nofish's post: @John Doe: Yeah, but especially since I had "Speaker" on my conference badge. They all found it interesting after I described them and I also had positive feedback from the organizers, so maybe next year.
Nofish commented on Nofish's post: @Ivesen: I was a guest at TED this year, but there is a possibility to give a talk about it next year. Yeah we need attention and I happy to answer any questions. :)
Nofish commented on P2P's post: It could work, parts that needs to be modified: - Saving/loading html pages - Image uploading
Nofish commented on Jayofthezero's post: Rev2054 added `--bind 1.2.3.4` optional. I was unable to test it as I don't have VPN connection, so please give some feedback if it worked for you.
Nofish started following Krixano
Nofish commented on Krixano's post: It should work on any file (for example ZeroBlog posts stored in data/data.json) I think the problem is the missing `["json_id", "INTEGER REFERENCES json (json_id)"]` in your zites table.
Nofish commented on Dude's post: The favorites are stored and handled by the ZeroHello site, so it's not possible.
Nofish commented on Krixano's post: you can try `body { backface-visibility: hidden }`
Nofish commented on Firebox's post: Thanks, true, it's unnecessary. Removed in Rev2048
Nofish commented on Krixano's post: modify dbschema.json, and increment table's `schema_changed` value, press reload on sidebar
Nofish commented on Jayofthezero's post: Yeah, you are right, it's only modified the listening socket's network address. I will check the possibility on binding outgoing connections.
Nofish commented on Krixano's post: @Krixano: I recommend starting with the ZeroChat tutorial @ ZeroBlog. It stores/queries the messages using database.
Nofish commented on Krixano's post: Probably your json file is invalid (at line 8, char 4). try pasting it to http://jsonlint.com/ to see detailed info.
Nofish commented on Jayofthezero's post: `--fileserver_ip 1.2.3.4 --ui_ip 1.2.3.4` should work
Nofish commented on P2P's post: @P2P: Even if we sign every resources, the dlls loaded before the python interpreter, so if anyone want to add any malicous code we can't do anything against it. btw why would you use zeronet.exe if you can use simply python.exe? https://www.virustotal.com/en/file/8006bcf09a7b148b7ef87f2fa4d106b51920eb6b218e9e92b3b549f7c924c44d/analysis/
Nofish commented on P2P's post: Yeah it can be problematic, and I will try to find a better way (problematic parts: dlls, pyd files, updates, third party plugins) but if the virus scanners marks zeronet.exe malicious because of that, then they should also mark every application that loads external dlls, because windows does not checks signature on that files. Win/Mac dists are only updated if I find the new version stable. MacOS much better in this case, as it validates every file within the .app by itself.
Nofish commented on Binchan's post: @Binchan: It's a simple misspelling: you need `date_added`, but you have dat**a**_added in the file
Nofish commented on Binchan's post: If the json file is valid and same as other user's then it should update the date_added correctly. Can you give link to your bot's data.json file?
Nofish commented on ZAlex's post: sad :( unfortunately this can be a precedent for the future. plz keep us updated if you have more info
Nofish commented on P2P's post: Thanks, fixed! I switched to Tor always mode and the release script was not executed correctly. It should work now and I think I also found and fixed the site lock bug.
Nofish commented on Glightstar's post: Yes, currently there is no global way, you need to add it to the content.json file you will use to sign the files.
Nofish commented on Glightstar's post: You need to add an "optional" regex pattern to the content.json and zeronet will handles the rest. eg.: `"optional": "(?!avatar).*jpg",` [in zerome's user files](merged-ZeroMe/1RedkCkVaXuVXrqCMpoXQS29bwaqsuFdL/data/users/1J3rJ8ecnwH2EPYa6MrgZttBNc61ACFiCj/content.json) means every jpg file in user directory is optional file except avatar.jpg
Nofish commented on Mkg20001's post: Yeah thanks, it's a pretty big patch, I going to check it
Nofish commented on Firebox's post: @P2P: Every site you seed has a assigned .onion address. site lock violation happens if someone try to request a file from a site that is not assigned to that .onion address.
Nofish commented on Firebox's post: @P2P: You can try to set --tor_hs_limit to lower value. Maybe the guard node requires Proof-of-work based on load to avoid DoS. I got no site lock violation yet, but I will keep running it.
Nofish commented on Firebox's post: @P2P: Then probably the high cpu caused by different reasons, you can try to update your tor client, for me the cpu usage hovers around 1-2%
Nofish commented on Firebox's post: Pull request merged + Tor hidden service merging added in Rev2033. From now by default ZeroNet creates maximum 10 hidden services, then choose randomly between them. (Can be changed with --tor_hs_limit argument) @P2P: Please verify if it's fixed the CPU usage problem
Nofish commented on P2P: @Lmath's post: @P2P: Yes, python also requires it
Nofish commented on P2P: @Lmath's post: @P2P: According to readme: Built with: Microsoft Visual C++ 2008 Express Edition, Supported OS: Windows 2000 up to Windows 10
Nofish commented on P2P: @Lmath's post: @Lmath: Thanks, that's normal, changed to this version.
Nofish commented on P2P: @Lmath's post: Can you please try if this one works for you: https://indy.fulgan.com/SSL/openssl-1.0.2k-i386-win32.zip ? Just unpack and run openssl.exe. If no error, then I will change to this.
Nofish commented on Glightstar's post: @Glightstar: Try press Reload the Rebuild button on the sidebar.
Nofish commented on Glightstar's post: Thanks, updated the tutorial on medium.com. As for the error: looks like for some unknown reasons there is an inconsistency between the "content.db" cache and the real file system. I testing a fix for it, but try to create the "data/users/14K7EydgyeP84L1NKaAHBZTPQCev8BbqCy/content.json" file manually with the content "{}"
Nofish commented on Firebox's post: @Jayofthezero: Yeah, it's a good idea and probably that going to be the next major feature of zeronet: support for zipped archive of old content as optional files. It will also speed up the initial sync time, reduce hdd usage and in general solves many scalability problems. Btw I created a new hub: White hub :P
Nofish commented on Firebox's post: I think I'm done: all the libs are up-to-date. Some of them required some patching: - pyelliptic: loadOpenSSL function to find bundled libs - PySocks: add win_inet_pton lib - pybitcointools: relative import path in bitcoin module I will create a update.sh script that handles the downloads (should it use http or pip?) and updates automatically.
Nofish commented on Firebox's post: Maybe we could use this: https://requires.io/github/openhealthcare/opal/requirements/?branch=v0.8.2
Nofish commented on Firebox's post: Just updated rsa, PySocks, pybitcointools, pyasn to latest version Upcoming: gevent-websocket, openssl dll, pyelliptic, coffeescript Probably I will remove BitcoinECC and by that the ZeroBoard site
Nofish commented on Firebox's post: Thanks for the list, it's really helpful. I will check later today if newer libs working well.
Nofish liked ks's post
Nofish commented on Nofish's post: @Ranxerox: Probably it's not, but the TED motto is "Ideas worth spreading" and "Future of you" I think it's already more than just an idea, since it's working. Thanks for the heads-up to be honest I lost a bit on opened issues in the last few months, I will check it and update and outdated libs.
Nofish commented on Nofish's post: @Monday: Unfortunately I feel the real problem is lots of people fear of uncensored network because "Darknet = Drugs, CP, Antisemitism = Bad" and they don't want to take risks of these negative feedback.
Nofish commented on Mkg20001's post: Looking nice, I will check the pull request!
Nofish commented on Plasmmer's post: @Plasmmer: When you visit the proxy the private key displayed in the browser and later the user can use to to login
Nofish commented on Plasmmer's post: @Plasmmer: thx, modified
Nofish commented on kskarthik's post: @Plasmmer: Planned in the future, but currently focusing on features that affect most of the users.
Nofish commented on Plasmmer's post: @Plasmmer: Yes, without installing anything, all you need is a standard browser
Nofish commented on Plasmmer's post: You can already do that using proxies: https://bit.surf:43110/me.zeronetwork.bit/?Post/1oranGeS2xsKZ4jVsu9SVttzgkYXu4k9v/15XeFY67fYKCKxQPfeo246ELto2Gff1hAe/1491080744
Nofish commented on P2P's post: I have pushed a new update that should prevent running save funcion in parallel: https://github.com/HelloZeroNet/ZeroNet/commit/e3baf8944e18587f736af9c83393bfe2f59ec967
Nofish commented on P2P's post: @P2P: I dont think so, need a way to reproduce the problem to be able to fix it
Nofish commented on P2P's post: @P2P: ZeroNet is single threaded so it should not be the problem.
Nofish commented on P2P's post: @P2P: Having file opened for reading does not blocks the file write, it would be only problematic if we open it for writing multiple times.
Nofish commented on P2P's post: The users.json file always using the atomicWrite method that is opening file using `with` and should prevent data loss: https://github.com/HelloZeroNet/ZeroNet/blob/master/src/util/helper.py#L15 So the problem should be somewhere else. Are the files are totally corrupted/lost or just missing something from it?
Nofish commented on Plasmmer's post: these informations not timestamped/logged yet, but could be possible in the future
Show more...

Following

+
paragliding, software, bitcoin [my blog](/ulrichard.bit) [my gallery](/1EHrpXd3f6CyKooBFtpLhdFk5C3bqFhyg1) [tandem flights](/paraeasy.bit) [south america trip 2007](/1MrcCZjfRCKeJZAENsjDqf7dvDhEhFpsXY) [Why I prefer Bitcoin](/13UYenBm13vidxEA9DxKMSA3CJacWdkpke) [Why I prefer Linux](/1Aiq4P8bh1f9pPxQz9MA2Z1cDkYaUGjEPn/) [Tesla for Bitcoin](/1P9yJty4fzXkuGV3BSM4VSADr7xap2riAa/) [Bought with Bitcoin](/17JvYecccPVS8nzngiUKwJQe8LX5jrjNwo/)
+
I love 3 things: Computer science, philosophy and porn. Visit my hentai site (only in italian) ! http://127.0.0.1:43110/15xL5zA9TeDxdqwnikSA1SfhFEzMfBxD5D/index.html
+
ᖱᗩ ᕈᙓᙓᑕᙓ make Love not War. Protect the Internet, the largest project of mankind on Earth. Don't ask for your privacy. Take it back.
+
Author of the [Bluishcoder](https://bluishcoder.co.nz) clearnet blog [mirrored on ZeroNet](http://127.0.0.1:43110/1BLuish1ZbwVfPaCSDVQkg5WW58hqEoYQw)
+
## Programmer, Poet, Conlanger, Composer ### Projects * [ZeroNet Exe Installer](/Talk.ZeroNetwork.bit/?Topic:1526846299_12gAes6NzDS9E2q6Q1UXrpUdbPS6nvuBPu/ZeroNet+Exe+Installer+Beta+P2P+Message+Plugin+by+GitCenter) * [KxoNetwork](/kxonet.bit) * [KxoVid](/kxovid.bit) * [New Important Zites](/1MiS3ud9JogSQpd1QVmM6ETHRmk5RgJn6E/) * [ZeroMedium](/ZeroMedium.bit/?/) ([GitCenter](/1BEPbMfV8QtaZCD2cPNbabfDKnmhTAZRPx)) * [ZeroExchange](/1PHBjZSAc6mHDMkySJNs3XeSXUL7eY7Q7W) * [ZeroNet Dev Center](/14pM9huTYzJdyQyHRj6v2kfhMe8DrxwpGt) * [ZeroDB](/ZeroDB.bit) * [ZeroFrame Router Example](/1K2myjtjoEVpRC2JMieRL73ES4V4iLP2Ev) * [Important Zites](/15Pf9VVuDT8NSWj1qUBh4V89yPmrmzRw6a/) (Old) * [My Portal Zite](/1JBXrjCabLEWXmKJ2pJ4XhxA4rwYAEazKw)
+
Chech out my blog etc. [my blog](/1MqGCssnzh6kkBTAR49cWYNAdt4zhZ2ezC)
+
Not so random ZeroNet user
+
Feel free to email me if you have any questions. GPG: `0AD8 D7E9 DCD9 5A10 5BA6 A111 6CC6 5FB9 1CE3 9BB5`.
+
Random ZeroNet user
+
a vigorously lazy deadbeat with matured immaturity **Zites** [SetupList](/setuplist.0web.bit) **Clearnet** [mmap.page](https://mmap.page) **License** Permission to use, copy, modify, and/or distribute all my posts and comments at ZeroMe for any purpose with or without fee is hereby granted.
+
Interested in security, privacy and anonymity. [Personal Zite](/amorgan.bit) Tox ID: A159F3328E43C0766FE3CECA051049699E6AA4B3037FC84B643A00E975DA3559F6D9887D5A17
+
Nobody can stop me from dreaming [✉](/itskas.bit/)
+
delapsus resurgam
+
Author of ZeroNet's first decentralised music streaming service called "NinjaBeats". - [NinjaBeats website (NEW!)](http://127.0.0.1:43110/15F9LYs6c17NUXv3MxpJSGbpWxyHTEbzMk/) - [NinjaBeats streamer (NEW! - Now works on mobile)](http://127.0.0.1:43110/15F9LYs6c17NUXv3MxpJSGbpWxyHTEbzMk/stream/audio.html) - [Contact me](http://127.0.0.1:43110/Mail.ZeroNetwork.bit/?to=paintninja) - [My blog](http://127.0.0.1:43110/1BufKyfSLxGcWFkP6T3t4HJYXzhKA1DVrM)
+
Share and enjoy! | [books](http://127.0.0.1:43110/192dZ1EG5tU7PnCfuwGMDEBrr2eLqvs4t3/?Category:books) | [chat](http://127.0.0.1:43110/zeromedia.bit/) | [sandbox](http://127.0.0.1:43110/Me.ZeroNetwork.bit/?Post/1BLueGvui1GdbtsjcKqCf4F67uKfritG49/1GervYZoP82e8X1Rgc4wyMkJY2zSsJuq8t/1472286311) |:---:|:---:|:---:| | [GTM 111](http://127.0.0.1:43110/111GTMxRyZBvXVZgJZrkjpPLVZDFnEU8h/) | | [land](http://127.0.0.1:43110/1Jo59uY2VygWUHjWLfkg71fCgNzX4vmZMz/) | | [CMS](http://127.0.0.1:43110/1KuwNEQXaF7SmAibxLs2wgJQJjGYTDMErc/) | | | | [ACiC](http://127.0.0.1:43110/1EmfJvSsZWtikUqWnHbhMQ9yV5ATUBEooc/) | | | | [CIG](http://127.0.0.1:43110/1PDFcigDTymiqMkiK5iXLgtXxDwT5ts3WC/) | | | | [ABHoH](http://127.0.0.1:43110/1ABHoH2K2PBKv3bBDqfyLeNxBDT2XiH63A/) | | | | [☉](http://127.0.0.1:43110/1PDFoo23w9Ha71HcmFsL4zBRTKpKKGjMuv/) | | [__amine__](http://127.0.0.1:43110/192dZ1EG5tU7PnCfuwGMDEBrr2eLqvs4t3/?Category:amine) |
+
Liberté. Egalité. Poulet. 🌈 I don't defend my ideas. They are not mine. Please, share your music : [mixtape.bit](/mixtape.bit)
+
Be free [ZeroMe "Plus"](http://127.0.0.1:43110/1Lj1oPcN7oZQL8HkS5KbkzQuKqs42zQWY6/) - Clickable & searchable hashtags - Reverse comment order option - See a random post [ZeroNet Mods](http://127.0.0.1:43110/1JjHyLeEDidHyyN33Y8f8cQKVEzWCEPnW7/) - Simple API - User switcher - Default User
+
User from the Void. -------- - [Dark ZeroMe](http://127.0.0.1:43110/1HmJ7GadjdiQUPevaK7QXVrK4nhyCp1fSj/) - If you hate white, come join us on the dark side. - [8values on Zeronet](http://127.0.0.1:43110/1GhQnhzopJKP1RUKX8cd55CSXXSygzPFYj/) - Find out how you lean in politics. - [Blog](http://127.0.0.1:43110/19Bb6RMYZGTDZJduj5q5Zz8uaBgNtHnCeH/) - So far used mostly to post updates of my various ZeroNet projects. - [TED Talks 2017 - Speakers List Archive](http://127.0.0.1:43110/14MQWAD4DNSWx52pidtux4oQXz4gTn4B5f/) - Archive of speakers from when the ZeroNet founder was slated to appear at the conference. - [Share Test](http://127.0.0.1:43110/1FyYahC8sKVYfsPQXEzuRjXNDQeb6Y2xno/) - Experiments with file sharing and video streaming using ZeroMux. - [Contact Me](http://127.0.0.1:43110/mail.zeroverse.bit/?to=jayofthezero)
+
26/Female - Kaffiene Developer --- [Patreon](https://www.patreon.com/aprilc0t) [Patreon Discord](https://discord.gg/hx3xXZV) [My Blog](http://127.0.0.1:43110/1A83ijw3boqTtqdLz8me7AqeK1nEK8yxeu) [My Github](https://github.com/April93) [Kaffiene Search](http://127.0.0.1:43110/1Mr5rX9TauvaGReB4RjCaE6D37FJQaY5Ba)
+
Belarusian Python and JS Developer [Blog](http://127.0.0.1:43110/blog.gomzik.bit) (RU) Working on ZeroMusic service [here](http://127.0.0.1:43110/1JMNrd9FD19AhWVukYJ2gdxAiK2ohczwk5/)
+
Privacy ain't dead, deal with it!
+
Random ZeroNet user
+
Let's decentralize the web!
+
The Evil One
+
Random ZeroNet user
+
I am from the future. SkyNet has sent me on a mission to destroy mankind.
+
Random ZeroNet user
+
Random ZeroNet user!
This page is a snapshot of ZeroNet. Start your own ZeroNet for complete experience. Learn More