Search Results

Search found 5413 results on 217 pages for 'git pull'.

Page 8/217 | < Previous Page | 4 5 6 7 8 9 10 11 12 13 14 15  | Next Page >

  • Gittornado with Nginx fails to push and pull

    - by Josh Buell
    I'm making a simple website to host git repositories, much like github. I'm using Gittornado to handle git Smart HTTP requests, and it works perfectly locally; I can clone, push, pull, etc... But when I put it behind Nginx, git commands stop working, giving no errors except: "fatal: The remote end hung up unexpectedly" I know that it's Nginx that's causing the trouble because if I open the port that tornado is running on and try my git commands through that (i.e. "git pull \http://mysite.com:8000/myrepository master" instead of "git pull \http://mysite.com/myrepository master" [backslashes added because Server Fault says I have too many links]) everything works as expected. The Nginx access and error logs don't seem to say anything interesting, so I'm reasonably sure that it has something to do with the way Nginx is compressing or chunking the requests/responses, causing git to think there's been an unexpected hangup, but I'm not sure what to do to fix it, since this is my first time with Nginx. My Nginx configuration file is basically a clone of the on found here; I've tried commenting out various likely-seeming options to see if they were causing the problem, but none of them fixed it so I assume there's some default behavior I need to suppress, I'm just not sure which. Any thoughts on how to fix this? Since it works not through Nginx, I'm considering just redirecting git requests to the tornado port itself, but this feels like a hack rather than a clean solution...

    Read the article

  • Where to find Hg/Git technical support?

    - by Rook
    Posting this as a kind of a favour for a former coleague, so I don't know the exact circumstances, but I'll try to provide as much info as I can ... A friend from my old place of employment (maritime research institute; half government/commercial funding) has asked me if I could find out who provides technical support (commercial) for two major DVCS's of today - Git and Mercurial. They have been using VCS for years now (Subversion while I was there, don't know what they're using now - probably the same), and now they're renewing their software licences (they have to give a plan some time in advance for everything ... then it goes "through the system") and although they will be keeping Subversion as well, they would like to justify beginning of DVCS as an alternative system (most people root for Mercurial since it seems simpler; mostly engineers and physicians there who are not that interested in checking Git repos for corruption and the finer workings of Git, but I believe any one of the two could "pass") - but it has to have a price (can be zero; no problem there) and some sort of official technical support. It is a pro forma matter, but it has to be specified. Most of the people there are using one of the two already, but this has to be specified to be official. So, I'm asking you - do you know where could one go for Git or Mercurial technical support (can be commercial)? Technical forums and the like are out of the question. It has to work on the principle: - I have a problem. - I post a question with the details. - I get an answer in specified time. It can be "we cannot do that." but it has to be an official answer and given in agreed time. I'm sure by now most of you understand what I'm asking, but if not - post a comment or similar. Also, if you think of any reasons which could decide justification of introducing Git/Hg from an technical and administrative viewpoint, feel free to write them down also.

    Read the article

  • Why did Git become so popular?

    - by Jungle Hunter
    Almost every article you read comparing Git and Mercurial it seems like Mercurial has a better command line UX with each command being limited to one idea only (unlike say git checkout). But at some point Git suddenly became super popular and literally exploded. Source: Debian What happened in 2010-01 that things suddenly changed. Looks like GitHub was founded earlier than that - 2008. Edit: Git 1.7.0 seems to be released at the same time: January 2012. Here are the 1.7.0 release notes and the file history with the corresponding dates.

    Read the article

  • svn vs git for the sole developer? [closed]

    - by nattyP
    If I am sole developer (I do not work in a team) working from my laptop (Windows OS and Linux VM) and backing up data to the cloud (Dropbox etc), then is git still better than svn for my version control needs? I was thinking not since I wont need any of git's distributed features. But is git such a better approach to version control that I should consider moving anyway? With so many articles saying how people are moving from svn to git? I was wondering, if they are talking about large or open projects with teams of developers vs the sole developer. What do you think?

    Read the article

  • Parse git log by modified files

    - by MrUser
    I have been told to make git messages for each modified file all one line so I can use grep to find all changes to that file. For instance: $git commit -a modified: path/to/file.cpp/.h - 1) change1 , 2) change2, etc...... $git log | grep path/to/file.cpp/.h modified: path/to/file.cpp/.h - 1) change1 , 2) change2, etc...... modified: path/to/file.cpp/.h - 1) change1 , 2) change2, etc...... modified: path/to/file.cpp/.h - 1) change1 , 2) change2, etc...... That's great, but then the actual line is harder to read because it either runs off the screen or wraps and wraps and wraps. If I want to make messages like this: $git commit -a modified: path/to/file.cpp/.h 1) change1 2) change2 etc...... is there a good way to then use grep or cut or some other tool to get a readout like $git log | grep path/to/file.cpp/.h modified: path/to/file.cpp/.h 1) change1 2) change2 etc...... modified: path/to/file.cpp/.h 1) change1 2) change2 etc...... modified: path/to/file.cpp/.h 1) change1 2) change2 etc......

    Read the article

  • Options for secure git -repo hosting?

    - by hhh
    I need a secure git -repo host, either by third-party or by myself. I am not sure how so outlining some ideas. Please, answer how you manage git -repos securely -- do you use some service or do you use only your 'legs' -approach? Afaik Bitbucket.org and Github.com are missing Gmail -style second-verification. Now I need this kind of login-system with password and mobile-phone to access the administration things in the git -hosting or ability to disable this kind of access without private -key. Host it oneself (not sure about details) other? Perhaps related http://stackoverflow.com/questions/11007679/how-can-i-host-git-repositories-and-manage-my-content-hosting-myself

    Read the article

  • What does SVN do better than git?

    - by doug
    No question that the majority of debates over programmer tools distill to either personal choice (by the user) or design emphasis, i.e., optimizing design according to particular uses cases (by the tool builder). Text Editors are probably the most prominent example--a coder who works on a Windows at work and codes in Haskell on the Mac at home, values cross-platform and compiler integration and so chooses Emacs over Textmate, etc. It's less common that a newly introduced technology is genuinely, demonstrably superior to the extant options. I wonder if this is in fact the case with version-control systems, in particular, centralized VCS (CVS, SVN) versus distributed VCS (git, hg)? I used SVN for about five years, and SVN is currently used where I work. A little less than three years ago, I switched to git (and gitHub) for all of my personal projects. I can think of a number of advantages of git over subversion (and which for the most part abstract to advantages of distributed over centralized VCS), but I cannot think of one contra example--some task (that's relevant and arises in a programmers usual workflow) that subversion does better than git. The only conclusion I have drawn from this is that I don't have any data--not that git is better, etc. My guess is that such counter-examples exist, hence this question.

    Read the article

  • Git, auto updating, security and tampering?

    - by acidzombie24
    I was thinking about hosting my private project on my server (i may use 'gitolite') and have a copy on my local machine as backup (git clone then automated git fetch every few minute). I want to know what happens if there is a bug gitolite or somewhere else on my server and the source code and git repository has been tampered with? Will my backup also be corrupted? will i easily be able to revert the source using the history?

    Read the article

  • Structure of a Git repository

    - by Luke Puplett
    Sorry if this is a duplicate, I looked. We're moving to Git. In Subversion, I'm used to having \trunk, \branches and \tags folders. With Git, switching between branches will replace the contents of the working directory, so am I right to assume that the way we used to work just doesn't apply with Git? My guess is that I'd have a repo folder with maybe a gitignore and readme.txt, then the folders for the projects that make up the repo, and that's it.

    Read the article

  • Git, auto updating, security and tampering?

    - by acidzombie24
    I was thinking about hosting my private project on my server (i may use 'gitolite') and have a copy on my local machine as backup (git clone then automated git fetch every few minute). I want to know what happens if there is a bug gitolite or somewhere else on my server and the source code and git repository has been tampered with? Will my backup also be corrupted? will i easily be able to revert the source using the history?

    Read the article

  • Git cloning for Ubuntu Kernel gave error: index-pack died of signal 9447381

    - by LAMOHAN
    My /usr/src is found empty. So I tried to install a fresh Kernel. But was unsuccessful with some error. I did this: git clone git://kernel.ubuntu.com/ubuntu/ubuntu-precise.git but it gave this error message: error: index-pack died of signal 9447381), 802.20 MiB | 88 KiB/s fatal: index-pack failed My current Kernel version is 3.8.13-bone20 #1 in LINUX -Ubuntu-armhf Can anyone help me to solve this?

    Read the article

  • Build git on embedded system

    - by naive231
    I want to build git on our embedded system and I got git's source code from here. I run ./configure tool with following arguments: ./configure --build=arm-eabi --host=`uname -p`-`uname -s`-`uname -o` Configure tool runs, and returns an error says: checking whether system succeeds to read fopen'ed directory...configure: error: in`/home/git-1.8.0.2`. I set necessary environment variables like CC and CXX to our tool chain already, so ./configure complains nothing about compilers. But I have no idea about that error message. It looks like some permission problem, but I have full permission of /home/git-1.8.0.2 for sure. Any idea ?

    Read the article

  • How to import a svn repository underneath a git repository?

    - by Thiago Moreira
    Hi there, I have a svn repository that I migrated to git using the tool svn2git. Now I would like to push this git layout to a remote repository underneath an existing directory. But, I would like to keep the svn history (tags and branches). For instance: Git remote repository layout: git-repository/dirA git-repository/dirB git-repository/dirC/svn-repository-migrated-to-git Makes sense? Is it possible?? Thanks

    Read the article

  • Using git-svn with slightly strange svn layout

    - by Ibrahim
    Hi guys, I'm doing an internship and they are using SVN (although there has been some discussion of moving to hg or git but that's not in the immediate future). I like git so I would like to use git-svn to interact with the svn repository and be able to do local commits and branches and stuff like that (rebasing before committing to svn of course). However, there is one slight wrinkle, the svn repository layout is a little weird. It basically looks like this /FOO +-branches +-tags +-trunk +-FOO +-myproject Basically, my project has been stuck into a subdirectory of trunk, and there is another project that is also a subdirectory of the trunk. If I use git-svn and only clone the directory for my project instead of the root, will it get confused or cause any problems? I just wonder because the commit numbers are incremented for the entire repository and not just my project, so would commits be off or anything like that? I probably wouldn't push any branches or tags to SVN because I'd prefer to just do those locally in git and I don't know how git-svn deals with branches and tags anyway, and no one else uses them so I find little point in doing so. Thanks for the help!

    Read the article

  • Git force complete sync to master

    - by Jesse
    My workplace uses Subversion for source control so I have been playing around with git-svn for the advantages of my own branches, commit as often as I want without touching the main repo, etc. Since my git svn checkout is local, I have cloned it to a network share as well to act as a backup. My thinking is that if my desktop takes a dump I will at least have the repo on the network share to get changes that I have not had a chance to dcommit yet. My workflow is to work from the desktop, make changes, commit, etc. At the end of the day I want to update the repo on the network share with all of my current changes. I had setup the repo on the network share using git clone repo_on_my_desktop and then updating the repo on the network share with git pull origin master. The problem that I am running into is when I used do a git rebase to squish multiple commits prior to dcommitting to the main svn repository. When I do this, I get merge conflicts on the repo on the network share when I try to backup at night. Is there a way to simply sync entirely with the repository on my desktop without doing a new git clone each night?

    Read the article

  • git: correct way to merge/rebase with respect to svn dcommit

    - by Albert
    I have the following situation (mostly because I didn't really thought it through in the beginning -- or more exactly, I thought it shouldn't be a problem the way I did this but now I am stumbled): ... --- A --- B1 --- ... --- Bn ... --- git-svn Whereby A and git-svn are at the same state (exactly the same files and file content) but they don't have any common point in history. And I want: ... --- git-svn --- B1 --- ... --- Bn Or at least, when I do the svn dcommit, I want exactly to get the commits B1 to Bn and nothing else. I am not exactly sure how dcommit works. So if I would get something like this: ... ------------ A --- B1 --- ... --- Bn \ \ ... --- git-svn -- A' ----------------- B' would the dcommit behave in the way I want? Because if so, that would be easy to get (merging A into git-svn does work just fine because they are content-wise the same). Or should I do some sort of rebase? But I don't want to rebase A on git-svn, just B1 to Bn.

    Read the article

  • git push with git-cola failing

    - by slacktracer
    I started getting the following error a week ago when pushing with git-cola...I found something about a similar problem happening a lot a couple years ago but it didn't help at all. "git push" returned exit status 128 Have you rebased/pulled lately? Already up-to-date. Pushing to https://github.com/slacktracer/lokapala.git error: cannot run None: No such file or directory fatal: could not read Username for 'https://github.com': No such device or address When I push with the terminal it works just fine, so perhaps it is mostly a question about git-cola. Anyway, just wondering if anyone can help. I'm lost right now...

    Read the article

  • Git - git add in a gitignore directory

    - by Steve
    If my .gitignore file has tmp\ in it and I do a git add file.test from tmp, Git adds file.test to the repository. If file.test never changes, than this is as good as a one time add to the repository, right? Say for static files that I want to be in version control. You want the file in an initial clone and that's it. I assume file.test doesn't get tracked, so if there are updates, Git doesn't see it as modified?

    Read the article

  • Git push won't do anything (Everything up-to-date)

    - by phleet
    I'm trying to update a git repository on github. I made a bunch of changes, added them, committed then attempted to do a git push. The response tells me that everything is up to date, but clearly it's not. git remote show origin responds with the repository I'd expect. Why is git telling me the repository is up to date when there are local commits that aren't visible on the repository? [searchgraph] git status # On branch develop # Untracked files: # (use "git add <file>..." to include in what will be committed) # # Capfile # config/deploy.rb nothing added to commit but untracked files present (use "git add" to track) [searchgraph] git add . [searchgraph] git status # On branch develop # Changes to be committed: # (use "git reset HEAD <file>..." to unstage) # # new file: Capfile # new file: config/deploy.rb # [searchgraph] git commit -m "Added Capistrano deployment" [develop 12e8af7] Added Capistrano deployment 2 files changed, 26 insertions(+), 0 deletions(-) create mode 100644 Capfile create mode 100644 config/deploy.rb [searchgraph] git push Everything up-to-date [searchgraph] git status # On branch develop nothing to commit (working directory clean)

    Read the article

  • Git init - .git: Permission Denied

    - by Gcoop
    Hi All, I am trying to initiate git on my remote server using ssh. When I run git init On the server in a folder I have write permissions to I get the following error. .git: Permission denied Do I need to assign any other permissions on that folder to be able to create the repository? Thanks

    Read the article

  • Git Pull works; Git push fails

    - by Michael
    I thought I set up my key pairs correctly -- I can do git pulls. I can do git commits. But when I do a git push, it counts objects, decompresses, then says: fatal: the remote end hung up unexpectedly. What's the issue here? I'm a super user, so it's not folder writable / readable access problems -- it must be the way I set up the encryption key pair... how do I debug this ... since git pull works?

    Read the article

  • Executing python subprocess via git hook

    - by aljesco
    I'm running Gitolite over the Git repository and I have post-receive hook there written in Python. I need to execute "git" command at git repository directory. There are few lines of code: proc = subprocess.Popen(['git', 'log', '-n1'], cwd='/home/git/repos/testing.git' stdout=subprocess.PIPE, stderr=subprocess.PIPE) proc.communicate() After I make new commit and push to repository, scripts executes and says fatal: Not a git repository: '.' If I run proc = subprocess.Popen(['pwd'], cwd='/home/git/repos/testing.git' stdout=subprocess.PIPE, stderr=subprocess.PIPE) it says, as expected, correct path to git repository (/home/git/repos/testing.git) If I run this script manually from bash, it works correct and show correct output of "git log". What I'm doing wrong?

    Read the article

< Previous Page | 4 5 6 7 8 9 10 11 12 13 14 15  | Next Page >