CODEWITHSUNDEEP
gitgit-pull
Martin AJ
Martin AJ

Reputation: 6697

How can I git pull on the server optimally?

It's two days I'm trying to pull my project (which is on git) on my server. My server has 1GB ram and uses Ubuntu OS. (usually 200mb of RAM is used by other processes).

When I run git pull origin master on the server, It throws this:

fatal: Out of memory, malloc failed (tried to allocate 533517295 bytes)

fatal: index-pack failed

enter image description here

Also here is the result of nano .git/config:

[core]
    repositoryformatversion = 0
    filemode = true
    bare = false
    logallrefupdates = true
    packedGitLimit = 512m
    packedGitWindowSize = 512m

[remote "origin"]
    url = https://***@bitbucket.org/***/***.g$
    fetch = +refs/heads/*:refs/remotes/origin/*

[pack]
  threads = 1
  deltaCacheSize = 512m
  packSizeLimit = 512m
  windowMemory = 512m

Honestly I cannot get more RAM for my server. Any idea how can I pull my project on the server?

Noted that in reality there are real names instead of ***.

Upvotes: 0

Views: 313

Answers (1)

Ferrybig
Ferrybig

Reputation: 18834

To limit the amount of memory used during a fetch operation (git pull uses git fetch), you can use the --depth flag.

Just run git pull --depth=10 and increase to number to fetch more and more history (or decrease it if you also run out of memory). Once you have enough objects loaded in history, you can request the remainder object using git fetch --unshallow

The downside of pulling with depth is the fact that your repository will take more space on disk, as it also has to account for the multiple states of the repo

Upvotes: 1

Related Questions