slc deploy always giving RPC failed; result=55, HTTP code = 0

[idlescan]

Hi everyone, I'm getting this error trying to deploy to a remote box. The box has SLC PM installed as a service and I've verified that it's running.

Can anyone give some direction on this?

slc deploy -s testdeploy http://mybox Counting objects: 6707, done. Delta compression using up to 4 threads. Compressing objects: 100% (4945/4945), done. Writing objects: 100% (6707/6707), 10.28 MiB | 0 bytes/s, done. Total 6707 (delta 1297), reused 6702 (delta 1295) error: RPC failed; result=55, HTTP code = 0 fatal: The remote end hung up unexpectedly fatal: The remote end hung up unexpectedly Everything up-to-date Failed to deploy 'deploy' as 'testdeploy' to 'http://mybox:8701/' via 'http://mybox:8701/'

I've tried increasing git buffer size, this has no effect. git config http.postBuffer 524288000

I've successfully deployed to PM on my local machine, but cannot get this to deploy to the remote environment.

I've also run it with the following parameters, output below:

GIT_CURL_VERBOSE=1 GIT_TRACE=1 slc deploy -s testdeploy http://mybox

POST /api/services/2/deploy/default/git-receive-pack HTTP/1.1 Host: mybox:8701 User-Agent: git/2.3.2 (Apple Git-55) Accept-Encoding: gzip Content-Type: application/x-git-receive-pack-request Accept: application/x-git-receive-pack-result Content-Length: 10780126

Recv failure: Connection reset by peer Closing connection 1 error: RPC failed; result=56, HTTP code = 0 fatal: The remote end hung up unexpectedly fatal: The remote end hung up unexpectedly Everything up-to-date Failed to deploy 'deploy' as 'testdeploy' to 'http://mybox:8701/' via 'http://mybox:8701/'

Thanks!

[sam-github]

@rmg have you seen this? @idlescan I wonder if using ssh tunneling would allow you basically "be" local on the remote system and bypass this?

see https://github.com/strongloop/strong-deploy/blob/master/bin/sl-deploy.txt#L28-L36

[rmg]

@idlescan what version of node and strong-pm are running on the remove system you are trying to deploy to?

[idlescan]

@rmg

slc -v strongloop v6.0.1 (node v4.2.6) ├── strong-arc@1.8.8 (fb756f2) ├── strong-build@2.1.2 (a98f310) ├── strong-deploy@3.1.4 (9b29df0) ├── strong-mesh-models@8.1.0 (62e539b) ├── strong-pm@5.2.1 (6ab604c) ├── strong-registry@1.1.6 (f8c2339) ├── strong-start@1.3.4 (28c261d) ├─┬ strong-supervisor@3.3.3 │ └── strong-agent@2.0.3 (ec2d95b) ├── generator-loopback@1.20.6 (c53788b) ├── node-inspector@0.7.4 └── nodefly-register@0.3.3

node -v v4.2.6

It's running on latest Ubuntu on DigitalOcean. I previously had a bunch of issues getting strongloop installed, had to do --unsafe-perms on any npm installation.

@sam-github I can successfully deploy to the host if I do it from the host itself - check out the repo onto the box, and do slc deploy from there. I've tried the http+ssh://REMOTEHOST deployment with environment variables set, but get the following: ssh connection error: { [Error: All configured authentication methods failed] level: 'client-authentication' }

[idlescan]

Hi everyone,

Any advice on how to approach this issue?

I'm currently SSH-ing onto the remote server, checking out the repository and deploying locally, due to the above issue.

Is that a safe way to deploy my application in a production environment? Any downsides to this?

Thanks!

[idlescan]

@rmg Sorry to keep on about this one but can you confirm if deploying locally on a production server is safe/secure for production and if there are any potential issues I need to consider while the above issue is open?

[bachirelkhoury]

related to #335 checkout my solution.

[rmg]

@idlescan that should be fine.

[idlescan]

@bachirelkhoury Thanks - I saw your issue previously but there doesn't actually seem to be many large files that I can delete. It's a 10MB deploy, which doesn't seem so big that it should cause an issue like this.

@rmg thanks, I'll do local deploys on the production server for now

[bachirelkhoury]

@idlescan this seems to be happening again. #335 turned out to be a random success. How are you deploying locally on production? using sl-pm? any help would be really appreciated.

[idlescan]

@bachirelkhoury yeah, I've got strong-pm service running on the production machine and I'm running slc build --npm followed by slc deploy on that machine - It's been working fine for the last while (though I haven't updated strongloop etc recently)

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[stale[bot]]

This issue has been closed due to continued inactivity. Thank you for your understanding.