Undefined error when deploying from Web UI

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Undefined error when deploying from Web UI

MichaelCurtissPatrick
Hi,

I'm working with a brand new, very vanilla, apt-get installation Artifactory (4.2.2).  Whenever I try to deploy using the Web UI Deploy button I get an error saying the file is "undefined".  Steps to reproduct:

- Bring up artifactory web page.
- Log in as an admin user
- Click Artifacts
- Select repository (the repostiory is type generic)
- Click Deploy button
- Select a file
- Immediately get error message.  See attached screenshot:  
Reply | Threaded
Open this post in threaded view
|

Re: Undefined error when deploying from Web UI

denebeim
MichaelCurtissPatrick wrote
I'm working with a brand new, very vanilla, apt-get installation Artifactory (4.2.2).
I'm the guy who did the artifactory debian packaging.

Looking at your screen shot doesn't really tell me what you're trying to do.  How are you doing the upload.  Also the logs are usually very useful, can you please provide ARTIFACTORY_HOME/logs/artifactory.log.

We'll need to reproduce it, however I don't think that this has anything to do with the debian packaging.  Once we see this we can figure out where to go from there.
Reply | Threaded
Open this post in threaded view
|

Re: Undefined error when deploying from Web UI

sblaes
I'm seeing this too. I'm just trying to upload a jar to one of my Maven local repositories. I tailed all the logs just now while I did it, but there are no log entries at all (in the access logs, artifactory.log, or anything else in /var/opt/jfrog/artifactory/logs) while this happens.

I'm using Artifactory 4.10.0 rev 40236.

Upon further digging, the request is getting rejected by nginx. I'm using the docker version of Artifactory. I tailed all the nginx logs, and I see the following as soon as I drop a file onto the deploy frame:

==> error.log <==
2016/07/31 08:44:53 [error] 33#33: *2700 client intended to send too large body: 2482497 bytes, client: 10.128.20.21, server: _, request: "POST /artifactory/ui/artifact/upload HTTP/1.1", host: "artifactory.scs.skybox.com", referrer: "https://artifactory.scs.skybox.com/artifactory/webapp/"

==> access.log <==
10.128.20.21 - - [31/Jul/2016:08:44:55 +0000] "POST /artifactory/ui/artifact/upload HTTP/1.1" 413 601 "https://artifactory.scs.skybox.com/artifactory/webapp/" "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/51.0.2704.103 Safari/537.36" "-"

This implies that we're hitting the default server that is mapped to 443 and 80 on the box, and I noticed that you did not set the property client_max_body_size 0; and chunked_transfer_encoding on; for that one, like you did on all of the others. I added this to mine, and it resolved the problem. You should fix this in your docker containers for the future though.