Doc missing: env http_proxy must be disabled when starting nova daemons by hand

[termie]

Hi,

I am trying to setup a multi-node installation of nova. After installing nova and other related tools I tried to bundle and upload the image using euca2ools. Here I am able to bundle but unfortunately not able to upload. An exception is thrown when I run the following command.

euca-upload-bundle -m /tmp/kernel.manifest.xml -b mybucket

Checking bucket: mybucket Warning: failed to parse error message from AWS: :7:2: mismatched tag Traceback (most recent call last): File "/usr/bin/euca-upload-bundle", line 231, in main() File "/usr/bin/euca-upload-bundle", line 214, in main bucket_instance = ensure_bucket(conn, bucket, canned_acl) File "/usr/bin/euca-upload-bundle", line 87, in ensure_bucket bucket_instance = connection.get_bucket(bucket) File "/usr/lib/pymodules/python2.6/boto/s3/connection.py", line 275, in get_bucket rs = bucket.get_all_keys(headers, maxkeys=0) File "/usr/lib/pymodules/python2.6/boto/s3/bucket.py", line 204, in get_all_keys headers=headers, query_args=s) File "/usr/lib/pymodules/python2.6/boto/s3/connection.py", line 342, in make_request data, host, auth_path, sender) File "/usr/lib/pymodules/python2.6/boto/connection.py", line 459, in make_request return self._mexe(method, path, data, headers, host, sender) File "/usr/lib/pymodules/python2.6/boto/connection.py", line 435, in _mexe raise BotoServerError(response.status, response.reason, body) boto.exception.BotoServerError: BotoServerError: 503 Service Unavailable

Network Error (tcp_error)

A communication error occurred: "Operation timed out"

The Web Server may be down, too busy, or experiencing other problems preventing it from responding to requests. You may wish to try again at a later time.

For assistance, contact your network support team.

Before running the above command, I sourced the "novarc" file. But still I am getting the same exception. Could anyone help me on this?

Regards, Raaj

---

Just copied the old question. Maybe it´s only a cosmetic feature but when you are running the single daemons e.g. "nova-api --nodaemon --verbose" you must unset http_proxy and https_proxy (both?) or you will receive an unknown error. Just the same when using the euca-tools.

---

Imported from Launchpad using lp2gh.

• date created: 2011-02-14T17:57:41Z
• owner: naehring
• the launchpad url was https://bugs.launchpad.net/bugs/718900

[termie]

(by ttx) What is the bug you're trying to report ? that we should clear the env so that http_proxy is not honored by nova when you start with it ?

[termie]

(by naehring) The problem can be divided into two parts. The first one is that you will receive unknown errors when you start the nova-services from command line, maybe including "--nodaemon" to have the output directly. When a proxy is set before you start, you'll receive the unknown error. Unsetting the variable resolves the problem.

The other part belongs to the use of euca-describe-instances, euca-describe-images and maybe the rest of the tools. If here the environment proxy is set, you will receive the traceback (compare the doubled bugreport).

I think the easiest way is to ignore (or clear) http_proxy and maybe https_proxy before starting the services or using the euca2ools.

[termie]

(by philippe-berthault) Le 23/02/2011 14:33, Thierry Carrez a écrit :

What is the bug you're trying to report ?

The bug is that this restriction (« http_proxy must be disabled » ) isn't indicated in the "OpenStack Compute Administration Manuel", nor on the Wiki pages.

There is just one sentence in the administration manuel about proxy: « You may also need to check your proxy settings to see if they are causing problems with the novarc creation. » which doesn't explain what to do with the proxy.

[termie]

(by ttx) OK, editing title and adding doc tag.