Exceptions from monitor-ec2-build-slaves script

[Paul Sokolovsky]

We've been having quite a few of exception from build instance monitoring script lately, e.g.:

Traceback (most recent call last):
  File "linaro-aws-tools/monitor-ec2-build-slaves", line 146, in <module>
    process_instance(i)
  File "linaro-aws-tools/monitor-ec2-build-slaves", line 121, in process_instance
    slave_info = check_build_slave(i.id, owner, run_time)
  File "linaro-aws-tools/monitor-ec2-build-slaves", line 95, in check_build_slave
    slave_info = get_slave_info(instance_id, owner)
  File "linaro-aws-tools/monitor-ec2-build-slaves", line 70, in get_slave_info
    f = urllib2.urlopen("https://%s/jenkins/computer/%s/api/json?depth=2" % (host, instance_id))
  File "/usr/lib/python2.7/urllib2.py", line 126, in urlopen
    return _opener.open(url, data, timeout)
  File "/usr/lib/python2.7/urllib2.py", line 398, in open
    response = meth(req, response)
  File "/usr/lib/python2.7/urllib2.py", line 511, in http_response
    'http', request, response, code, msg, hdrs)
  File "/usr/lib/python2.7/urllib2.py", line 436, in error
    return self._call_chain(*args)
  File "/usr/lib/python2.7/urllib2.py", line 370, in _call_chain
    result = func(*args)
  File "/usr/lib/python2.7/urllib2.py", line 519, in http_error_default
    raise HTTPError(req.get_full_url(), code, msg, hdrs, fp)
urllib2.HTTPError: HTTP Error 404: Not Found

[Paul Sokolovsky (pfalcon)]

One obvious reason for these are slaves started by sandboxes. But generally, exceptions should be handled more gracefully.

[Alexander Sack (asac)]

On Fri, Jun 1, 2012 at 5:46 AM, Paul Sokolovsky
<email address hidden> wrote:
> One obvious reason for these are slaves started by sandboxes. But
> generally, exceptions should be handled more gracefully.

What are sandboxes?

--
Alexander Sack
Technical Director, Linaro Platform Teams
http://www.linaro.org | Open source software for ARM SoCs
http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

[Paul Sokolovsky (pfalcon)]

Hello Alexander,

On Tue, 5 Jun 2012 09:34:22 +0200
Alexander Sack <email address hidden> wrote:

> On Fri, Jun 1, 2012 at 5:46 AM, Paul Sokolovsky
> <email address hidden> wrote:
> > One obvious reason for these are slaves started by sandboxes. But
> > generally, exceptions should be handled more gracefully.
>
> What are sandboxes?

Sandboxes are test environments, more or less completely mimicking
production system:
http://en.wikipedia.org/wiki/Sandbox_%28software_development%29

In our case, we have sandboxing capability for Android Build System,
with individual sandboxes started on separate ec2 instances. As they
reuse configuration from production (including key names which we use
to find out slave ownership), we have this issue, when monitoring
script looks for a slave info on production server, and gets 404 Not
Found instead.

--
Best Regards,
Paul

Linaro.org | Open source software for ARM SoCs
Follow Linaro: http://www.facebook.com/pages/Linaro
http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

[Paul Sokolovsky (pfalcon)]

Corresponding changes were deployed.