All of lore.kernel.org
 help / color / mirror / Atom feed
* Redmine problem
@ 2017-11-08 21:54 Nathan Cutler
  2017-11-08 21:55 ` David Galloway
  0 siblings, 1 reply; 5+ messages in thread
From: Nathan Cutler @ 2017-11-08 21:54 UTC (permalink / raw)
  To: David Galloway; +Cc: ceph-devel, Dachary, Loic

Hi David:

In order to automate creation of backport tracker issues, we use a tool 
written by Loic that is essentially a glorified python script. It has 
worked flawlessly for years. Today, all of a sudden, it started failing 
with

requests.exceptions.ConnectionError: ('Connection aborted.', error(104, 
'Connection reset by peer'))

This error message is emanating from the python-requests module, which 
is used to communicate with Redmine (probably via its REST API) over 
http. Apparently, the other side (i.e. tracker.ceph.com) is resetting 
the connection without warning.

Did something change recently on the server side? Like, in the last two 
or three days? If so, I'm curious to hear what, since it would probably 
help me to debug this issue.

Thanks,
Nathan

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Redmine problem
  2017-11-08 21:54 Redmine problem Nathan Cutler
@ 2017-11-08 21:55 ` David Galloway
  2017-11-08 22:03   ` Nathan Cutler
  0 siblings, 1 reply; 5+ messages in thread
From: David Galloway @ 2017-11-08 21:55 UTC (permalink / raw)
  To: Nathan Cutler; +Cc: ceph-devel, Dachary, Loic


On 11/08/2017 04:54 PM, Nathan Cutler wrote:
> Hi David:
> 
> In order to automate creation of backport tracker issues, we use a tool
> written by Loic that is essentially a glorified python script. It has
> worked flawlessly for years. Today, all of a sudden, it started failing
> with
> 
> requests.exceptions.ConnectionError: ('Connection aborted.', error(104,
> 'Connection reset by peer'))
> 
> This error message is emanating from the python-requests module, which
> is used to communicate with Redmine (probably via its REST API) over
> http. Apparently, the other side (i.e. tracker.ceph.com) is resetting
> the connection without warning.
> 
> Did something change recently on the server side? Like, in the last two
> or three days? If so, I'm curious to hear what, since it would probably
> help me to debug this issue.
> 

Could it possibly be because Loic no longer has Redmine Administrator
privileges per his request?

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Redmine problem
  2017-11-08 21:55 ` David Galloway
@ 2017-11-08 22:03   ` Nathan Cutler
  2017-11-08 22:40     ` David Galloway
  0 siblings, 1 reply; 5+ messages in thread
From: Nathan Cutler @ 2017-11-08 22:03 UTC (permalink / raw)
  To: David Galloway; +Cc: ceph-devel, Dachary, Loic

> Could it possibly be because Loic no longer has Redmine Administrator
> privileges per his request?

I don't think so. Was the revocation of Loic's admin privs done within 
the past 72 hours?

(The script uses the Redmine key provided on the command line, and I'm 
providing my own key since that's the only one I have.)

Nathan

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Redmine problem
  2017-11-08 22:03   ` Nathan Cutler
@ 2017-11-08 22:40     ` David Galloway
  2017-11-08 22:50       ` Nathan Cutler
  0 siblings, 1 reply; 5+ messages in thread
From: David Galloway @ 2017-11-08 22:40 UTC (permalink / raw)
  To: Nathan Cutler; +Cc: ceph-devel, Dachary, Loic



On 11/08/2017 05:03 PM, Nathan Cutler wrote:
>> Could it possibly be because Loic no longer has Redmine Administrator
>> privileges per his request?
> 
> I don't think so. Was the revocation of Loic's admin privs done within
> the past 72 hours?

Yes.

> 
> (The script uses the Redmine key provided on the command line, and I'm
> providing my own key since that's the only one I have.)
> 

Can you share the script without the API key and maybe some timestamps
so I can check out the logs?

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: Redmine problem
  2017-11-08 22:40     ` David Galloway
@ 2017-11-08 22:50       ` Nathan Cutler
  0 siblings, 0 replies; 5+ messages in thread
From: Nathan Cutler @ 2017-11-08 22:50 UTC (permalink / raw)
  To: David Galloway; +Cc: ceph-devel, Dachary, Loic

> Can you share the script without the API key and maybe some timestamps
> so I can check out the logs?

The Redmine bits of the script are in this file: 
http://ceph-workbench.dachary.org/root/ceph-workbench/blob/master/ceph_workbench/wbredmine.py

I first started seeing this at around 11:40 a.m. CET (10:40 UTC) and it 
was happening all afternoon - the script would run for some time and 
then fail with the same error, but not at the same spot. Sometimes it 
would process 10 or 20 pending backport issues, other times it would get 
through more.

I just ran it again and it completed twice in a row without error, so I 
guess it was a transient issue. Timestamp after completing the second run:

$ date --utc
Wed Nov  8 22:48:34 UTC 2017

Thanks for looking into it.

Nathan

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2017-11-08 22:50 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-11-08 21:54 Redmine problem Nathan Cutler
2017-11-08 21:55 ` David Galloway
2017-11-08 22:03   ` Nathan Cutler
2017-11-08 22:40     ` David Galloway
2017-11-08 22:50       ` Nathan Cutler

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.