Bug 126285 - proxy: don't spam during an outage/traceback
Summary: proxy: don't spam during an outage/traceback
Keywords:
Status: CLOSED DUPLICATE of bug 309071
Alias: None
Product: Red Hat Satellite Proxy 5
Classification: Red Hat
Component: Server
Version: 420
Hardware: All
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Miroslav Suchý
QA Contact: Brandon Perkins
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2004-06-18 13:57 UTC by Jack Neely
Modified: 2008-07-29 15:47 UTC (History)
1 user (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2008-07-29 14:12:13 UTC
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)
06/18/04 -- SSL connect error (9.08 KB, text/plain)
2004-06-18 13:58 UTC, Jack Neely 		no flags Details
06/18/04 -- SSL read error (8.29 KB, text/plain)
2004-06-18 13:59 UTC, Jack Neely 		no flags Details
RHN Traceback (16.13 KB, text/plain)
2004-08-16 14:00 UTC, Jack Neely 		no flags Details
RHN Proxy traceback (4.59 KB, text/plain)
2004-08-23 13:33 UTC, Jack Neely 		no flags Details
View All

Description Jack Neely 2004-06-18 13:57:33 UTC 
Description of problem:
I got quite a few emails tonight from my RHN Proxy with a couple
different SSL errors connecting with the RHN parent.  This seems to be
a regular occurance.  The date in question is 

   Fri, 18 Jun 2004 06:33:25 -0400

I'm attaching a couple emails of the different tracebacks.
Comment 1 Jack Neely 2004-06-18 13:58:36 UTC 
Created attachment 101244 [details]
06/18/04 -- SSL connect error
Comment 2 Jack Neely 2004-06-18 13:59:41 UTC 
Created attachment 101245 [details]
06/18/04 -- SSL read error

This error happens quite often.
Comment 3 Todd Warner 2004-06-20 15:54:19 UTC 
We had a bit of downtime when we upgraded code on the hosted side at
those times. If everything is operating smoothly again, then... all
should be well.

Not overly pretty error messaging I know.. alas...

Lemme know if it happens again please.

Moving to a NEEDINFO status for the time being.
Comment 4 Jack Neely 2004-08-16 14:00:36 UTC 
Created attachment 102761 [details]
RHN Traceback

I'm assuming there was maintenance on RHN last night.  Got several hundred of
the attached email.  The feature request in this case is to have a system that
understands the difference between downtime and failure and have the Proxy
servers handle it properly.
Comment 5 Jack Neely 2004-08-23 13:33:36 UTC 
Created attachment 102981 [details]
RHN Proxy traceback

I got 400 copies of the above email this morning.
Comment 6 Jack Neely 2004-08-23 13:35:11 UTC 
Again, if there was a maintenance window last night 1) your users
would like to know and 2) the RHN PRoxy servers should be able to
handle the difference between maintenance and failures.
Comment 7 Todd Warner 2004-09-16 04:40:58 UTC 
outage policy is in the midst of revision. I assume you are on
rhn-outage-list?

Spamming admins: Yup. Changing summary to reflect this is the driving
                 issue here.

"socket.sslerror with Proxy connecting to parent"
--> "proxy: don't spam during an outage"

I.e., we need to figure out a way to inform, but not over-inform.
Probably need to digest emails maybe for tracebacks? Dunno.

Punting to you misa; it's your bag now.
Comment 9 Red Hat Bugzilla 2007-04-12 00:07:16 UTC 
User bnackash's account has been closed
Comment 10 Red Hat Bugzilla 2007-05-03 04:16:06 UTC 
User pvetere's account has been closed
Comment 12 Miroslav Suchý 2008-04-30 12:57:25 UTC 
I think this could we doable. We should request from hosted team some method to
indicate that is outage (like definition of sorry page).
Comment 13 Miroslav Suchý 2008-07-29 14:12:13 UTC 
Duplicate of 309071
Already fixed in RHN Proxy 5.1
We do not send email if there is connection error, just write up message to log.

*** This bug has been marked as a duplicate of 309071 ***
Note You need to log in before you can comment on or make changes to this bug.