Bug 1399512 - Bug to track Pulp issues with the Satellite 6 Get well plan
Summary: Bug to track Pulp issues with the Satellite 6 Get well plan
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Red Hat Satellite
Classification: Red Hat
Component: Pulp
Version: Unspecified
Hardware: All
OS: All
unspecified
unspecified vote
Target Milestone: Unspecified
Assignee: satellite6-bugs
QA Contact: Katello QA List
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2016-11-29 08:47 UTC by Sam Nelson
Modified: 2017-01-12 18:14 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2017-01-12 18:14:21 UTC
Target Upstream Version:


Attachments (Terms of Use)

Description Sam Nelson 2016-11-29 08:47:44 UTC
Description of problem: Pulp component is one of the components which generates cases resulting in negative customer unsatisfaction and frustration. It is also one of the components that engineers ned to spent a lot of time ojn

This bug is being opened as part of the Sat 6 Get well program and will be used to track Pulp issues

Some of the common bugs and generic issues are :

https://bugzilla.redhat.com/show_bug.cgi?id=1223023: [RFE] Allow Pulp to force resync a repository
- hit by many customers
- basic robustness requirement


https://bugzilla.redhat.com/show_bug.cgi?id=1374723:
[Improvement] prevent running pulp-manage-db when pulp workers are up
- without that, easy to corrupt pulp DB during upgrade
- Sat6.2 upgrades were affected by lack of it


(still) improve generate applicability task performance/scalability
- even in Sat6.2.4, pulp related tasks spend big deal of time in reg.app.
- either implement bz1309673 (make calculation incremental), or allow the tasks running in parallel (by celery workers)


(generic): supportability of pulp-admin?
- officially, we do not support the tool (it has some limitations/drawbacks)
- BUT we often use it as workarounds/diagnosis
- either make it "official" or replace by something else


(generic, not such prio, issue) - handle own exceptions better (than skipping them)
- worth logging what exception (that often means something wrong happened) happened
- example: bz1327522


(no bugzilla): get rid of mongodb in favour of a real (relational) DB
- pulp requires relational DB what mongo isnt (this affects also workarounds we offer to custs)
- unable to build DB schema for pulp database (affects investigation)
- I know about plans to migrate to psql, not sure the state


Actual results: High Resolution times, frustration from customers regarding functionality issues. not functional DB- mongodb . Buggy . need to use unofficial pulpadmin


Expected results: Lower resolution times, user and customer module, lesser bugs


Additional info:

Comment 1 Brad Buckingham 2016-11-29 19:09:26 UTC
Sam, should this be defined as a 'Tracker' bugzilla with 'blocks' references to each of the bugzillas being tracked?

Comment 3 Sam Nelson 2016-12-12 12:34:50 UTC
Hello Michael,


The purpose of the bug is for tracking purpose for the management of both the GSS team and the Software engineering team as part of the Satelite 6 Get well plan. Yes, other feedbacks and discussions are being held, but this bug opened for us to ensure that pulp issues is being tracked and will be fixed,.


Note You need to log in before you can comment on or make changes to this bug.