by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
Dennis Metzger
@dmetzger57
Can we not just take these changes for Tuesday’s Beta 1 build? One less build, more time to review / merge. I believe QE has plenty to test with this weeks build already. Rushing review on this big a change is probably risky
Brett Thurber
@bthurber
@dmetzger57 per our email thread earlier this week QE wanted to get the bits as-soon-as possible. This specific question came up. @smallamp responded he would like to have them in an alpha build for tomorrow. If beta 1 makes more sense I am fine with that if QE acks.
Dennis Metzger
@dmetzger57
What wasn’t clear (at least not to me) was the scale of change, hence my hesitation now. If folks think the can thoroughly review and have any comments addressed by tomorrow I’m OK, just raising a flag.
smallamp
@smallamp
@dmetzger57 @bthurber I agree.. If we are not ready yet with changes/reviews, there is no point cutting the build.. Would rather take a beta build that has gone thru thorough reviews and tests than a rushed up alpha build..
Brett Thurber
@bthurber
@dmetzger57 +1. @gmcculloug if you feel we need a few extra days we will go with that. @smallamp :thumbsup:
Greg McCullough
@gmcculloug
I haven't not had a chance to look at the PR much more then basic stats and assigning it. I'll make sure Tina/Madhu get some time on it today, but Madhu is almost fully committed to Service Portal work and is on PTO starting next Wed so I need to get some items from him for that project. Either way, if we request changes I assume @fdupont-redhat will work on them for tomorrow and we will need to review those as well. On top of that I'm taking PTO tomorrow.
Brett Thurber
@bthurber
ack...he will be available if/when issues come up during review. Beta 1 is critical for us to hit. If we miss that we move to red and are at risk for 1.1 GA.
Kedar Kulkarni
@kedark3
@michaelkro https://10.8.196.109/migration#/ @AparnaKarve
stalled migration
Michael Ro
@michaelkro
taking a look, thanks KK :)
Kedar Kulkarni
@kedark3
its stuck in Pre-migration Status: Acquire Transformation Host
for 1 hour +
Michael Ro
@michaelkro
@kedark3, when I open the mapping wizard, it looks like the target cluster does not have a conversion host configured
Kedar Kulkarni
@kedark3
yes
so migration should ideally notice it and fail
because that is what used to happen in previous builds
Michael Ro
@michaelkro
is that possibly an automate thing? The UI changes state depending on the status of each plan request
Kedar Kulkarni
@kedark3
@michaelkro I'll file a bz
Michael Ro
@michaelkro
I'll take a look at the data and see if I see anything out of place
Kedar Kulkarni
@kedark3
yes let me know if you find anything specific
Michael Ro
@michaelkro
will do. Looking now :)
Kedar Kulkarni
@kedark3
thanks a lot :)
Michael Ro
@michaelkro
@kedark3 it looks like cancellation was requested for the plan named plan_2COTLJiUTg
was it supposed to stop migration?
Kedar Kulkarni
@kedark3
yes I requested that after first 5 mins of watching it trying to acquire the host
Michael Ro
@michaelkro
hmm, that's not good
was this the first and only time this plan has been run?
Kedar Kulkarni
@kedark3
started two more plans to see if it gets stuck same way
1 failed, 1 going on
asdetrewt
Michael Ro
@michaelkro
interestingly, there are two requests for plan_2COTLJiUTg
Aparna Karve
@AparnaKarve
both in-progress plans have 2 requests each. The most recent request (which is currently being executed) for each plan has an empty tasks array
Kedar Kulkarni
@kedark3
can you check asdetrewt ??
that is latest, just started, cancel operation is not requested unlike other two
Michael Ro
@michaelkro
that one only has one
I wonder if this is the change that is causing it to stall
message: "State5 is not finished yet [13/86400 retries]."
how long does each retry take?
this is the entire object
/state5:
status: "active"
weight: 1
description: "Acquire Transformation Host"
message: "State5 is not finished yet [15/86400 retries]."
started_on: "2018-10-18T19:38:57Z"
percent: 0.017361111111111112
updated_on: "2018-10-18T19:43:58Z"
Kedar Kulkarni
@kedark3
just for checking it, I added one transformation host
Aparna Karve
@AparnaKarve
asdetrewt seems to have failed due to this -
"/State11/State2": {
              "status": "failed",
              "weight": 5,
              "description": "Convert disks",
              "message": "undefined method `storage' for nil:NilClass",
              "started_on": "2018-10-18T19:44:56Z",
              "percent": 100.0,
              "updated_on": "2018-10-18T19:44:57Z"
            }
Kedar Kulkarni
@kedark3
@AparnaKarve that one failed as soon as I added 1 conversion host
so real issue is, should we really be trying 86400 times for conversion host?
Michael Ro
@michaelkro
those retries seem awfully slow
Kedar Kulkarni
@kedark3
about the "undefined methodstorage' for nil:NilClass"` QE is going to run all migrations and if we see this as an issue we will file a bz.
Kedar Kulkarni
@kedark3
@bthurber yesterday I migrated an ansible tower VM from vmware to rhv with 5.10.0.20.20181016163900_fe677b4 successfully
and post migration it was up and running tower
Brett Thurber
@bthurber
@kedark3 awesome!
Kedar Kulkarni
@kedark3