Bugzilla will be upgraded to version 5.0. The upgrade date is tentatively scheduled for 2 December 2018, pending final testing and feedback.
Bug 1076310 - [Docs] [Feature] Maximum Downtime per guest during migration now configurable
[Docs] [Feature] Maximum Downtime per guest during migration now configurable
Status: CLOSED CURRENTRELEASE
Product: Red Hat Enterprise Virtualization Manager
Classification: Red Hat
Component: Documentation (Show other bugs)
3.4.0
Unspecified Unspecified
unspecified Severity unspecified
: ---
: ---
Assigned To: Lucy Bopf
Laura Novich
:
Depends On:
Blocks:
  Show dependency treegraph
 
Reported: 2014-03-14 00:16 EDT by Zac Dover
Modified: 2016-01-26 20:58 EST (History)
7 users (show)

See Also:
Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2014-09-11 14:19:19 EDT
Type: Bug
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---


Attachments (Terms of Use)

  None (edit)
Description Zac Dover 2014-03-14 00:16:43 EDT
2. What is the nature and description of the request?
Currently maximum allowed downtime during live migrations is controlled by 'migration_downtime' global setting in vdsm.conf and applies to all guests.

Customer needs the ability to configure maximum allowed downtime during live migrations on a per guest basis.

3. Why does the customer need this? (List the business requirements here)
Different guests run different workloads and have different SLAs.

For very I/O or memory intensive guests default value of 0.5 seconds is not enough for live migration to converge and complete successfully.

For very critical guests higher downtimes than 0.5 seconds can cause noticeable downtime (e.g. guests are expelled from JBoss clusters thus causing evident disruption to end user whose session is lost).

4. How would the customer like to achieve this? (List the functional requirements here) 
Maximum allowed downtime during live migrations becomes a configurable property of each guest.

5. For each functional requirement listed in question 4, specify how Red Hat and the customer can test to confirm the requirement is successfully implemented. 
Customer can configure different maximum allowed downtimes during live migrations for different guests according to their workload and SLA requirements.
Maximum allowed downtime for each guest is honoured.
  -- https://bugzilla.redhat.com/show_bug.cgi?id=994391

See also:
https://bugzilla.redhat.com/show_bug.cgi?id=105543
Comment 1 Lucy Bopf 2014-04-10 21:13:40 EDT
New check box "Use custom migration downtime" appears in Hosts tab of New/Edit Virtual Machine window. Allows administrator to enter a value (in milliseconds).

Updated Virtual Machine Host Settings section of Admin Guide to include the new check box and its functionality.

Added additional tab names ('System' and 'Host') to Editing Virtual Machine Properties section of Admin Guide.
Comment 2 Lucy Bopf 2014-04-16 01:45:28 EDT
Documentation Link
------------------------------
http://documentation-devel.engineering.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.4/html-single/Administration_Guide/index.html#Virtual_Machine_Host_settings_explained

http://documentation-devel.engineering.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Virtualization/3.4/html-single/Administration_Guide/index.html#Editing_Virtual_Machine_Properties1

What Changed
------------------------------
The following topics were revised to include this feature. Table 9.7 was updated to include the 'Use custom migration downtime' check box and its function. Step 2 of procedure 9.9 was updated to include additional tabs.

Virtual Machine Host Settings Explained [ 10396-629233 ]

Editing Virtual Machine Properties [ 12677-629235 ]

Updated revision history: [ 30102-629236 ]

NVR
------------------------------
Red_Hat_Enterprise_Virtualization-Administration_Guide-3.4-web-en-US-3.4-19.el6eng

Moving to ON_QA.

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