Bug 1558194 - Missing "how to restart a dead/failed brick" in the troubleshooting section
Summary: Missing "how to restart a dead/failed brick" in the troubleshooting section
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Red Hat Gluster Storage
Classification: Red Hat Storage
Component: Documentation
Version: unspecified
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
: RHGS 3.4.z Batch Update 4
Assignee: Pratik Mulay
QA Contact: Anees Patel
Chandrakanth Pai
URL:
Whiteboard:
Depends On:
Blocks: 1672843
TreeView+ depends on / blocked
 
Reported: 2018-03-19 19:44 UTC by Shyamsundar
Modified: 2019-06-03 05:12 UTC (History)
10 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2019-06-03 05:12:09 UTC
Embargoed:


Attachments (Terms of Use)

Description Shyamsundar 2018-03-19 19:44:13 UTC
Document URL: 
I searched in this guide: https://access.redhat.com/documentation/en-us/red_hat_gluster_storage/3.3/html-single/administration_guide/#Troubleshooting1

The above link is the only place I found how to restart a down brick process.

Section Number and Name: 
None, as the requested data is missing

Describe the issue: 
We need a troubleshooting or recover step that details how to start a brick that is no longer up and running.

The solution seems to be "gluster volume start <volname> force", but is not documented anywhere.

Suggestions for improvement: 
As above

Additional information: 
While we are at it, it may serve us to check and see what failure recovery steps for other such scenarios are covered in the documentation.

Comment 4 Atin Mukherjee 2019-03-11 14:56:44 UTC
I keep on forgetting the credential for access.qa.redhat.com, so can't log into it till you pass back the user/password, Pratik.

In the mean time can you please paste the delta here?

Comment 5 Atin Mukherjee 2019-03-11 15:10:21 UTC
Thanks Pratik for passing down the credential.

While the changes look mostly good, I think it'd be worth to mention (as a note) that in case of an OCS converged and independent mode where brick multiplexing is enabled by default, we'd need to force start all the volumes for which the bricks are multiplexed into a single process which went down. Makes sense?

Comment 10 Anees Patel 2019-03-13 06:31:25 UTC
Looks good to me,
Moving this to verified.


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