Bug 1475821

Summary: Got disconnected from server with blk_update_request: I/O error, dev dm-0 error on server's console.
Product: [oVirt] ovirt-engine Reporter: Nikolai Sednev <nsednev>
Component: BLL.HostedEngineAssignee: Doron Fediuck <dfediuck>
Status: CLOSED WONTFIX QA Contact: Nikolai Sednev <nsednev>
Severity: high Docs Contact:
Priority: unspecified    
Version: 4.1.3.5CC: bugs, tnisan
Target Milestone: ---Flags: sbonazzo: ovirt-4.3-
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-07-16 08:39:01 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: Storage RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
sosreport from SPM host alma04 (host on which the deployment was initially started as on first host)
none
sosreport from the engine
none
sosreport from alma03 (second host)
none
screenshot of an error from remote console
none
screencast of disconnected alma03 after making changes to network none

Description Nikolai Sednev 2017-07-27 12:38:37 UTC
Description of problem:
Got disconnected from server with blk_update_request: I/O error, dev dm-0 error on server's console.

I had an open ssh connection to SPM ha-host, which I've turned in to maintenance from UI and attached one additional nene-required network to it, then I've tried to activate the server back and got disconnected. 
After some time I've power-cycled the server and after few minutes (around 10) it got active within the UI.

Version-Release number of selected component (if applicable):


How reproducible:
Sometimes - I do not have any specific reproduction steps for this one, just attaching the sosreport from host and engine to document this here.

Steps to Reproduce:
1.Deploy hosted-engine over iSCSI, on pair of hosts.
2.Add one iSCSI data storage domain.
3.Add one or two none-required networks.
4.Create multipath iSCSI bond to HE-storage domain.
5.Set in to maintenance the SPM host from UI.
6.Attach one of none-required networks to one of empty NICs on host in maintenance.
7.Activate back host.
8.Host loosing connectivity.

Actual results:
Host loosing connectivity.

Expected results:
Host should not get disconnected.

Additional info:
Sosreports from engine and SPM host being attached.

Comment 1 Nikolai Sednev 2017-07-27 12:41:26 UTC
Created attachment 1305351 [details]
sosreport from SPM host alma04 (host on which the deployment was initially started as on first host)

Comment 2 Nikolai Sednev 2017-07-27 12:42:29 UTC
Created attachment 1305352 [details]
sosreport from the engine

Comment 3 Nikolai Sednev 2017-07-27 12:44:10 UTC
Created attachment 1305353 [details]
sosreport from alma03 (second host)

Comment 4 Nikolai Sednev 2017-07-27 12:46:50 UTC
Created attachment 1305354 [details]
screenshot of an error from remote console

Comment 5 Nikolai Sednev 2017-07-27 13:42:44 UTC
Created attachment 1305379 [details]
screencast of disconnected alma03 after making changes to network

Comment 6 Tal Nisan 2018-07-16 08:39:01 UTC
Closing old bugs, feel free to reopen if still needed.