[pve-devel] Live storage migration v4

[Alexandre DERUMIER]

I need to do a fix for my last patches,

as block-job-complete is async according to 
https://bugzilla.redhat.com/show_bug.cgi?id=1119173

(pending write are flushed later)


so I need to check in the query-block-job if the job is correctly stopped.
(That's the correct behavior in current proxmox code)



----- Mail original -----
De: "aderumier" <aderumier at odiso.com>
À: "pve-devel" <pve-devel at pve.proxmox.com>
Envoyé: Lundi 24 Octobre 2016 23:24:05
Objet: [pve-devel] Live storage migration v4

changelog: 

-increase block-job-cancel && compplete qmp timeout. Previously it's "worked" because 
we called qmp query-block just before (with a timeout of 10min). 
qmp socket is busy when block jobs are running, so we need to increase timeout 
for cancel. 

- qemu_blockjobs_cancel : simply cancel jobs list. (with qmp timeout increased it's working 100%) 

- qemu_drive_mirror_monitor : only check mirroring jobs and not all jobs 

- vm start with targetstorage : cleanup code from wolfgang comments 

- use qmp for nbd start (missing data field) 



unchanged: 
- skipcomplete logic, feel free to improve it ;) 



_______________________________________________ 
pve-devel mailing list 
pve-devel at pve.proxmox.com 
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel