On 12/1/20 4:39 PM, Ján Tomko wrote:
On a Tuesday in 2020, Michal Privoznik wrote:
If there is an error getting info from guest agent, then the control on qemuDomainGetGuestInfo() jumps onto 'exitagent' label and subsequently continues on 'endagentjob'. Both labels are hit also in success case, which is why there is a code that tries to match info obtained from the guest agent with domain definition.
I'm confused by 'exitagent' and 'exitagentjob' being above code that is only done (or only makes sense) on success. And ret being set to zero so early - I guess that's due to the nature of the best-effort information gathering here. But I think it would be perfectly fine to error out if we fail to get a query job or the domain dies in the meantime.
Moving the exitagent and endagentjob labels after the cleanup block would remove the need to check ret. (i.e. duplicating ExitAgent and EndAgentJob calls - one pair that would be exectued on success and one pair only on failure)
Fair enough. Will post v2. Michal