On Mon, Jan 23, 2012 at 14:30:54 +0100, Jiri Denemark wrote:
We already provide ways to detect when a domain has been paused as a result of I/O error, but there was no way of getting the exact error or even the device that experienced it. This new API may be used for both. ... + +/** + * virDomainGetIoError: + * @dom: a domain object + * @dev: disk device to be inspected or NULL + * @flags: extra flags; not used yet, so callers should always pass 0 + * + * The @disk parameter is either the device target (the dev attribute of + * target subelement), such as "vda", or NULL, in which case all disks will be + * inspected for errors. If only one disk experienced an I/O error, that error + * will be returned. However, if there are more disks with I/O errors, this + * function will fail and return -2, requiring the caller to check every + * device explicitly. + * + * Returns -2 if @dev is NULL and there are multiple disks with errors, -1 if + * the function fails to do its job, the I/O error (virDomainIoError) observed + * on the specified disk (or any disk if @dev is NULL). Namely, 0 is returned + * when there is no error on the disk. + */ +int +virDomainGetIoError(virDomainPtr dom, + const char *dev, + unsigned int flags)
Actually, after talking a bit more about this with Federico on IRC, this API should rather provide a list of disks with error codes. I'll create a v2, in which I will also address Michal's comments. Jirka