Hi all, FUJITA Tomonori wrote: > On Mon, 20 Apr 2009 11:55:52 +0200 > Hannes Reinecke <hare@xxxxxxx> wrote: > >> Hi Mike, >> [ .. ] >>> I was thinking that maybe using kobject_uevent_env would be better. The >>> info that gets passed to userspace would be the decoded sense and >>> asc/ascq based on values from the drivers/scsi/constants.c. >>> >> No. This patch has the possibility of generating _huge_ amounts of >> messages, most of which are information only and of no influence >> to the actual operation. >> udev would be flooded with it and won't be able to react to 'important' >> messages while processing them. > > Do we really have huge amount of messages, errors, unit attentions, > etc? > There is a potential for that. Things which would normally ignored/retried silently (like UNIT ATTENTION) will suddenly be visible. And what's more, we'll be likely be getting plenty of errors if the target has some failure. And these will get multiplied when having a multipathed setup. But this is exactly when we rely on udev to process any 'real' events like device remove etc in time. So I'd rather use a separate mechanism for this. > We already have a mechanism to send events to user space, > sdev_evt_send(). Could we simply use (or extend) it? > No, not really for above reasons. Cheers, Hannes -- Dr. Hannes Reinecke zSeries & Storage hare@xxxxxxx +49 911 74053 688 SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Markus Rex, HRB 16746 (AG Nürnberg) -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
