Hi Dan and Charles

Dan, thanks again for all your hard work on NUT.
I really much appreciated that you satisfied all my previous comments!
I know how hard it is ;)

2013/11/6 [email protected] <[email protected]>

> In my opinion, there might be just one problem: the blzr driver
> doesn't provide anymore two splitted drivers (serial and usb), so this
> isn't going to be a 'silent update' for those using these drivers
> (should we provide symlinks or the like?).
> If this doesn't sound like a problem to you, I'll take care of the rename.
>

ah, good point! I completely missed that.

I must admit that I wasn't too thrilled at first by this merge.
however, considering the specific nature of the Q* family (and I can tell
that we've just seen the emerged part of the iceberg!), and the upcoming
transition, it makes sense!
the code overhead is low, and most if not all recent Q units come with USB
only (hence, the serial support overhead is ever lower)
however (again), considering the overall changes in the driver approach, we
can't force the usual transition!
we really need to collect a minimum feedback, with enough coverage, before
forcing the switch.

there remains only one option:
- keep blazer_ser and _usb,
- rename blzr to "blazer"
    => will collide on the manpage!
    => I would prefer "nutdrv_qx", to initiate a new driver naming effort
- suggest user to transition and report on "blazer" ("nutdrv_qx").
    => double HCL entries is the best way to advertise (i.e. "blazer or
blazer_{ser,usb}"), along with UPGRADING
    => call to user tests on the list

this last point brings another one:
why do you specify the voltronic protocol in HCL entries? (i.e. "blzr
protocol=voltronic")
as the default, you shouldn't need so!
or did I miss something?

Thanks and cheers,
Arno
_______________________________________________
Nut-upsdev mailing list
[email protected]
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/nut-upsdev

Reply via email to