Bug 21178
| Summary: | voltag selector, and unload support for chio(1). | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Base System | Reporter: | csg <csg> | ||||
| Component: | bin | Assignee: | Kenneth D. Merry <ken> | ||||
| Status: | Closed FIXED | ||||||
| Severity: | Affects Only Me | CC: | ajk | ||||
| Priority: | Normal | ||||||
| Version: | 4.0-STABLE | ||||||
| Hardware: | Any | ||||||
| OS: | Any | ||||||
| Attachments: |
|
||||||
Why doesn't somebody just merge the equivalent fixes from NetBSD's chio into
FreeBSD? Or where they already done?
>
> >Number: 21178
> >Category: bin
> >Synopsis: voltag selector, and unload support for chio(1).
> >Confidential: no
> >Severity: non-critical
> >Priority: low
> >Responsible: freebsd-bugs
> >State: open
> >Quarter:
> >Keywords:
> >Date-Required:
> >Class: change-request
> >Submitter-Id: current-users
> >Arrival-Date: Sun Sep 10 16:30:01 PDT 2000
> >Closed-Date:
> >Last-Modified:
> >Originator: C. Stephen Gunn
> >Release: FreeBSD 4.0-STABLE i386
> >Organization:
> Waterpout Communications, Inc.
> >Environment:
>
> Most versions of FreeBSD.
>
> >Description:
>
> chio(1) provides no mechanism to select media by "voltag."
> chio(1) provides no mechanism to return a media unit to its source.
>
> >How-To-Repeat:
>
> N/A - new feature request.
>
> >Fix:
>
> o A pseudo-element type selector "voltag" that can be used to select
> source elements:
>
> chio move voltag VOLUME-03 drive 0
>
> chio exchange voltag VOLUME-03 voltag VOLUME-04
>
> o A new sub-command "unload" that returns a media unit to the source
> element reported by 'chio status -a':
>
> chio unload drive 0
> chio unload voltag VOLUME-03
>
> A patch relative to 4.1-STABLE (should work on 5.0-CURRENT as well) is
> available at:
>
> http://www.physics.purdue.edu/~csg/FreeBSD/chio.patch
>
>
> >Release-Note:
> >Audit-Trail:
> >Unformatted:
>
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-bugs" in the body of the message
>
On Sun, 10 Sep 2000 16:40:33 MST, Matthew Jacob wrote: > Why doesn't somebody just merge the equivalent fixes from NetBSD's chio in > to FreeBSD? Or where they already done? This isn't actually a fix. It allows you to select a media unit to move by voltag. It should make it easier for products like amanda to drive a tape-changer with barcode reader on FreeBSD. You can simply ask for a specific tape by voltag: chio move voltag VOLUME-03 drive 0 The other feature is the "unload" command which checks the source parameter returned in the changer_element_status record. If there's a source parameter, we "unload" the media by moving it back where the drive reports it came from. This the logic to request a certain tape (if you have a barcode reader or equivalent): chio unload drive <target-drive> chio move voltag <label> drive <target-drive> If the second one fails, the tape isn't in the changer, raccoons have eaten your barcodes, or the kids put bubblegum on the barcode reader optics. ;-) - Steve PS. the patch URL was incorrect before, I've fixed it (on my end). -- C. Stephen Gunn URL: http://www.waterspout.com/ WaterSpout Communications, Inc. Email: csg@waterspout.com 427 North 6th Street Phone: +1 765.742.6628 Lafayette, IN 47901 Fax: +1 765.742.0646 hmm. Sounds good, actually. I'm not chio's maintainer tho. On Sun, Sep 10, 2000 at 18:19:40 -0500, csg@waterspout.com wrote: > >Description: > > chio(1) provides no mechanism to select media by "voltag." > chio(1) provides no mechanism to return a media unit to its source. > > >How-To-Repeat: > > N/A - new feature request. > > >Fix: > > o A pseudo-element type selector "voltag" that can be used to select > source elements: > > chio move voltag VOLUME-03 drive 0 > > chio exchange voltag VOLUME-03 voltag VOLUME-04 > > o A new sub-command "unload" that returns a media unit to the source > element reported by 'chio status -a': > > chio unload drive 0 > chio unload voltag VOLUME-03 > > A patch relative to 4.1-STABLE (should work on 5.0-CURRENT as well) is > available at: > > http://www.physics.purdue.edu/~csg/FreeBSD/chio.patch Looks like a pretty good idea. I took a cursory glance through the patch, and noticed the 'XXX' comment about looking through pickers and portals. I'd suggest looking at them as well, even if it won't be used very often. (I've moved things from pickers and portals often enough, though.) I also wonder about the 'unload' keyword. It almost sounds like something you might use to eject a tape or something. What about calling it 'return' instead? (I don't feel really strongly about this, but I figure we ought to make things intuitive, if possible.) Anyway, I'll try to give the patch a more thorough look-over in the next few days. I won't be able to do more than that before committing it, though, since I don't have a test changer anymore. (You've obviously tested it, though, so we should be okay in that department.) Ken -- Kenneth Merry ken@kdm.org On Mon, 11 Sep 2000 00:56:23 CST, "Kenneth D. Merry" wrote: > Looks like a pretty good idea. > > I took a cursory glance through the patch, and noticed the 'XXX' comment > about looking through pickers and portals. Yeah. I didn't do that because of the way the ioctl() works to fetch information from the drive. I guess it's not a big deal though. Since I don't have any pickers or portals on my test drive, I'm not familiar with when a tape might be in one. ( I guess I can figure the portal one out real easy ). > I'd suggest looking at them as well, even if it won't be used very often. > (I've moved things from pickers and portals often enough, though.) I can add this. I've got some other pending comments from my homeboys on the cc: line. I'll address those as well, and submit a more recent patch early this week. > I also wonder about the 'unload' keyword. It almost sounds like something > you might use to eject a tape or something. What about calling it 'return' > instead? (I don't feel really strongly about this, but I figure we ought > to make things intuitive, if possible.) <grin> I use it exclusively to "unload" a tape from the drive to the slot that it came from. In my cursory investigation, drives (and perhaps pickers/portals) are the only changer elements with a valid source parameter. > Anyway, I'll try to give the patch a more thorough look-over in the next > few days. I won't be able to do more than that before committing it, > though, since I don't have a test changer anymore. (You've obviously > tested it, though, so we should be okay in that department.) Let me fixup some stuff based input, and I'll update the pr, and cc: you. Thanks! - Steve -- C. Stephen Gunn URL: http://www.waterspout.com/ WaterSpout Communications, Inc. Email: csg@waterspout.com 427 North 6th Street Phone: +1 765.742.6628 Lafayette, IN 47901 Fax: +1 765.742.0646 On Mon, 11 Sep 2000 00:56:23 CST, "Kenneth D. Merry" wrote: I've put a new patch online, same location: http://www.physics.purdue.edu/~csg/FreeBSD/chio.patch Some cleanups, command name changes, sanity check some pointers. > I'd suggest looking at them as well, even if it won't be used very often. > (I've moved things from pickers and portals often enough, though.) Done. > I also wonder about the 'unload' keyword. It almost sounds like something > you might use to eject a tape or something. What about calling it 'return' > instead? (I don't feel really strongly about this, but I figure we ought > to make things intuitive, if possible.) I slept on this, and 'return' makes as much sense (if not more) than 'unload' since it will work on any media unit that has a source attribute. > Anyway, I'll try to give the patch a more thorough look-over in the next > few days. I won't be able to do more than that before committing it, > though, since I don't have a test changer anymore. (You've obviously > tested it, though, so we should be okay in that department.) Seems to work for me (tm). Those who looked at it the first time are on the cc: line. Have another look if you want. :) - Steve -- C. Stephen Gunn URL: http://www.waterspout.com/ WaterSpout Communications, Inc. Email: csg@waterspout.com 427 North 6th Street Phone: +1 765.742.6628 Lafayette, IN 47901 Fax: +1 765.742.0646 Responsible Changed From-To: freebsd-bugs->ken Ken asked for the new patch. On Mon, Sep 11, 2000 at 23:21:45 -0500, C. Stephen Gunn wrote: > On Mon, 11 Sep 2000 00:56:23 CST, "Kenneth D. Merry" wrote: > > I've put a new patch online, same location: > > http://www.physics.purdue.edu/~csg/FreeBSD/chio.patch > > Some cleanups, command name changes, sanity check some pointers. Well, I took a look through it this evening, and I don't see any code problems. It looks good! I've made some minor style changes, though, diffs against -current are attached. Ken -- Kenneth Merry ken@kdm.org State Changed From-To: open->suspended I have committed the latest version of these patches to -current. Once they've had a chance to sit in -current for a little while, I'll merge them to RELENG_4. Thanks Steve! State Changed From-To: suspended->closed This support is in RELENG_4. |
chio(1) provides no mechanism to select media by "voltag." chio(1) provides no mechanism to return a media unit to its source. Fix: o A pseudo-element type selector "voltag" that can be used to select source elements: chio move voltag VOLUME-03 drive 0 chio exchange voltag VOLUME-03 voltag VOLUME-04 o A new sub-command "unload" that returns a media unit to the source element reported by 'chio status -a': chio unload drive 0 chio unload voltag VOLUME-03 A patch relative to 4.1-STABLE (should work on 5.0-CURRENT as well) is available at: http://www.physics.purdue.edu/~csg/FreeBSD/chio.patch How-To-Repeat: N/A - new feature request.