I think it's fixed. I tried your scenario (on one machine however, with a
different second port) and get:
> /opt/owfs/bin/owdir -s 4305
/10.67C6697351FF
/26.4AEC29CDBAAB
/28.F2FBE3467CC2
/2C.66FE04000000
/2C.1B7900000000
/22.367A03000000
/12.C4DF1F000000
/01.5F50CA090000
/81.6B1B24000000
/89.BD29C300705E
/bus.1
/bus.0
/uncached
/settings
/system
/statistics
/structure
/simultaneous
/alarm
> /opt/owfs/bin/owdir -s 4305 bus.0
/bus.0/interface
/bus.0/2C.66FE04000000
/bus.0/2C.1B7900000000
/bus.0/22.367A03000000
/bus.0/12.C4DF1F000000
/bus.0/01.5F50CA090000
/bus.0/81.6B1B24000000
/bus.0/89.BD29C300705E
/bus.0/simultaneous
/bus.0/alarm
> /opt/owfs/bin/owdir -s 4305 bus.1
/bus.1/interface
/bus.1/10.67C6697351FF
/bus.1/26.4AEC29CDBAAB
/bus.1/28.F2FBE3467CC2
/bus.1/bus.0
/bus.1/uncached
/bus.1/settings
/bus.1/system
/bus.1/statistics
/bus.1/structure
/bus.1/simultaneous
/bus.1/alarm
The bus handling in the directory code had been rather sloppy. It was
recently tightened up, including better handling of the initial '/'
Paul Alfille
On Sun, May 17, 2009 at 10:32 PM, Rob Fugina <rob.fug...@gmail.com> wrote:
> This is from 2.7p19, so I apologize if it's been fixed since then...
>
> This involves two owserver instances, one of which is referencing and
> aggregating the other. So, two boxes. The host "slug" is running
> "owserver -p 4304 --fake=10,26,28" plus some debugging options. The
> host "troll" is running "owserver -p 4304 -u -s slug:4304". Both come
> up fine. To start with, running "owdir" on troll lists the real
> devices on its real 1-wire bus plus the 3 fake devices on slug, just
> as expected. Running "owdir bus.0" shows just the local stuff.
> Running "owdir bus.1" causes a seg fault on slug. The last debug
> message (at level 4) before the fault is "CALL: PARSENAME
> path=[bus.1]". If instead I run "owdir /bus.1", the owserver on slug
> doesn't crash, and everything goes as expected. The matching debug
> line shown on slug is "CALL: PARSENAME path=[]" in this case.
>
> So, I can kind of see what's not going right, and it seems to be on
> the server (troll) end. Is there any additional information I can
> provide that would be of help? Has it already been fixed, perhaps,
> and I should just upgrade?
>
> Thanks,
> Rob
>
>
> ------------------------------------------------------------------------------
> Crystal Reports - New Free Runtime and 30 Day Trial
> Check out the new simplified licensing option that enables
> unlimited royalty-free distribution of the report engine
> for externally facing server and web deployment.
> http://p.sf.net/sfu/businessobjects
> _______________________________________________
> Owfs-developers mailing list
> Owfs-developers@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/owfs-developers
>
------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables
unlimited royalty-free distribution of the report engine
for externally facing server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers
