Let's try the dev list.

Anyone have a spare USB UPS they can test against? The fix is for Tripp Lite 
and OS X, but I'm looking for other data points because the code is common to 
nearly *all* of the USB drivers, and could break them.

Thanks!

On Aug 16, 2014, at 5:25 PM, Charles Lepple <[email protected]> wrote:

> If you have a USB UPS, and can rebuild NUT from source (tarball or Git), 
> please read on.
> 
> I was trying to fix a problem specific to OS X and Tripp Lite UPSes, but it 
> turns out that the redundant call to usb_set_altinterface(0) is right in the 
> middle of the common NUT USB driver code, affecting pretty much any USB UPS 
> that we support. So I would like to get a little additional testing before 
> merging this. (Thanks to Steve Ballantyne for testing this on a different 
> Tripp Lite model.)
> 
> Basically, this call to usb_set_altinterface() should not be needed 
> (alternate interfaces are not often implemented in USB), but it appears that 
> OS X might be sending this control message internally, and the extra call in 
> the NUT source causes problems. Rather than rip it out completely, I made it 
> configurable, just in case someone runs across an UPS that needs the old 
> behavior.
> 
> The full branch is visible here: 
> https://github.com/networkupstools/nut/compare/issue_138_altinterface_value
> 
> A few other fixes and reorganizations crept into this patch, but they are 
> either just reorganization (08c721d...) or a workaround for the *.orig files 
> that patch typically leaves behind (4fc6f73...).
> 
> https://github.com/networkupstools/nut/commit/08c721d67ca232d52c6b7d5e67182d56009f8cc7
> https://github.com/networkupstools/nut/commit/4fc6f738478e60be9d306cff9eb0f1e38cb53d78
> 
> When you test, it would be best to unplug and re-plug the UPS USB connection 
> so that we are not relying on any previous USB setup. Please indicate your OS 
> and UPS type when testing.
> 
> Thanks,
> 
> -- 
> Charles Lepple
> clepple@gmail
> 
> 

Attachment: issue_138_usb_altinterface.patch
Description: Binary data

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

Reply via email to