I think I have fixed this problem for good, see 
http://git.mah.priv.at/gitweb/emc2-dev.git/shortlog/refs/heads/abort-toolchange

wrt to my further studies:
the reason was simply that during a pending toolchange or tool-prepare, emc 
would not properly send an EMC_TOOL_ABORT message, hence iocontrol had no 
chance to react to it (which it wouldnt have done properly in the first place 
if it gotten it ;-).
Btw nml queuing works despite the missing 'queue' argument - I have no idea why.

so the fix contains two aspects:
- make sure emc sends an EMC_TOOL_ABORT message if a change/prepare is pending
- make sure iocontrol does the right thing when it gets that abort message.

The 'right thing' for iocontrol to do in my view is to deassert the tool-change 
and tool-prepare pins if they were set, beyond turning off lube and mist. With 
this patch it does that now. A consequence of deasserting 
iocontrol.0.tool-change is that the toolchanger window disappears on abort, 
which I think is the right way of handling this.

I'd appreciate a review of this patch.


-Michael



------------------------------------------------------------------------------
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and, 
should the need arise, upgrade to a full multi-node Oracle RAC database 
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to