EPICS Controls Argonne National Laboratory

Experimental Physics and
Industrial Control System

1994  1995  1996  1997  1998  1999  2000  2001  2002  <20032004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  <20032004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: RE: Channel Access disconnect/reconnect
From: "Jeff Hill" <[email protected]>
To: "'Nick Rees'" <[email protected]>, <[email protected]>
Date: Mon, 14 Apr 2003 17:53:49 -0600
Nick,

Are you using switches or hubs? We have seen problems like that
on hub based systems resulting from Ethernet collision chains
rarely (once in many transmissions) becoming long enough to push
the Ethernet collision retry delay exponential back off into a
run permit heartbeat timeout situation. 

Your sniffer will probably tell you what the maximum collision
retry count has been.

When this is running do you see unprocessed message bytes waiting
around continuously in either Solaris or vxWorks? To find out run
"netstat | grep 5064" on Solaris and inetstatShow on vxWorks.

Jeff

> -----Original Message-----
> From: Nick Rees [mailto:[email protected]]
> Sent: Monday, April 14, 2003 4:45 PM
> To: [email protected]
> Subject: Channel Access disconnect/reconnect
> 
> 
> We currently have an application running under R3.13.4 in which
> there is a
> tightly coupled 20 Hz loop between two systems. The CA client,
> running on
> solaris, receives monitors at 20 Hz, and the does some
> calculations and
> returns a demand to the CA server (running on a Heurikon
> baja68k board) -
> for total of 40 CA transactions/second.
> 
> About 2-4 times a day Channel Access reports a disconnect in
> the solaris
> client. However, as far as we can see with a network sniffer,
> the traffic
> continues to flow in both directions - the monitors certainly
> keep on
> coming in since the code processes, and it needs the monitors
> to trigger
> this. Nothing appears in the iocLogServer log file.
> 
> Has this been seen by anyone else and/or has it bee fixed since
> R3.13.4? I
> have looked at the release notes up to R3.13.8 and find some
> things which
> are maybe close, but most seem to be IOC issues, not client
> ones.
> 
> Cheers,
> 
> --
> Nick Rees
> 
> Head of Software and Computing Services
> Joint Astronomy Centre                   Ph:         +1 (808)
> 961-3756
> 660 N. Aohoku Place                      Fax:        +1 (808)
> 961-6516
> Hilo, HI.  96720                         Email:
> [email protected]



References:
Channel Access disconnect/reconnect Nick Rees

Navigate by Date:
Prev: Re: recDynLink.c for R3.14.1 ? Tim Mooney
Next: RE: drvAscii R3.14 version Kevin Tsubota
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  <20032004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: Channel Access disconnect/reconnect Nick Rees
Next: db_parse returned -1 M.C.Shao
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  <20032004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
ANJ, 10 Aug 2010 Valid HTML 4.01! · Home · News · About · Base · Modules · Extensions · Distributions · Download ·
· Search · EPICS V4 · IRMIS · Talk · Bugs · Documents · Links · Licensing ·