EPICS Home

Experimental Physics and Industrial Control System


 
1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  <20112012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024  Index 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  <20112012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
<== Date ==> <== Thread ==>

Subject: Re: vxWorks network problems
From: "Steven M. Hartman" <[email protected]>
To: Dirk Zimoch <[email protected]>
Cc: "[email protected]" <[email protected]>
Date: Tue, 24 May 2011 09:19:41 -0400
Hi Dirk--

Dirk Zimoch wrote:

I tried to tune the network buffers. But once I when I saw with inetstatShow that send queues filled up, I increased their size from the default 8k to 64k. This seems to be a bad decision, because now I don't have enough mbufs any more.

Were you tuning to address a particular problem you were experiencing?

In the scenario you described where the server had a full send queue and multiple clients had large receive queues, increasing the queue size is unlikely to help much. At best, it may just slightly delay the inevitable (or starve the server of resources first).

I would be more curious as to what was happening (client, server, network) that led to this peculiar situation. (One end of the connection briefly queuing up is not unusual for a busy system, but both ends seems strange.)

It was user shift, so hopefully nobody had unplugged any cables.

Could still be a flaky edge network switch. Or something unrelated.

Probably ca_search UDP packages are related, but the time resolution of our network diagnosis is not so good that I can see if the IOC crashes because of search broadcasts or if the clients broadcast because the IOC crashed.

I would guess the search broadcasts were the result of the IOC crash, not the cause.


--
Steven Hartman
[email protected]

References:
vxWorks network problems Dirk Zimoch
Re: vxWorks network problems Steven M. Hartman
Re: vxWorks network problems Dirk Zimoch

Navigate by Date:
Prev: Re: vxWorks network problems Steven M. Hartman
Next: Re: Suggestion for IOC log server Ralph Lange
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  <20112012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024 
Navigate by Thread:
Prev: Re: vxWorks network problems Dirk Zimoch
Next: Re: vxWorks network problems Andrew Johnson
Index: 1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  <20112012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022  2023  2024