Experimental Physics and
Industrial Control System

[email protected] (Bill Brown) · Thu, 19 Oct 95 07:17:19 PDT

> Error codes and system wide error handler is another topic we have talked about
> for several years but never progressed past the existing vxWorks error codes
> and the system wide log file.
> We would all benefit if a nice solution is designed and implemented.

I'm also in agreement that this area could use some work.  We've done a simple
hack that checks the socket to the iocLogServer "every-so-often."  If it finds
an error it deletes the socket and re-opens it.  (This is from the ioc point
of view).  THerefore, if something happens to the iocLogServer, when it is
restarted the iocs will reconnect to it which does nnot happen under the
current arrangement.

We also have a cron job that runs a script every 24 hours which deletes the
iocLogServer, renames the iocLog.text file (to something like iocLog.yymmdd)
and restarts the iocLogserver.

This really isn't a great solution as a few log messages may get lost in the
process, but it does avoid the problem of having iocLogServer wedge when the
log file grows beyond a certain size (> !MB, I think?).  And it makes it a
bit easier to review the log files for clues as to what went wrong when if
bad things happen.

The issue was addressed at the last meeting (in Dallas) but I haven't heard of
any progress.  Of course the code (such as it is!) is available for the asking.
I'd kind of like to get it included in the next release, but it's so trivial
that it doesn't take much to tack it on locally.

Disclaimer:  Any opinions are my own and have	    |  -bill
    nothing to do with the official policy or the   |   [email protected]
    management of L.B.L, who probably couldn't      |   Berkeley, CA
    care less about employees who play with trains. |   aka [email protected]

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System