On Wed, 25 Oct 2006 11:15:35 -0400 Robert wrote: RS> On Wed, 25 Oct 2006 06:36:57 -0700 Wes wrote: RS> WH> >>>>> "AN" == Anthony Novatsis <[EMAIL PROTECTED]> writes: RS> WH> AN> For Net-SNMP version 5.3.1 (sending a SIGTERM to the pass-through RS> WH> AN> executable). RS> WH> RS> WH> In order for it to go into 5.4 I'll need to get feedback from other RS> WH> developers as well since we're in lock-down mode on that code branch. RS> WH> To me your patch looks good and like the right thing to do. To the RS> WH> other developers: please chime in if you agree (or not). RS> RS> Hmm.. it seems to me that the 'right' thing to do would be to extend the RS> persist protocol to send a shutdown messages and give them a little time to RS> process it. Or in this case, pass the HUP signal down. Killing the process RS> seems a bit rude.
Actually, after looking at the code, the closing of the pipe/fd is a signal to the persist app, which should terminate on that condition. Thus, I think this is a bug in the persist script, not the agent. Something that closed the pipes, then set a timer to give the process time to exit gracefully and *then* kill it would be ok, I think. But closing and then immediately killing is wrong. -1 on the patch. ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Net-snmp-coders mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
