On Tue, 23 Feb 2010 17:59:41 +0100 Bart wrote: BVA> The above is indeed only the case if NETSNMP_DS_LIB_ALARM_DONT_USE_SIG has BVA> been set to zero (the default value is one), such that Net-SNMP alarms BVA> trigger SIGALRM.
good. BVA> Forking off a child process might introduce more issues than it solves. As BVA> an example, what should happen with the child process when snmpd receives BVA> SIGHUP ? I wouldn't think anything. What do you think should happen? BVA> It is not clear to me why a thread has been created to listen for BVA> RTM_NEWADDR, RTM_DELADDR and RTM_NEWPREFIX messages instead of processing BVA> the netlink socket data via snmp_sess_select() ? excellent point. BVA> Another remark: the code that handles the netlink messages is not correct. BVA> The queue used by the Linux kernel to pass data from the kernel to user BVA> space has a fixed size (configurable via SO_RCVBUF) and hence can overflow. BVA> When such an overflow happens some of the RTM_* messages are lost. The BVA> process that is listening for RTM_* messages must detect overflows and if an BVA> overflow happened it must poll the current state such that the in-kernel BVA> state and the state in the user process are again in sync. Overflows are BVA> reported by the kernel by sending a message with the type NLMSG_ERROR and BVA> setting an appropriate error code. again, excellent point. I think we need a bug report for both of these. ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Net-snmp-coders mailing list Net-snmp-coders@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/net-snmp-coders
