On Apr 28, 2009, at 11:27 AM, kingpin wrote: > Hi, as a temporary workaround solution I've decided to automate the > restart of the zenmodeler daemon when it hangs by means of en event > command. > > This is my event command: > set kpid; kpid=`cat /usr/local/zenoss/zenoss/var/zenmodeler- > localhost.pid`; kill -9 $kpid; zenmodeler start > > The event command is triggered correctly but then the command itself > fails. I always get this Cmd/Fail event : > Error running: zenmodeler restart: /bin/sh: line 1: : command not > found > > Error running: zenmodeler restart: /bin/sh: /usr/local/zenoss/zenoss/ > bin/zenmodeler start > : No such file or directory > /bin/sh: line 1: > : command not found > > > It does work when I run it on a command line terminal as the zenoss > user but fails when triggered as an event command
The ticket below likely explains why your zenmodeler hangs. Unfortunately the patch is in one of our libraries, so a zenpatch command can't be used to fix it. What you need to do is edit netsnmp.py on your system. vi `find $ZENHOME -name netsnmp.py` You're looking for a line that looks like the following: fdset = c_long * (MAXFD/32) You need to change the c_long to c_int32 so that it looks like this: fdset = c_int32 * (MAXFD/32) Once you save the file, you should restart zenmodeler, zenprocess and zenperfsnmp. Ticket #2914: python-pynetsnmp: get hangs in snmp_read on amd64 http://dev.zenoss.org/trac/ticket/2914 _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
