External Check times out, then becomes "unsupported" item

[freak]

I wrote a small script giving back the last successful hop of a traceroute output.

I defined an External Check traceroute.sh[] (traceroute.sh in /etc/zabbix/externalscripts/ is called with the selected hostname/ip automatically).

Since traceroute needs quite a while to complete, I experienced that the item after some time gets in "Not supported" state. I guess that zabbix times if the script doesn't reply after a certain amount of time.
Can this be tweaked?

Additionally, it seems the script is being called seperately by more than one zabbix_server instance at the same time. Instead of just calling it once.
Is this a bug?

I experienced this behaviour in recent trunk (@5231).

[JoelG]

Similar Problem

I have a short external script that is showing a similar problem. It runs fine for a few samples, over a couple of minutes, then changes to the 'not supported' state. The script runs just fine from the command line. Under the worst network conditions, I don't expect it to take more then a second or two to complete.

This is a problem that has been reported multiple times, and I have not seen a real solution to the problem.

In addition, the external script parser does not seem to like floating point values as input, so I have my script multiply by 10, then have Zabbix do an internal conversion with a multiplier of 0.1. It works, but is clearly a work around.

Running version 1.4.4

[BusteR81]

same here

Additionally, it seems the script is being called seperately by more than one zabbix_server instance at the same time. Instead of just calling it once.
Is this a bug?

I experienced this behaviour in recent trunk (@5231).

i have a sh that send an traceroute email using sendmail if connectivity Down. however if a link is Down, i will receive 2 emails instead of 1. anyone knows how to stop "script is being called seperately by more than one zabbix_server instance at the same time." ??