[GE users] time configuration parameters

tamara sgesystem at live.com
Wed Mar 25 19:20:57 GMT 2009


    [ The following text is in the "iso-8859-1" character set. ]
    [ Your display is set for the "ISO-8859-10" character set.  ]
    [ Some special characters may be displayed incorrectly. ]

Dear all,

i configure the parameters: load_report_time, max_unheard, reschedule_unknown and i get the best result for the fault tolerance about 7 seconds and max 16 seconds (it needs few seconds for the job to change its status from running to pending but it needs almost 1 second or less to reschedule the job (from pending to running on another host))
i was working with my system for more than one week but suddenly today the network was coming and going and so the master host was disconnecting .. connecting .. disconnecting...
when the network stabilized and tried the system again with the fault tolerance it was to slow it takes more seconds to detect the failure host and it need sometimes for 10 seconds to reschedule the job to another host, although the three parameters that i mentioned in the beginning didn't change

what could be the problem? should i configure another parameter?


regards,
tamara

________________________________
Get news, entertainment and everything you care about at Live.com. Check it out!<http://www.live.com/getstarted.aspx>



More information about the gridengine-users mailing list