alanchow Posted November 19, 2012 Share Posted November 19, 2012 I've been testing the driver all day (new project, just a HC800 and the driver) with 15 second interval.Load Average currently is 0.02 0.07 0.12. 3% CPU usage and 7% CPUWill keep the monitor up. Unfortunately if the driver utilises such a heavy load at such a low interval i may have to increase the minimum interval to something like 2 minutes.Your dealer will need to install the correct version of the driver that works with gmail.Before making any changes my recommendation is to let me run my test bench controller for another few days to see if my load average on my controller increasing rises to 1. Link to comment Share on other sites More sharing options...
alanchow Posted November 19, 2012 Share Posted November 19, 2012 sorry meant 3% memory. Link to comment Share on other sites More sharing options...
alanchow Posted November 19, 2012 Share Posted November 19, 2012 i've replicated the problem. Took a day for the load average to increase. Give me a week or two to resolve it. Unfortunately replicating this problem takes a day each time so testing will take a while. Link to comment Share on other sites More sharing options...
pstuart Posted November 19, 2012 Share Posted November 19, 2012 Is the problem only with the POP driver or is the SMTP driver effected as well? Link to comment Share on other sites More sharing options...
alanchow Posted November 19, 2012 Share Posted November 19, 2012 Only with Pop. SMTP doesn't share any code with it and is only called when you actually send an email (in fact theres no timers or anythign that can cause a memory leak in that driver). Link to comment Share on other sites More sharing options...
alanchow Posted November 21, 2012 Share Posted November 21, 2012 i've updated the driver and have optimised the code. Have reduced load average from 0.20 to 0.00-0.05 (0.05 when the driver is actually polling). Dealers can download the latest driver from our website. Link to comment Share on other sites More sharing options...
digitaltrader Posted November 22, 2012 Author Share Posted November 22, 2012 i've updated the driver and have optimised the code. Have reduced load average from 0.20 to 0.00-0.05 (0.05 when the driver is actually polling). Dealers can download the latest driver from our website.Thanks Alan, Ill see if my dealer can update me. Any idea if this still has the CPU drag issue if left at 15 second poll interval after 2 days? Link to comment Share on other sites More sharing options...
alanchow Posted November 22, 2012 Share Posted November 22, 2012 I left it for two days and top gives the following on a 15 second poll. Note that 15 seconds is still a higher setting then i would use myself (i personally have it set to a minute).Mem: 105048K used, 917932K free, 0K shrd, 3684K buff, 56480K cachedCPU: 0% usr 0% sys 0% nic 99% idle 0% io 0% irq 0% sirqLoad average: 0.04 0.03 0.05 1/84 12718 PID PPID USER STAT VSZ %MEM %CPU COMMAND 2874 1 root S 2748 0% 0% /control4/bin/sddpd 2888 1 root S 56172 5% 0% director Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.