AlexJ Posted January 10, 2011 Posted January 10, 2011 Does anyone know Linux command, by which I can know total memory usage of a particular process? Example: 14275 xxxxx 20 0 565m 62m 32m S 0 0.8 0:01.10 apache2 14316 xxxxx 20 0 561m 47m 19m S 0 0.6 0:00.28 apache2 14334 xxxxx 20 0 549m 37m 21m S 0 0.5 0:00.16 apache2 14338 xxxxx 20 0 563m 44m 14m S 0 0.6 0:00.54 apache2 14339 xxxxx 20 0 542m 13m 3496 S 0 0.2 0:00.00 apache2 14343 xxxxx 20 0 559m 40m 14m S 0 0.5 0:00.28 apache2 14345 xxxxx 20 0 549m 26m 9m S 0 0.3 0:00.00 apache2 14346 xxxxx 20 0 549m 28m 12m S 0 0.4 0:00.06 apache2 14347 xxxxx 20 0 554m 32m 11m S 0 0.4 0:00.06 apache2 14348 xxxxx 20 0 561m 41m 13m S 0 0.5 0:00.56 apache2 14349 xxxxx 20 0 562m 44m 14m S 0 0.6 0:00.48 apache2 14356 xxxxx 20 0 548m 24m 9228 S 0 0.3 0:00.04 apache2 I would like to know total memory usage of all apache2 process together rather then as % of total memory.
Gary. Posted January 10, 2011 Posted January 10, 2011 In SSH type in:top Then press:shift + M This will show it based on memory usuage, Now depending on the spawns, I would not let apache use more than 5mb per process and 0% CPU, If you exceed that then you may encounter issues and optimization is needed. Hope it helps.
AlexJ Posted January 10, 2011 Author Posted January 10, 2011 This will show it based on memory usuage, Now depending on the spawns, I would not let apache use more than 5mb per process and 0% CPU, If you exceed that then you may encounter issues and optimization is needed. Hope it helps. 0% CPU? If process has spawned will it not use CPU? Also, I think mine uses more then 5Mb. Do you know what issue can be?
Gary. Posted January 10, 2011 Posted January 10, 2011 Possibly lots of connections. whats your current connections to apache
AlexJ Posted January 10, 2011 Author Posted January 10, 2011 Possibly lots of connections. whats your current connections to apache How to find current connections to apache? I tried this 'netstat -an | grep :80 | sort' and it gives me 32. Any other way to find connections? Also, ps aux | grep apache| wc -l gives me value of 20.
Gary. Posted January 11, 2011 Posted January 11, 2011 netstat -an | grep 80 | wc -l Thats fine, 32 is not alot... Infact it's probably idle. whats the output of:netstat -ntu | awk '{print $5}' | cut -d: -f1 | sort | uniq -c | sort -n Any unusual amount of connections in there, And anything that you do not reconise when running:tail -f /var/log/messages
.Nuno. Posted January 12, 2011 Posted January 12, 2011 Hello, I use a small scrip called ps_mem.py to list all running processes and related memory. just call the scrip in the comand line: ./ps_mem.py
AlexJ Posted January 13, 2011 Author Posted January 13, 2011 Gary, what does this means? Jan 10 06:25:02 server rsyslogd: [origin software="rsyslogd" swVersion="3.18.6" x-pid="3116" x-info="http://www.rsyslog.com"] restart Jan 11 06:25:01 server kernel: imklog 3.18.6, log source = /proc/kmsg started. Jan 11 06:25:01 server rsyslogd: [origin software="rsyslogd" swVersion="3.18.6" x-pid="3116" x-info="http://www.rsyslog.com"] restart Jan 12 02:53:10 server kernel: [1139013.562766] NETDEV WATCHDOG: eth0: transmit timed out Jan 12 02:53:11 server kernel: [1139013.579201] r8169: eth0: link up Jan 12 06:25:02 server kernel: Kernel logging (proc) stopped. Jan 12 06:25:02 server kernel: imklog 3.18.6, log source = /proc/kmsg started. Jan 12 06:25:02 server rsyslogd: [origin software="rsyslogd" swVersion="3.18.6" x-pid="3116" x-info="http://www.rsyslog.com"] restart Jan 12 07:02:34 server kernel: [1157380.949499] NETDEV WATCHDOG: eth0: transmit timed out Jan 12 07:02:35 server kernel: [1157380.965811] r8169: eth0: link up
Gary. Posted January 13, 2011 Posted January 13, 2011 I remember that error roughly from a few years back and it turned out to be an old kernal bug, Whats the current kernal version ? Type in SSH:uname -r
Recommended Posts
Archived
This topic is now archived and is closed to further replies.